940:
domain of citation recommender systems, users typically do not rate a citation or recommended article. In such cases, offline evaluations may use implicit measures of effectiveness. For instance, it may be assumed that a recommender system is effective that is able to recommend as many articles as possible that are contained in a research article's reference list. However, this kind of offline evaluations is seen critical by many researchers. For instance, it has been shown that results of offline evaluations have low correlation with results from user studies or A/B tests. A dataset popular for offline evaluation has been shown to contain duplicate data and thus to lead to wrong conclusions in the evaluation of algorithms. Often, results of so-called offline evaluations do not correlate with actually assessed user-satisfaction. This is probably because offline training is highly biased toward the highly reachable items, and offline testing data is highly influenced by the outputs of the online recommendation module. Researchers have concluded that the results of offline evaluations should be viewed critically.
1135:(AI) applications in recommendation systems are the advanced methodologies that leverage AI technologies, to enhance the performance recommendation engines. The AI-based recommender can analyze complex data sets, learning from user behavior, preferences, and interactions to generate highly accurate and personalized content or product suggestions. The integration of AI in recommendation systems has marked a significant evolution from traditional recommendation methods. Traditional methods often relied on inflexible algorithms that could suggest items based on general user trends or apparent similarities in content. In comparison, AI-powered systems have the capability to detect patterns and subtle distinctions that may be overlooked by traditional methods. These systems can adapt to specific individual preferences, thereby offering recommendations that are more aligned with individual user needs. This approach marks a shift towards more personalized, user-centric suggestions.
677:, content-based filtering, and other approaches. There is no reason why several different techniques of the same type could not be hybridized. Hybrid approaches can be implemented in several ways: by making content-based and collaborative-based predictions separately and then combining them; by adding content-based capabilities to a collaborative-based approach (and vice versa); or by unifying the approaches into one model. Several studies that empirically compared the performance of the hybrid with the pure collaborative and content-based methods and demonstrated that the hybrid methods can provide more accurate recommendations than pure approaches. These methods can also be used to overcome some of the common problems in recommender systems such as cold start and the sparsity problem, as well as the knowledge engineering bottleneck in
1057:, et al. criticized that "it is currently difficult to reproduce and extend recommender systems research results," and that evaluations are "not handled consistently". Konstan and Adomavicius conclude that "the Recommender Systems research community is facing a crisis where a significant number of papers present results that contribute little to collective knowledge often because the research lacks the evaluation to be properly judged and, hence, to provide meaningful contributions." As a consequence, much research about recommender systems can be considered as not reproducible. Hence, operators of recommender systems find little guidance in the current research for answering the question, which recommendation approaches to use in a recommender systems.
441:. Collaborative filtering is based on the assumption that people who agreed in the past will agree in the future, and that they will like similar kinds of items as they liked in the past. The system generates recommendations using only information about rating profiles for different users or items. By locating peer users/items with a rating history similar to the current user or item, they generate recommendations using this neighborhood. Collaborative filtering methods are classified as memory-based and model-based. A well-known example of memory-based approaches is the user-based algorithm, while that of model-based approaches is
1045:, IJCAI), has shown that on average less than 40% of articles could be reproduced by the authors of the survey, with as little as 14% in some conferences. The articles considers a number of potential problems in today's research scholarship and suggests improved scientific practices in that area. More recent work on benchmarking a set of the same methods came to qualitatively very different results whereby neural methods were found to be among the best performing methods. Deep learning and neural methods for recommender systems have been used in the winning solutions in several recent recommender system challenges, WSDM,
1066:
recommendation algorithms or scenarios led to strong changes in the effectiveness of a recommender system. They conclude that seven actions are necessary to improve the current situation: "(1) survey other research fields and learn from them, (2) find a common understanding of reproducibility, (3) identify and understand the determinants that affect reproducibility, (4) conduct more comprehensive experiments (5) modernize publication practices, (6) foster the development and use of recommendation frameworks, and (7) establish best-practice guidelines for recommender-systems research."
1161:(CF) is one of the most commonly used recommendation system algorithms. It generates personalized suggestions for users based on explicit or implicit behavioral patterns to form predictions. Specifically, it relies on external feedback such as star ratings, purchasing history and so on to make judgments. CF make predictions about users' preference based on similarity measurements. Essentially, the underlying theory is: "if user A is similar to user B, and if A likes item C, then it is likely that B also likes item C."
829:. From 2006 to 2009, Netflix sponsored a competition, offering a grand prize of $ 1,000,000 to the team that could take an offered dataset of over 100 million movie ratings and return recommendations that were 10% more accurate than those offered by the company's existing recommender system. This competition energized the search for new and more accurate algorithms. On 21 September 2009, the grand prize of US$ 1,000,000 was given to the BellKor's Pragmatic Chaos team using tiebreaking rules.
318:. Collaborative filtering approaches build a model from a user's past behavior (items previously purchased or selected and/or numerical ratings given to those items) as well as similar decisions made by other users. This model is then used to predict items (or ratings for items) that the user may have an interest in. Content-based filtering approaches utilize a series of discrete, pre-tagged characteristics of an item in order to recommend additional items with similar properties.
430:
1210:(ANN), is a deep learning model structure which aims to mimic a human brain. They comprise a series of neurons, each responsible for receiving and processing information transmitted from other interconnected neurons. Similar to a human brain, these neurons will change activation state based on incoming signals (training input and backpropagated output), allowing the system to adjust activation weights during the network learning phase. ANN is usually designed to be a
750:
learning that is of particular use in the area of recommender systems is the fact that the models or policies can be learned by providing a reward to the recommendation agent. This is in contrast to traditional learning techniques which rely on supervised learning approaches that are less flexible, reinforcement learning recommendation techniques allow to potentially train models that can be optimized directly on metrics of engagement, and user interest.
737:
current user session. Domains, where session-based recommendations are particularly relevant, include video, e-commerce, travel, music and more. Most instances of session-based recommender systems rely on the sequence of recent interactions within a session without requiring any additional details (historical, demographic) of the user. Techniques for session-based recommendations are mainly based on generative sequential models such as
1080:
636:
recommendation system is significantly less than when other content types from other services can be recommended. For example, recommending news articles based on news browsing is useful. Still, it would be much more useful when music, videos, products, discussions, etc., from different services, can be recommended based on news browsing. To overcome this, most content-based recommender systems now use some form of the hybrid system.
576:. Content-based filtering methods are based on a description of the item and a profile of the user's preferences. These methods are best suited to situations where there is known data on an item (name, location, description, etc.), but not on the user. Content-based recommenders treat recommendation as a user-specific classification problem and learn a classifier for the user's likes and dislikes based on an item's features.
1345:. Therefore, there is a risk that the market could become fragmented, leaving it to the viewer to visit various locations and find what they want to watch in a way that is time-consuming and complicated for them. By using a search and recommendation engine, viewers are provided with a central 'portal' from which to discover content from several sources in just one location.
759:
predict a rating for unexplored items of u by exploiting preference information on multiple criteria that affect this overall preference value. Several researchers approach MCRS as a multi-criteria decision making (MCDM) problem, and apply MCDM methods and techniques to implement MCRS systems. See this chapter for an extended introduction.
2845:
768:
pushing recommendations in certain circumstances, for instance, during a professional meeting, early morning, or late at night. Therefore, the performance of the recommender system depends in part on the degree to which it has incorporated the risk into the recommendation process. One option to manage this issue is
1150:. These advanced methods enhance system capabilities to predict user preferences and deliver personalized content more accurately. Each technique contributes uniquely. The following sections will introduce specific AI models utilized by a recommendation system by illustrating their theories and functionalities.
1247:
Natural language processing is a series of AI algorithms to make natural human language accessible and analyzable to a machine. It is a fairly modern technique inspired by the growing amount of textual information. For application in recommendation system, a common case is the Amazon customer review.
963:– In some situations, it is more effective to re-show recommendations, or let users re-rate items, than showing new items. There are several reasons for this. Users may ignore items when they are shown for the first time, for instance, because they had no time to inspect the recommendations carefully.
855:
A number of privacy issues arose around the dataset offered by
Netflix for the Netflix Prize competition. Although the data sets were anonymized in order to preserve customer privacy, in 2007 two researchers from the University of Texas were able to identify individual users by matching the data sets
687:
is a good example of the use of hybrid recommender systems. The website makes recommendations by comparing the watching and searching habits of similar users (i.e., collaborative filtering) as well as by offering movies that share characteristics with films that a user has rated highly (content-based
301:
articles to television. As operators compete to be the gateway to home entertainment, personalized television is a key service differentiator. Academic content discovery has recently become another area of interest, with several companies being established to help academic researchers keep up to date
1010:
is a measure of "how surprising the recommendations are". For instance, a recommender system that recommends milk to a customer in a grocery store might be perfectly accurate, but it is not a good recommendation because it is an obvious item for the customer to buy. " serves two purposes: First, the
916:
Evaluating the performance of a recommendation algorithm on a fixed test dataset will always be extremely challenging as it is impossible to accurately predict the reactions of real users to the recommendations. Hence any metric that computes the effectiveness of an algorithm in offline data will be
448:
A key advantage of the collaborative filtering approach is that it does not rely on machine analyzable content and therefore it is capable of accurately recommending complex items such as movies without requiring an "understanding" of the item itself. Many algorithms have been used in measuring user
341:) to seed a "station" that plays music with similar properties. User feedback is used to refine the station's results, deemphasizing certain attributes when a user "dislikes" a particular song and emphasizing other attributes when a user "likes" a song. This is an example of a content-based approach.
1040:
in recommender systems publications. The topic of reproducibility seems to be a recurrent issue in some
Machine Learning publication venues, but does not have a considerable effect beyond the world of scientific publication. In the context of recommender systems a 2019 paper surveyed a small number
939:
The effectiveness of recommendation approaches is then measured based on how well a recommendation approach can predict the users' ratings in the dataset. While a rating is an explicit expression of whether a user liked a movie, such information is not available in all domains. For instance, in the
583:
is built to indicate the type of item this user likes. In other words, these algorithms try to recommend items similar to those that a user liked in the past or is examining in the present. It does not rely on a user sign-in mechanism to generate this often temporary profile. In particular, various
371:
created the first recommender system in 1979, called Grundy. She looked for a way to recommend users books they might like. Her idea was to create a system that asks users specific questions and classifies them into classes of preferences, or "stereotypes", depending on their answers. Depending on
261:
generators for video and music services, product recommenders for online stores, or content recommenders for social media platforms and open web content recommenders. These systems can operate using a single type of input, like music, or multiple inputs within and across platforms like news, books
794:
to offer personalized, context-sensitive recommendations. This is a particularly difficult area of research as mobile data is more complex than data that recommender systems often have to deal with. It is heterogeneous, noisy, requires spatial and temporal auto-correlation, and has validation and
767:
The majority of existing approaches to recommender systems focus on recommending the most relevant content to users using contextual information, yet do not take into account the risk of disturbing the user with unwanted notifications. It is important to consider the risk of upsetting the user by
749:
The recommendation problem can be seen as a special instance of a reinforcement learning problem whereby the user is the environment upon which the agent, the recommendation system acts upon in order to receive a reward, for instance, a click or engagement by the user. One aspect of reinforcement
1218:
ANN is widely used in recommendation systems for its power to utilize various data. Other than feedback data, ANN can incorporate non-feedback data which are too intricate for collaborative filtering to learn, and the unique structure allows ANN to identify extra signal from non-feedback data to
810:
to generate driving routes for taxi drivers in a city. This system uses GPS data of the routes that taxi drivers take while working, which includes location (latitude and longitude), time stamps, and operational status (with or without passengers). It uses this data to recommend a list of pickup
758:
Multi-criteria recommender systems (MCRS) can be defined as recommender systems that incorporate preference information upon multiple criteria. Instead of developing recommendation techniques based on a single criterion value, the overall preference of user u for the item i, these systems try to
635:
A key issue with content-based filtering is whether the system can learn user preferences from users' actions regarding one content source and use them across other content types. When the system is limited to recommending content of the same type as the user is already using, the value from the
333:
Last.fm creates a "station" of recommended songs by observing what bands and individual tracks the user has listened to on a regular basis and comparing those against the listening behavior of other users. Last.fm will play tracks that do not appear in the user's library, but are often played by
1273:
An emerging market for content discovery platforms is academic content. Approximately 6000 academic journal articles are published daily, making it increasingly difficult for researchers to balance time management with staying up to date with relevant research. Though traditional tools academic
985:
is particularly notable for the detailed personal information released in its dataset. Ramakrishnan et al. have conducted an extensive overview of the trade-offs between personalization and privacy and found that the combination of weak ties (an unexpected connection that provides serendipitous
798:
There are three factors that could affect the mobile recommender systems and the accuracy of prediction results: the context, the recommendation method and privacy. Additionally, mobile recommender systems suffer from a transplantation problem – recommendations may not apply in all regions (for
644:
of items, because as they also reflect aspects of the item like metadata, extracted features are widely concerned by the users. Sentiments extracted from the reviews can be seen as users' rating scores on the corresponding features. Popular approaches of opinion-based recommender system utilize
615:
representation (also called vector space representation). The system creates a content-based profile of users based on a weighted vector of item features. The weights denote the importance of each feature to the user and can be computed from individually rated content vectors using a variety of
736:
These recommender systems use the interactions of a user within a session to generate recommendations. Session-based recommender systems are used at YouTube and Amazon. These are particularly useful when history (such as past clicks, purchases) of a user is not available or not relevant in the
639:
Content-based recommender systems can also include opinion-based recommender systems. In some cases, users are allowed to leave text reviews or feedback on the items. These user-generated texts are implicit data for the recommender system because they are potentially rich resources of both
924:
In A/B tests, recommendations are shown to typically thousands of users of a real product, and the recommender system randomly picks at least two different recommendation approaches to generate recommendations. The effectiveness is measured with implicit measures of effectiveness such as
1065:
conducted a study of papers published in the field, as well as benchmarked some of the most popular frameworks for recommendation and found large inconsistencies in results, even when the same algorithms and data sets were used. Some researchers demonstrated that minor variations in the
3828:
Jannach, Dietmar; Lerche, Lukas; Gedikli, Fatih; Bonnin, Geoffray (June 10, 2013). "What
Recommenders Recommend – an Analysis of Accuracy, Popularity, and Sales Diversity Effects". In Carberry, Sandra; Weibelzahl, Stephan; Micarelli, Alessandro; Semeraro, Giovanni (eds.).
1290:
that takes a researchers' authorized paper and citations as input. Whilst these recommendations have been noted to be extremely good, this poses a problem with early career researchers which may be lacking a sufficient body of work to produce accurate recommendations.
610:
Basically, these methods use an item profile (i.e., a set of discrete attributes and features) characterizing the item within the system. To abstract the features of the items in the system, an item presentation algorithm is applied. A widely used algorithm is the
2842:
249:
that provides suggestions for items that are most pertinent to a particular user. Recommender systems are particularly useful when an individual needs to choose an item from a potentially overwhelming number of items that a service may offer.
563:
originally used collaborative filtering to recommend new friends, groups, and other social connections by examining the network of connections between a user and their friends. Collaborative filtering is still used as part of hybrid systems.
1282:
provide a readily accessible database of journal articles, content recommendation in these cases are performed in a 'linear' fashion, with users setting 'alarms' for new publications based on keywords, journals or particular authors.
349:
problem, and is common in collaborative filtering systems. Whereas
Pandora needs very little information to start, it is far more limited in scope (for example, it can only make recommendations that are similar to the original seed).
3056:, pp. 377-408, The Adaptive Web, Peter Brusilovsky, Alfred Kobsa, Wolfgang Nejdl (Ed.), Lecture Notes in Computer Science, Springer-Verlag, Berlin, Germany, Lecture Notes in Computer Science, Vol. 4321, May 2007, 978-3-540-72078-2.
3942:
Langer, Stefan (September 14, 2015). "A Comparison of
Offline Evaluations, Online Evaluations, and User Studies in the Context of Research-Paper Recommender Systems". In Kapidakis, Sarantos; Mazurek, Cezary; Werla, Marcin (eds.).
1214:
model. Unlike regular machine learning where the underlying theoretical components are formal and rigid, the collaborative effects of neurons are not entirely clear, but modern experiments has shown the predictive power of ANN.
2494:
4009:
Beel, Joeran; Genzmehr, Marcel; Langer, Stefan; Nürnberger, Andreas; Gipp, Bela (January 1, 2013). "A comparative analysis of offline and online evaluations and discussion of research paper recommender system evaluation".
912:
are useful to assess the quality of a recommendation method. Diversity, novelty, and coverage are also considered as important aspects in evaluation. However, many of the classic evaluation measures are highly criticized.
1027:(CTR) for recommendations labeled as "Sponsored" were lower (CTR=5.93%) than CTR for identical recommendations labeled as "Organic" (CTR=8.86%). Recommendations with no label performed best (CTR=9.87%) in that study.
1017:– A recommender system is of little value for a user if the user does not trust the system. Trust can be built by a recommender system by explaining how it generates recommendations, and why it recommends an item.
1248:
Amazon will analyze the feedbacks comments from each customer and report relevant data to other customers for reference. The recent years have witnessed the development of various text analysis models, including
994:– Beel et al. found that user demographics may influence how satisfied users are with recommendations. In their paper they show that elderly users tend to be more interested in recommendations than younger users.
548:: The number of items sold on major e-commerce sites is extremely large. The most active users will only have rated a small subset of the overall database. Thus, even the most popular items have very few ratings.
920:
User studies are rather a small scale. A few dozens or hundreds of users are presented recommendations created by different recommendation approaches, and then the users judge which recommendations are best.
542:: There are millions of users and products in many of the environments in which these systems make recommendations. Thus, a large amount of computation power is often necessary to calculate recommendations.
5290:
3049:
345:
Each type of system has its strengths and weaknesses. In the above example, Last.fm requires a large amount of information about a user to make accurate recommendations. This is an example of the
4296:
844:
Many benefits accrued to the web due to the
Netflix project. Some teams have taken their technology and applied it to other markets. Some members from the team that finished second place founded
257:, such as what product to purchase, what music to listen to, or what online news to read. Recommender systems are used in a variety of areas, with commonly recognised examples taking the form of
2336:
2531:
1049:. Moreover, neural and deep learning methods are widely used in industry where they are extensively tested. The topic of reproducibility is not new in recommender systems. By 2011,
1305:
In contrast to an engagement-based ranking system employed by social media and other digital platforms, a bridging-based ranking optimizes for content that is unifying instead of
4351:
948:
Typically, research on recommender systems is concerned with finding the most accurate recommendation algorithms. However, there are a number of factors that are also important.
410:
et al. discussed the problems of offline evaluations. Beel et al. have also provided literature surveys on available research paper recommender systems and existing challenges.
357:
since they help users discover items they might not have found otherwise. Of note, recommender systems are often implemented using search engines indexing non-traditional data.
285:
in order to discover and recommend appropriate content, whilst reducing ongoing maintenance and development costs. A content discovery platform delivers personalized content to
832:
The most accurate algorithm in 2007 used an ensemble method of 107 different algorithmic approaches, blended into a single prediction. As stated by the winners, Bell et al.:
4202:
4070:. 41st Annual International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR 2018). Ann Arbor, Michigan, USA: ACM. pp. 415–424.
2003:
Bobadilla, J.; Ortega, F.; Hernando, A.; Alcalá, J. (2011). "Improving collaborative filtering recommender system results and performance using genetic algorithms".
1337:
As the connected television landscape continues to evolve, search and recommendation are seen as having an even more pivotal role in the discovery of content. With
860:. As a result, in December 2009, an anonymous Netflix user sued Netflix in Doe v. Netflix, alleging that Netflix had violated United States fair trade laws and the
4906:
Konstan, Joseph A.; Adomavicius, Gediminas (January 1, 2013). "Toward identification and adoption of best practices in algorithmic recommender systems research".
403:
5294:
3095:
Chen, Minmin; Beutel, Alex; Covington, Paul; Jain, Sagar; Belletti, Francois; Chi, Ed (2018). "Top-K Off-Policy
Correction for a REINFORCE Recommender System".
1062:
1011:
chance that users lose interest because the choice set is too uniform decreases. Second, these items are needed for algorithms to learn and improve themselves".
552:
One of the most famous examples of collaborative filtering is item-to-item collaborative filtering (people who buy x also buy y), an algorithm popularized by
1928:
Chen, Hung-Hsuan; Chen, Pu (January 9, 2019). "Differentiating
Regularization Weights -- A Simple Mechanism to Alleviate Cold Start in Recommender Systems".
4771:
3751:. In: Proceedings of the 33rd International ACMSIGIR Conference on Research and Development in Information Retrieval, SIGIR 2010, pp. 210–217. ACM, New York
3069:
Hidasi, Balázs; Karatzoglou, Alexandros; Baltrunas, Linas; Tikk, Domonkos (March 29, 2016). "Session-based
Recommendations with Recurrent Neural Networks".
3046:
406:
provided a new, alternate overview of recommender systems. Herlocker provides an additional overview of evaluation techniques for recommender systems, and
3722:
2244:
1041:
of hand-picked publications applying deep learning or neural methods to the top-k recommendation problem, published in top conferences (SIGIR, KDD, WWW,
616:
techniques. Simple approaches use the average values of the rated item vector while other sophisticated methods use machine learning techniques such as
4680:
Rendle, Steffen; Krichene, Walid; Zhang, Li; Anderson, John (September 22, 2020). "Neural
Collaborative Filtering vs. Matrix Factorization Revisited".
1050:
1046:
3793:
Chen, Hung-Hsuan; Chung, Chu-An; Huang, Hsin-Chien; Tsui, Wen (September 1, 2017). "Common Pitfalls in Training and Evaluating Recommender Systems".
5549:
532:: For a new user or item, there is not enough data to make accurate recommendations. Note: one commonly implemented solution to this problem is the
5401:
3479:
838:
Our experience is that most efforts should be concentrated in deriving substantially different approaches, rather than refining a single technique.
2222:
BEEL, Joeran, et al. Paper recommender systems: a literature survey. International Journal on Digital Libraries, 2016, 17. Jg., Nr. 4, S. 305–338.
584:
candidate items are compared with items previously rated by the user, and the best-matching items are recommended. This approach has its roots in
4524:
2078:
1175:: Create a n-dimensional space where each axis represents a user's trait (ratings, purchases, etc.). Represent the user as a point in that space.
4865:
Ekstrand, Michael D.; Ludwig, Michael; Konstan, Joseph A.; Riedl, John T. (January 1, 2011). "Rethinking the recommender research ecosystem".
3627:
3559:
Pimenidis, Elias; Polatidis, Nikolaos; Mouratidis, Haralambos (August 3, 2018). "Mobile recommender systems: Identifying the major concepts".
2557:
1500:
334:
other users with similar interests. As this approach leverages the behavior of users, it is an example of a collaborative filtering technique.
3338:
Xin, Xin; Karatzoglou, Alexandros; Arapakis, Ioannis; Jose, Joemon (2020). "Self-Supervised Reinforcement Learning for Recommender Systems".
2326:." In Proceedings of the SIGCHI conference on Human factors in computing systems, pp. 194–201. ACM Press/Addison-Wesley Publishing Co., 1995.
2306:." In Proceedings of the SIGCHI conference on Human factors in computing systems, pp. 210–217. ACM Press/Addison-Wesley Publishing Co., 1995.
407:
208:
5317:
5158:
Wu, L. (May 2023). "A Survey on Accuracy-Oriented Neural Recommendation: From Collaborative Filtering to Information-Rich Recommendation".
2282:
5416:
3680:
2442:
Herlocker, J. L.; Konstan, J. A.; Terveen, L. G.; Riedl, J. T. (January 2004). "Evaluating collaborative filtering recommender systems".
1058:
640:
feature/aspects of the item and users' evaluation/sentiment to the item. Features extracted from the user-generated reviews are improved
321:
The differences between collaborative and content-based filtering can be demonstrated by comparing two early music recommender systems,
442:
105:
17:
1575:
Gupta, Pankaj; Goel, Ashish; Lin, Jimmy; Sharma, Aneesh; Wang, Dong; Zadeh, Reza (2013). "WTF: the who to follow service at Twitter".
955:– Users tend to be more satisfied with recommendations when there is a higher intra-list diversity, e.g. items from different artists.
2319:
2982:
The Deep Learning–Based Recommender System "Pubmender" for Choosing a Biomedical Publication Venue: Development and Validation Study
2780:
1023:– User satisfaction with recommendations may be influenced by the labeling of the recommendations. For instance, in the cited study
5524:
1197:: The system will analyze the similar preference of the k neighbors. The system will make recommendations based on that similarity
936:
Offline evaluations are based on historic data, e.g. a dataset that contains information about how users previously rated movies.
4770:
Schifferer, Benedikt; Deotte, Chris; Puget, Jean-François; de Souza Pereira, Gabriel; Titericz, Gilberto; Liu, Jiwei; Ak, Ronay.
4449:"Do not blame it on the algorithm: an empirical assessment of multiple recommender systems and their impact on content diversity"
3359:
Ie, Eugene; Jain, Vihan; Narvekar, Sanmit; Agarwal, Ritesh; Wu, Rui; Cheng, Heng-Tze; Chandra, Tushar; Boutilier, Craig (2019).
973:
using collaborative filtering can be problematic from a privacy point of view. Many European countries have a strong culture of
266:. Recommender systems have also been developed to explore research articles and experts, collaborators, and financial services.
5652:
379:
at Columbia University, and implemented at scale and worked through in technical reports and publications from 1994 onwards by
1325:
planned to pilot in 2024. Aviv Ovadya also argues for implementing bridging-based algorithms in major platforms by empowering
1164:
There are many models available for collaborative filtering. For AI-applied collaborative filtering, a common model is called
969:– Recommender systems usually have to deal with privacy concerns because users have to reveal sensitive information. Building
722:: One recommendation technique is applied and produces some sort of model, which is then the input used by the next technique.
5559:
5534:
5063:
5022:
4923:
4882:
4828:
4746:
4709:
4649:
4326:
4185:
4037:
3993:
3960:
3918:
3864:
3409:
3314:
3269:
3193:
3138:
2906:
2870:
2819:
2582:
2517:
2303:
1987:
1851:
1483:
1360:
1042:
849:
182:
2986:
297:. A large range of content discovery platforms currently exist for various forms of content ranging from news articles and
3119:
Yifei, Ma; Narayanaswamy, Balakrishnan; Haibin, Lin; Hao, Ding (2020). "Temporal-Contextual Recommendation in Real-Time".
2264:
2174:
Herz, Frederick SM. "Customized electronic newspapers and advertisements." U.S. Patent 7,483,871, issued January 27, 2009.
981:
can result in a negative customer response. Much research has been conducted on ongoing privacy issues in this space. The
3537:
2495:"A comparative analysis of offline and online evaluations and discussion of research paper recommender system evaluation"
2085:
2075:
1097:
1036:
Recommender systems are notoriously difficult to evaluate offline, with some researchers claiming that this has led to a
2956:
1805:
716:: Recommenders are given strict priority, with the lower priority ones breaking ties in the scoring of the higher ones.
5442:
5376:
4852:
3889:
Proceedings of the 24th annual international ACM SIGIR conference on Research and development in information retrieval
3763:
Proceedings of the 24th annual international ACM SIGIR conference on Research and development in information retrieval
5588:
5517:
5247:
2097:
1726:
1694:
1592:
1119:
201:
100:
4062:
2210:
2184:
1260:(LDA), etc. Their uses have consistently aimed to provide customers with more precise and tailored recommendations.
4146:
Ziegler CN, McNee SM, Konstan JA, Lausen G (2005). "Improving recommendation lists through topic diversification".
1365:
346:
80:
1610:"A capable multimedia content discovery platform based on visual content analysis and intelligent data enrichment"
799:
instance, it would be unwise to recommend a recipe in an area where all of the ingredients may not be available).
4782:
4232:
978:
1191:: Based on the computed distances, find k nearest neighbors of the user to which we want to make recommendations
5642:
5087:
Khanal, S.S. (July 2020). "A systematic review: machine learning based recommendation systems for e-learning".
3726:
2241:
1341:-connected devices, consumers are projected to have access to content from linear broadcast sources as well as
1313:
and Remesh which have been used around the world to help find more consensus around specific political issues.
1101:
4908:
Proceedings of the International Workshop on Reproducibility and Replication in Recommender Systems Evaluation
4352:"The Impact of Demographics (Age and Gender) and Other User Characteristics on Evaluating Recommender Systems"
4012:
Proceedings of the International Workshop on Reproducibility and Replication in Recommender Systems Evaluation
3660:
2566:
Proceedings of the International Workshop on Reproducibility and Replication in Recommender Systems Evaluation
2502:
Proceedings of the International Workshop on Reproducibility and Replication in Recommender Systems Evaluation
4064:
Should I Follow the Crowd? A Probabilistic Analysis of the Effectiveness of Popularity in Recommender Systems
2394:"Toward the Next Generation of Recommender Systems: A Survey of the State-of-the-Art and Possible Extensions"
1165:
785:
658:
450:
314:
and content-based filtering (also known as the personality-based approach), as well as other systems such as
3433:
Lakiotaki, K.; Matsatsinis; Tsoukias, A (March 2011). "Multicriteria User Modeling in Recommender Systems".
2688:
Koren, Yehuda; Volinsky, Chris (August 1, 2009). "Matrix Factorization Techniques for Recommender Systems".
2349:
Montaner, M.; Lopez, B.; de la Rosa, J. L. (June 2003). "A Taxonomy of Recommender Agents on the Internet".
2339:." In Proceedings of the 1994 ACM conference on Computer supported cooperative work, pp. 175–186. ACM, 1994.
1828:
986:
recommendations) and other data sources can be used to uncover identities of users in an anonymized dataset.
375:
Another early recommender system, called a "digital bookshelf", was described in a 1990 technical report by
5647:
5637:
4727:"Are We Evaluating Rigorously? Benchmarking Recommendation for Reproducible Evaluation and Fair Comparison"
4510:
Proceedings of the first international joint conference on Autonomous agents and multiagent systems: part 1
1565:
Resnick, Paul, and Hal R. Varian. "Recommender systems." Communications of the ACM 40, no. 3 (1997): 56–58.
1395:
1329:
that are representative of the platform's users to control the design and implementation of the algorithm.
1253:
861:
246:
194:
5048:
2020 2nd International Conference on Advances in Computing, Communication Control and Networking (ICACCCN)
3947:. Lecture Notes in Computer Science. Vol. 9316. Springer International Publishing. pp. 153–168.
5565:
2163:
System and method for providing recommendation of goods and services based on recorded purchasing history
1719:
Proceedings of the 19th National Conference on Innovative Applications of Artificial Intelligence, vol. 2
1257:
1147:
131:
4532:
Proceedings of the 17th International Conference on Theory and Practice of Digital Libraries (TPDL 2013)
4473:
4359:
Proceedings of the 17th International Conference on Theory and Practice of Digital Libraries (TPDL 2013)
4210:
Proceedings of the 17th International Conference on Theory and Practice of Digital Libraries (TPDL 2013)
3761:
Turpin, Andrew H; Hersh, William (2001). "Why batch and user evaluations do not give the same results".
3608:
3486:
262:
and search queries. There are also popular recommender systems for specific topics like restaurants and
4530:. In Trond Aalberg, Milena Dobreva, Christos Papatheodorou, Giannis Tsakonas, Charles Farrugia (eds.).
4357:. In Trond Aalberg; Milena Dobreva; Christos Papatheodorou; Giannis Tsakonas; Charles Farrugia (eds.).
4208:. In Trond Aalberg; Milena Dobreva; Christos Papatheodorou; Giannis Tsakonas; Charles Farrugia (eds.).
2124:
1355:
909:
629:
396:
136:
4381:
4203:"Persistence in Recommender Systems: Giving the Same Recommendations to the Same Users Multiple Times"
4448:
4262:
3696:
3365:
Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence (IJCAI-19)
1207:
4612:"Are we really making much progress? A worrying analysis of recent neural recommendation approaches"
4020:
3901:
3447:
3384:
Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining
3299:
Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining
3121:
Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining
3000:
2702:
2410:
1300:
881:
Evaluation is important in assessing the effectiveness of recommendation algorithms. To measure the
4772:"Using Deep Learning to Win the Booking.com WSDM WebTour21 Challenge on Sequential Recommendations"
3634:
2456:
1508:
1405:
1249:
865:
738:
498:
461:
402:
Montaner provided the first overview of recommender systems from an intelligent agent perspective.
95:
90:
5468:
5346:
4309:
3847:
2793:
Felício, Crícia Z.; Paixão, Klérisson V.R.; Barcelos, Celia A.Z.; Preux, Philippe (July 9, 2017).
2649:. In Proceedings of the Fourteenth conference on Uncertainty in artificial intelligence (UAI'98).
1306:
4553:
Ferrari Dacrema, Maurizio; Boglio, Simone; Cremonesi, Paolo; Jannach, Dietmar (January 8, 2021).
4282:
4165:
2162:
1375:
1370:
1158:
1132:
1090:
901:
674:
438:
424:
315:
311:
254:
141:
85:
49:
4997:
Said, Alan; Bellogín, Alejandro (October 1, 2014). "Comparative recommender system evaluation".
4283:
Naren Ramakrishnan; Benjamin J. Keller; Batul J. Mirza; Ananth Y. Grama; George Karypis (2001).
3892:
3884:
2231:
RICH, Elaine. User modeling via stereotypes. Cognitive science, 1979, 3. Jg., Nr. 4, S. 329–354.
2089:
2069:
2063:
1463:
710:: Recommendations from different recommenders are presented together to give the recommendation.
337:
Pandora uses the properties of a song or artist (a subset of the 400 attributes provided by the
4304:
4300:
4284:
4015:
3896:
3842:
3748:
3677:
3442:
2898:
2697:
2451:
2405:
1425:
1415:
1385:
1326:
1037:
617:
522:
460:
When building a model from a user's behavior, a distinction is often made between explicit and
110:
3838:
3523:
Yong Ge; Hui Xiong; Alexander Tuzhilin; Keli Xiao; Marco Gruteser; Michael J. Pazzani (2010).
904:, the latter having been used in the Netflix Prize. The information retrieval metrics such as
1400:
1232:: sequence of pages visited, time spent on different parts of a website, mouse movement, etc.
1000:– When users can participate in the recommender system, the issue of fraud must be addressed.
650:
589:
585:
372:
users' stereotype membership, they would then get recommendations for books they might like.
59:
5618:
5264:
4263:"Evaluating recommender systems from the user's perspective: survey of the state of the art"
3830:
3301:. KDD '18. London, United Kingdom: Association for Computing Machinery. pp. 1831–1839.
3238:
Li, Jing; Ren, Pengjie; Chen, Zhumin; Ren, Zhaochun; Lian, Tao; Ma, Jun (November 6, 2017).
3168:
Proceedings of the 27th ACM International Conference on Information and Knowledge Management
2890:
5493:
4807:
Volkovs, Maksims; Rai, Himanshu; Cheng, Zhaoyue; Wu, Ga; Lu, Yichao; Sanner, Scott (2018).
4429:
3524:
3246:. CIKM '17. Singapore, Singapore: Association for Computing Machinery. pp. 1419–1428.
2798:
1679:
Proceedings of the 11th Annual International ACM/IEEE Joint Conference on Digital Libraries
1671:
1182:
905:
338:
146:
3837:. Lecture Notes in Computer Science. Vol. 7899. Springer Berlin Heidelberg. pp.
2980:
X.Y. Feng, H. Zhang, Y.J. Ren, P.H. Shang, Y. Zhu, Y.C. Liang, R.C. Guan, D. Xu, (2019), "
2316:
8:
5265:""Extending and Customizing Content Discovery for the Legal Academic Com" by Sima Mirkin"
4349:
4200:
3532:. Proceedings of the 16th ACM SIGKDD Int'l Conf. on Knowledge Discovery and Data Mining.
1967:
1420:
1410:
1342:
974:
802:
One example of a mobile recommender system are the approaches taken by companies such as
454:
5547:
5540:
4433:
2671:
2393:
2211:
Automated detection and exposure of behavior-based relationships between browsable items
5352:
5208:
5167:
5104:
5069:
5028:
4979:
4950:
Breitinger, Corinna; Langer, Stefan; Lommatzsch, Andreas; Gipp, Bela (March 12, 2016).
4929:
4888:
4834:
4752:
4685:
4655:
4619:
4592:
4566:
4486:
4406:
4332:
4243:
4128:
4043:
3924:
3810:
3586:
3568:
3460:
3415:
3387:
3361:"SlateQ: A Tractable Decomposition for Reinforcement Learning with Recommendation Sets"
3339:
3320:
3275:
3247:
3218:
3199:
3171:
3144:
3096:
3070:
2825:
2806:. UMAP '17. Bratislava, Slovakia: Association for Computing Machinery. pp. 32–40.
2715:
2650:
2627:
2588:
2523:
2469:
2423:
2389:
2366:
2144:
1945:
1907:
1786:
1639:
1024:
930:
897:
773:
654:
533:
278:
167:
5055:
4951:
4555:"A Troubling Analysis of Reproducibility and Progress in Recommender Systems Research"
4212:. Lecture Notes of Computer Science (LNCS). Vol. 8092. Springer. pp. 390–394
3217:
Kang, Wang-Cheng; McAuley, Julian (2018). "Self-Attentive Sequential Recommendation".
2781:
Discovery of Hidden Similarity on Collaborative Filtering to Overcome Sparsity Problem
2607:
2564:
5614:
5555:
5530:
5513:
5395:
5243:
5108:
5073:
5059:
5046:
Verma, P.; Sharma, S. (2020). "Artificial Intelligence based Recommendation System".
5018:
4971:
4919:
4878:
4824:
4756:
4742:
4705:
4659:
4645:
4596:
4505:
4490:
4478:
4322:
4181:
4132:
4033:
3989:
3956:
3914:
3860:
3831:
3405:
3380:"Reinforcement Learning to Optimize Long-term User Engagement in Recommender Systems"
3378:
Zou, Lixin; Xia, Long; Ding, Zhuoye; Song, Jiaxing; Liu, Weidong; Yin, Dawei (2019).
3310:
3265:
3189:
3148:
3134:
2902:
2891:
2886:
2866:
2815:
2631:
2578:
2513:
2473:
2427:
2148:
2093:
2034:
1983:
1911:
1899:
1847:
1778:
1722:
1690:
1631:
1588:
1479:
1287:
1054:
514:
Obtaining a list of items that a user has listened to or watched on his/her computer.
489:
115:
5032:
4838:
4725:
Sun, Zhu; Yu, Di; Fang, Hui; Yang, Jie; Qu, Xinghua; Zhang, Jie; Geng, Cong (2020).
3928:
3814:
3590:
3464:
3419:
3324:
3279:
2719:
2383:
2381:
2370:
1949:
1643:
776:. This system combines a content-based technique and a contextual bandit algorithm.
5586:
5291:"Mendeley, Elsevier and the importance of content discovery to academic publishers"
5218:
5177:
5138:
5096:
5051:
5010:
5002:
4963:
4911:
4892:
4870:
4816:
4734:
4695:
4637:
4629:
4584:
4576:
4525:"Sponsored vs. Organic (Research Paper) Recommendations and the Impact of Labeling"
4468:
4460:
4410:
4396:
4336:
4314:
4247:
4173:
4120:
4071:
4047:
4025:
3981:
3948:
3906:
3852:
3802:
3578:
3504:
3452:
3397:
3302:
3257:
3203:
3181:
3124:
3026:
2935:
2807:
2800:
Proceedings of the 25th Conference on User Modeling, Adaptation and Personalization
2707:
2619:
2592:
2570:
2527:
2505:
2461:
2415:
2358:
2335:
Resnick, Paul, Neophytos Iacovou, Mitesh Suchak, Peter Bergström, and John Riedl. "
2197:
2136:
2082:
2046:
2012:
1975:
1937:
1891:
1839:
1790:
1768:
1682:
1621:
1580:
1543:
1471:
1139:
621:
484:
Presenting two items to a user and asking him/her to choose the better one of them.
354:
298:
5197:"Explaining Deep Neural Networks and Beyond: A Review of Methods and Applications"
4983:
4933:
4464:
4447:
Möller, Judith; Trilling, Damian; Helberger, Natali; van Es, Bram (July 3, 2018).
4350:
Joeran Beel; Stefan Langer; Andreas Nürnberger; Marcel Genzmehr (September 2013).
4201:
Joeran Beel; Stefan Langer; Marcel Genzmehr; Andreas Nürnberger (September 2013).
3170:. CIKM '18. Torino, Italy: Association for Computing Machinery. pp. 843–852.
2829:
2140:
1843:
852:. 4-Tell, Inc. created a Netflix project–derived solution for ecommerce websites.
4808:
4726:
4177:
3952:
3684:
3379:
3244:
Proceedings of the 2017 ACM on Conference on Information and Knowledge Management
3053:
3003:, Semantic Web – Interoperability, Usability, Applicability 1 (2010) 1, IOS Press
2940:
2923:
2849:
2378:
2323:
2268:
2248:
2050:
2016:
1979:
1732:
1318:
926:
836:
Predictive accuracy is substantially improved when blending multiple predictors.
560:
465:
5548:
Jannach, Dietmar; Markus Zanker; Alexander Felfernig; Gerhard Friedrich (2010).
4233:"Is seeing believing?: how recommender system interfaces affect users' opinions"
3856:
2558:"Research paper recommender system evaluation: A quantitative literature survey"
1711:
1475:
1226:: what specify time and date or a season that a user interacts with the platform
5223:
5196:
5143:
5126:
5100:
4124:
4103:
3985:
2261:
1548:
1531:
1430:
1275:
845:
811:
points along a route, with the goal of optimizing occupancy times and profits.
678:
625:
517:
Analyzing the user's social network and discovering similar likes and dislikes.
380:
376:
151:
5443:"YouTube Adding Experimental Community Notes Feature to Battle Misinformation"
5181:
4967:
4401:
3164:"Recurrent Neural Networks with Top-k Gains for Session-based Recommendations"
2623:
2362:
2081:
Conference on Research and Development in Information Retrieval (SIGIR 2002).
1895:
1626:
1609:
632:
in order to estimate the probability that the user is going to like the item.
5631:
5593:
4975:
4482:
3582:
3533:
3477:
2922:
Wang, Donghui; Liang, Yanchun; Xu, Dong; Feng, Xiaoyue; Guan, Renchu (2018).
2556:
Beel, J.; Langer, S.; Genzmehr, M.; Gipp, B.; Breitinger, C. (October 2013).
1903:
1635:
1390:
1143:
982:
970:
882:
826:
820:
662:
326:
294:
290:
263:
177:
5622:
Proceedings of the Eighteenth National Conference on Artificial Intelligence
5006:
4915:
4874:
4820:
4738:
4700:
4633:
4075:
4029:
3806:
3776:
3401:
3306:
3261:
3185:
3129:
2811:
2735:"Application of Dimensionality Reduction in Recommender System A Case Study"
2734:
2574:
2509:
2035:"A survey of active learning in collaborative filtering recommender systems"
1686:
1584:
1286:
Google Scholar provides an 'Updates' tool that suggests articles by using a
481:
Asking a user to rank a collection of items from favorite to least favorite.
5587:
Robert M. Bell; Jim Bennett; Yehuda Koren & Chris Volinsky (May 2009).
3522:
1782:
1380:
596:
580:
392:
5598:
5318:"Social media algorithms can be redesigned to bridge divides — here's how"
4240:
Proceedings of the SIGCHI conference on Human factors in computing systems
4230:
3910:
3609:"A $ 1 Million Research Bargain for Netflix, and Maybe a Model for Others"
3294:
3239:
3163:
2794:
2669:
2465:
429:
5473:
Belfer Center for Science and International Affairs at Harvard University
4641:
4588:
4554:
4079:
3360:
2795:"A Multi-Armed Bandit Model Selection for Cold-Start User Recommendation"
2419:
1435:
1069:
1007:
890:
885:
of recommender systems, and compare different approaches, three types of
868:, led to the cancellation of a second Netflix Prize competition in 2010.
825:
One of the events that energized research in recommender systems was the
704:: Choosing among recommendation components and applying the selected one.
698:: Combining the score of different recommendation components numerically.
646:
388:
368:
302:
with relevant academic content and serendipitously discover new content.
5512:
Kim Falk (d 2019), Practical Recommender Systems, Manning Publications,
5014:
4318:
3456:
2711:
1869:
Content-based book recommendation using learning for text categorization
1710:
Felfernig, Alexander; Isak, Klaus; Szabo, Kalman; Zachar, Peter (2007).
1670:
Chen, Hung-Hsuan; Gou, Liang; Zhang, Xiaolong; Giles, Clyde Lee (2011).
5120:
5118:
4611:
4105:
3293:
Liu, Qiao; Zeng, Yifu; Mokhosi, Refuoe; Zhang, Haibin (July 19, 2018).
2672:
Empirical Analysis of Predictive Algorithms for Collaborative Filtering
2647:
Empirical analysis of predictive algorithms for collaborative filtering
2644:
2304:
Social information filtering: algorithms for automating "word of mouth"
2065:
1104: in this section. Unsourced material may be challenged and removed.
886:
791:
612:
553:
449:
similarity or item similarity in recommender systems. For example, the
437:
One approach to the design of recommender systems that has wide use is
4610:
Ferrari Dacrema, Maurizio; Cremonesi, Paolo; Jannach, Dietmar (2019).
2981:
2924:"A content-based recommender system for computer science publications"
2768:. AAAI Workshop in Semantic Web Personalization, San Jose, California.
2337:
GroupLens: an open architecture for collaborative filtering of netnews
2185:
System and method for providing access to data using customer profiles
1757:"How to tame the flood of literature : Nature News & Comment"
1658:
ExpertSeer: a Keyphrase Based Expert Recommender for Digital Libraries
5619:
Content-Boosted Collaborative Filtering for Improved Recommendations.
4552:
1532:"A systematic review and research perspective on recommender systems"
1338:
1211:
521:
Collaborative filtering approaches often suffer from three problems:
172:
64:
54:
5115:
4580:
3978:
Proceedings of the 2017 SIAM International Conference on Data Mining
3031:
3014:
2843:
Collaborative Recommendations Using Item-to-Item Similarity Mappings
2262:
Newsgroup Clustering Based On User Behavior-A Recommendation Algebra
1966:
Rubens, Neil; Elahi, Mehdi; Sugiyama, Masashi; Kaplan, Dain (2016).
1941:
1773:
1756:
1079:
360:
Recommender systems have been the focus of several granted patents.
5417:"YouTube's community notes feature rips a page out of X's playbook"
5213:
5172:
4690:
4624:
4571:
4231:
Cosley, D.; Lam, S.K.; Albert, I.; Konstan, J.A.; Riedl, J (2003).
3573:
3392:
3344:
3252:
3223:
3176:
3101:
3075:
2957:"Online Recommender Systems – How Does a Website Know What I Want?"
2198:
Playlist-based detection of similar digital works and work creators
1882:
Haupt, Jon (June 1, 2009). "Last.fm: People-Powered Online Radio".
1657:
641:
282:
274:
258:
5574:
Computing Taste: Algorithms and the Makers of Music Recommendation
4504:
Montaner, Miquel; López, Beatriz; de la Rosa, Josep Lluís (2002).
4148:
Proceedings of the 14th international conference on World Wide Web
4104:
Cañamares, Rocío; Castells, Pablo; Moffat, Alistair (March 2020).
2766:
Using viewing time to infer user preference in recommender systems
2655:
1577:
Proceedings of the 22nd International Conference on World Wide Web
5271:. Digital Commons @ American University Washington College of Law
4769:
4523:
Beel, Joeran, Langer, Stefan, Genzmehr, Marcel (September 2013).
2317:
Recommending and evaluating choices in a virtual community of use
1608:
Baran, Remigiusz; Dziech, Andrzej; Zeja, Andrzej (June 1, 2018).
1322:
1314:
1181:: 'Distance' measures how far apart users are in this space. See
744:
684:
322:
286:
3697:"Netflix Spilled Your Brokeback Mountain Secret, Lawsuit Claims"
2606:
Beel, J.; Gipp, B.; Langer, S.; Breitinger, C. (July 26, 2015).
606:
A history of the user's interaction with the recommender system.
4609:
4060:
1464:"Recommender Systems: Techniques, Applications, and Challenges"
1310:
1279:
579:
In this system, keywords are used to describe the items, and a
487:
Asking a user to create a list of items that he/she likes (see
4949:
4867:
Proceedings of the fifth ACM conference on Recommender systems
4809:"Two-stage Model for Automatic Playlist Continuation at Scale"
3760:
3068:
2315:
Hill, Will, Larry Stead, Mark Rosenstein, and George Furnas. "
864:
by releasing the datasets. This, as well as concerns from the
772:, a system which models the context-aware recommendation as a
673:
Most recommender systems now use a hybrid approach, combining
572:
Another common approach when designing recommender systems is
433:
An example of collaborative filtering based on a rating system
395:, also at MIT, whose work with GroupLens was awarded the 2010
4616:
Proceedings of the 13th ACM Conference on Recommender Systems
4446:
4168:. In Ricci, Francesco; Rokach, Lior; Shapira, Bracha (eds.).
3885:"Why batch and user evaluations do not give the same results"
2196:
Harbick, Andrew V., Ryan J. Snodgrass, and Joel R. Spiegel. "
2002:
1970:. In Ricci, Francesco; Rokach, Lior; Shapira, Bracha (eds.).
1466:. In Ricci, Francesco; Rokach, Lior; Shapira, Bracha (eds.).
4999:
Proceedings of the 8th ACM Conference on Recommender systems
4008:
3558:
3432:
3162:
Hidasi, Balázs; Karatzoglou, Alexandros (October 17, 2018).
2441:
2209:
Linden, Gregory D., Brent Russell Smith, and Nida K. Zada. "
2183:
Herz, Frederick, Lyle Ungar, Jian Zhang, and David Wachob. "
471:
Examples of explicit data collection include the following:
5624:(AAAI-2002), pp. 187–192, Edmonton, Canada, July 2002.
4679:
3123:. Association for Computing Machinery. pp. 2291–2299.
3118:
2283:"A digital bookshelf: original work on recommender systems"
857:
807:
803:
511:
Keeping a record of the items that a user purchases online.
384:
4864:
3827:
3625:
3337:
2792:
2123:
Bi, Xuan; Qu, Annie; Wang, Junhui; Shen, Xiaotong (2017).
1138:
Recommendation systems widely adopt AI techniques such as
840:
Consequently, our solution is an ensemble of many methods.
790:
Mobile recommender systems make use of internet-accessing
741:, Transformers, and other deep-learning-based approaches.
27:
Information filtering system to predict users' preferences
5522:
4952:"Towards reproducibility in recommender-systems research"
4813:
Proceedings of the ACM Recommender Systems Challenge 2018
4382:"Recommender systems: from algorithms to user experience"
4145:
2732:
2645:
John S. Breese; David Heckerman & Carl Kadie (1998).
2608:"Research Paper Recommender Systems: A Literature Survey"
1965:
1681:. Association for Computing Machinery. pp. 231–240.
1672:"CollabSeer: a search engine for collaboration discovery"
1579:. Association for Computing Machinery. pp. 505–514.
599:, the system mostly focuses on two types of information:
505:
Observing the items that a user views in an online store.
5001:. RecSys '14. New York, NY, USA: ACM. pp. 129–136.
4869:. RecSys '11. New York, NY, USA: ACM. pp. 133–140.
4423:
3976:
Basaran, Daniel; Ntoutsi, Eirini; Zimek, Arthur (2017).
3013:
Gomez-Uribe, Carlos A.; Hunt, Neil (December 28, 2015).
2733:
Sarwar, B.; Karypis, G.; Konstan, J.; Riedl, J. (2000).
2605:
1709:
1462:
Ricci, Francesco; Rokach, Lior; Shapira, Bracha (2022).
4503:
4164:
Castells, Pablo; Hurley, Neil J.; Vargas, Saúl (2015).
3478:
Gediminas Adomavicius; Nikos Manouselis; YoungOk Kwon.
3094:
2670:
Breese, John S.; Heckerman, David; Kadie, Carl (1998).
2348:
1806:"Netflix Revamps iPad App to Improve Content Discovery"
1712:"The VITA Financial Services Sales Support Environment"
310:
Recommender systems usually make use of either or both
4910:. RepSys '13. New York, NY, USA: ACM. pp. 23–28.
4779:
WSDM '21: ACM Conference on Web Search and Data Mining
4522:
3945:
Research and Advanced Technology for Digital Libraries
2555:
1070:
Artificial intelligence applications in recommendation
281:
which uses recommender system tools. It utilizes user
5269:
Articles in Law Reviews & Other Academic Journals
4014:. RepSys '13. New York, NY, USA: ACM. pp. 7–14.
3883:
Turpin, Andrew H.; Hersh, William (January 1, 2001).
3725:. Netflix Prize Forum. March 12, 2010. Archived from
3358:
2763:
2033:
Elahi, Mehdi; Ricci, Francesco; Rubens, Neil (2016).
753:
5377:"Elon Musk keeps Birdwatch alive — under a new name"
4260:
4163:
4106:"Offline Evaluation Options for Recommender Systems"
1219:
boost user experience. Following are some examples:
731:
668:
5160:
IEEE Transactions on Knowledge and Data Engineering
3975:
3292:
2764:Parsons, J.; Ralph, P.; Gallagher, K. (July 2004).
2398:
IEEE Transactions on Knowledge and Data Engineering
2387:
977:, and every attempt to introduce any level of user
5375:Smalley, Alex Mahadevan, Seth (November 8, 2022).
4905:
3969:
3792:
3747:Lathia, N., Hailes, S., Capra, L., Amatriain, X.:
3518:
3516:
3019:ACM Transactions on Management Information Systems
2187:." U.S. Patent 8,056,100, issued November 8, 2011.
2071:Methods and Metrics for Cold-Start Recommendations
1866:
1574:
1461:
1153:
5316:Thorburn, Luke; Ovadya, Aviv (October 31, 2023).
4806:
3509:(Ph.D.), Institut National des Télécommunications
3161:
2954:
2885:
2779:Sanghack Lee and Jihoon Yang and Sung-Yong Park,
2493:Beel, J.; Genzmehr, M.; Gipp, B. (October 2013).
1930:ACM Transactions on Knowledge Discovery from Data
1826:
1669:
1470:(3 ed.). New York: Springer. pp. 1–35.
762:
475:Asking a user to rate an item on a sliding scale.
5629:
5127:"Artificial intelligence in recommender systems"
4731:Fourteenth ACM Conference on Recommender Systems
4682:Fourteenth ACM Conference on Recommender Systems
4424:Ricci F, Rokach L, Shapira B, Kantor BP (2011).
2492:
2032:
1871:. In Workshop Recom. Sys.: Algo. and Evaluation.
1501:"How Computers Know What We Want — Before We Do"
353:Recommender systems are a useful alternative to
4276:
4159:
4157:
4061:Cañamares, Rocío; Castells, Pablo (July 2018).
3513:
3240:"Neural Attentive Session-based Recommendation"
3012:
2921:
2836:
2213:." U.S. Patent 9,070,156, issued June 30, 2015.
2200:." U.S. Patent 8,468,046, issued June 18, 2013.
2129:Journal of the American Statistical Association
2074:. Proceedings of the 25th Annual International
1820:
1607:
1561:
1559:
848:, a recommendation engine that's active in the
5315:
4166:"Novelty and Diversity in Recommender Systems"
3891:. SIGIR '01. New York, NY, USA: ACM. pp.
3833:User Modeling, Adaptation, and Personalization
3602:
3600:
3506:DRARS, A Dynamic Risk-Aware Recommender System
3496:
3377:
2854:
2165:." U.S. Patent 7,222,085, issued May 22, 2007.
1242:
745:Reinforcement learning for recommender systems
4724:
4172:(2 ed.). Springer US. pp. 881–918.
4097:
4054:
3526:An Energy-Efficient Mobile Recommender System
2687:
1974:(2 ed.). Springer US. pp. 809–846.
1961:
1959:
1834:. In Claude Sammut; Geoffrey I. Webb (eds.).
1268:
779:
202:
5400:: CS1 maint: multiple names: authors list (
4996:
4497:
4474:11245.1/4242e2e0-3beb-40a0-a6cb-d8947a13efb4
4379:
4154:
3237:
2122:
2028:
2026:
1556:
1498:
1317:has also used this approach for manging its
253:Typically, the suggestions refer to various
5240:Introduction to natural language processing
5045:
4373:
4139:
3882:
3628:"The BellKor solution to the Netflix Prize"
3597:
3216:
1703:
5237:
4956:User Modeling and User-Adapted Interaction
4516:
4389:User Modeling and User-Adapted Interaction
4270:User Modeling and User-Adapted Interaction
4224:
3754:
3502:
2612:International Journal on Digital Libraries
1956:
1750:
1748:
1746:
1744:
1742:
1568:
1457:
1455:
1453:
1451:
567:
443:matrix factorization (recommender systems)
418:
209:
195:
5222:
5212:
5171:
5142:
4699:
4689:
4623:
4570:
4472:
4417:
4400:
4343:
4308:
4194:
4019:
3900:
3846:
3749:Temporal diversity in recommender systems
3572:
3446:
3391:
3343:
3251:
3222:
3175:
3128:
3100:
3074:
3030:
2939:
2701:
2654:
2455:
2409:
2118:
2116:
2023:
1827:Melville, Prem; Sindhwani, Vikas (2010).
1772:
1663:
1650:
1625:
1547:
1120:Learn how and when to remove this message
5414:
4506:"Developing trust in recommender agents"
4453:Information, Communication & Society
3715:
3661:"Mátrixfaktorizáció one million dollars"
2860:
2638:
2280:
2057:
1968:"Active Learning in Recommender Systems"
1927:
1860:
1803:
1529:
1263:
428:
5374:
4559:ACM Transactions on Information Systems
3626:R. Bell; Y. Koren; C. Volinsky (2007).
3114:
3112:
3039:
2773:
2753:. International J. Man-Machine Studies.
2302:Shardanand, Upendra, and Pattie Maes. "
2064:Andrew I. Schein; Alexandrin Popescul;
1739:
1656:H. Chen, A. G. Ororbia II, C. L. Giles
1448:
876:
691:Some hybridization techniques include:
14:
5630:
5589:"The Million Dollar Programming Prize"
5466:
5344:
5340:
5338:
5262:
5086:
4945:
4943:
4855:, Deep Learning Re-Work SF Summit 2018
4297:IEEE Educational Activities Department
4285:"Privacy risks in recommender systems"
4254:
3941:
3658:
3001:The Knowledge Reengineering Bottleneck
2251:. Syslab Working Paper 179 (1990). "
2113:
5440:
5194:
5124:
4853:Deep Learning for Recommender Systems
4845:
4618:. RecSys '19. ACM. pp. 101–109.
3878:
3876:
3788:
3786:
3090:
3088:
3086:
2751:User Models: Theory, Method, Practice
2748:
2488:
2486:
2484:
2125:"A group-specific recommender system"
1923:
1921:
1881:
1361:ACM Conference on Recommender Systems
1301:Criticism of Google § Algorithms
1238:: information from outer social media
183:ACM Conference on Recommender Systems
5551:Recommender Systems: An Introduction
5523:Bharat Bhasker; K. Srikumar (2010).
3606:
3480:"Multi-Criteria Recommender Systems"
3109:
3064:
3062:
2987:Journal of Medical Internet Research
1754:
1530:Roy, Deepjyoti; Dutta, Mala (2022).
1102:adding citations to reliable sources
1073:
814:
5335:
4940:
3331:
391:at MIT, Will Hill at Bellcore, and
24:
5501:
5157:
3873:
3795:ACM SIGKDD Explorations Newsletter
3783:
3607:Lohr, Steve (September 22, 2009).
3371:
3352:
3083:
2955:Blanda, Stephanie (May 25, 2015).
2481:
1918:
1884:Music Reference Services Quarterly
1867:R. J. Mooney & L. Roy (1999).
1294:
1201:
1195:Forming Predictive Recommendations
1031:
943:
896:The commonly used metrics are the
754:Multi-criteria recommender systems
508:Analyzing item/user viewing times.
25:
5664:
5617:, and Ramadass Nagarajan. (2002)
5526:Recommender Systems in E-Commerce
5056:10.1109/ICACCCN51052.2020.9362962
4261:Pu, P.; Chen, L.; Hu, R. (2012).
3659:Bodoky, Thomas (August 6, 2009).
3059:
2863:Recommender Systems: The Textbook
1614:Multimedia Tools and Applications
732:Session-based recommender systems
669:Hybrid recommendations approaches
603:A model of the user's preference.
101:Item-item collaborative filtering
5487:
5460:
5434:
5415:Shanklin, Will (June 17, 2024).
5408:
5368:
5309:
5283:
5256:
5231:
5188:
5151:
5080:
5039:
4990:
4899:
4858:
4800:
4763:
4718:
4673:
4603:
4546:
3015:"The Netflix Recommender System"
2281:Karlgren, Jussi (October 2017).
1836:Encyclopedia of Machine Learning
1078:
5238:Eisenstein, J. (October 2019).
5131:Complex and Intelligent Systems
4440:
4002:
3935:
3821:
3769:
3741:
3689:
3671:
3652:
3619:
3552:
3471:
3426:
3386:. KDD '19. pp. 2810–2818.
3286:
3231:
3210:
3155:
3006:
2993:
2974:
2948:
2915:
2879:
2786:
2757:
2742:
2726:
2681:
2663:
2599:
2549:
2435:
2342:
2329:
2309:
2296:
2274:
2271:." SICS Research Report (1994).
2254:
2242:An Algebra for Recommendations.
2234:
2225:
2216:
2203:
2190:
2177:
2168:
2155:
1996:
1875:
1797:
1154:KNN-based collaborative filters
1089:needs additional citations for
726:
457:as first implemented by Allen.
5576:. University of Chicago Press.
4851:Yves Raimond, Justin Basilico
3561:Journal of Information Science
3047:Hybrid Web Recommender Systems
2351:Artificial Intelligence Review
1838:. Springer. pp. 829–838.
1804:Analysis (December 14, 2011).
1601:
1523:
1492:
891:online evaluations (A/B tests)
858:Internet Movie Database (IMDb)
763:Risk-aware recommender systems
13:
1:
5653:Social information processing
5467:Ovadya, Aviv (May 17, 2022).
5441:Novak, Matt (June 17, 2024).
5345:Ovadya, Aviv (May 17, 2022).
5263:Mirkin, Sima (June 4, 2014).
4465:10.1080/1369118X.2018.1444076
2961:American Mathematical Society
2677:(Report). Microsoft Research.
2141:10.1080/01621459.2016.1219261
1844:10.1007/978-0-387-30164-8_705
1767:(7516). Nature.com: 129–130.
1717:. In William Cheetham (ed.).
1499:Lev Grossman (May 27, 2010).
1442:
1332:
889:are available: user studies,
871:
786:Location based recommendation
659:Multimodal sentiment analysis
645:various techniques including
525:, scalability, and sparsity.
493:or other similar techniques).
413:
387:, and research groups led by
4426:Recommender systems handbook
4380:Konstan JA, Riedl J (2012).
4361:. Springer. pp. 400–404
4178:10.1007/978-1-4899-7637-6_26
4170:Recommender Systems Handbook
3953:10.1007/978-3-319-24592-8_12
3503:Bouneffouf, Djallel (2013),
2941:10.1016/j.knosys.2018.05.001
2051:10.1016/j.cosrev.2016.05.002
2017:10.1016/j.knosys.2011.06.005
1980:10.1007/978-1-4899-7637-6_24
1972:Recommender Systems Handbook
1468:Recommender Systems Handbook
1396:Information filtering system
1254:singular value decomposition
1168:. The ideas are as follows:
862:Video Privacy Protection Act
534:multi-armed bandit algorithm
247:information filtering system
7:
5125:Zhang, Q. (February 2021).
3857:10.1007/978-3-642-38844-6_3
3678:Rise of the Netflix Hackers
2861:Aggarwal, Charu C. (2016).
2068:; David M. Pennock (2002).
1476:10.1007/978-1-0716-2197-4_1
1348:
1258:latent Dirichlet allocation
1243:Natural language processing
1148:natural language processing
893:, and offline evaluations.
305:
132:Collaborative search engine
10:
5669:
5224:10.1109/JPROC.2021.3060483
5144:10.1007/s40747-020-00212-w
5101:10.1007/s10639-019-10063-9
4125:10.1007/s10791-020-09371-3
3986:10.1137/1.9781611974973.44
2783:, Discovery Science, 2007.
2267:February 27, 2021, at the
1755:jobs (September 3, 2014).
1549:10.1186/s40537-022-00592-5
1356:Algorithmic radicalization
1298:
1269:Academic content discovery
818:
783:
780:Mobile recommender systems
630:artificial neural networks
422:
397:ACM Software Systems Award
363:
271:content discovery platform
137:Content discovery platform
18:Content discovery platform
5182:10.1109/TKDE.2022.3145690
4968:10.1007/s11257-016-9174-x
4402:10.1007/s11257-011-9112-x
3683:January 24, 2012, at the
2624:10.1007/s00799-015-0156-0
1896:10.1080/10588160902816702
1627:10.1007/s11042-017-5014-1
1208:artificial neural network
1185:for computational details
856:with film ratings on the
739:recurrent neural networks
255:decision-making processes
5469:"Bridging-Based Ranking"
5347:"Bridging-Based Ranking"
5195:Samek, W. (March 2021).
4119:(4). Springer: 387–410.
3583:10.1177/0165551518792213
3435:IEEE Intelligent Systems
1660:, in arXiv preprint 2015
1406:Media monitoring service
1250:latent semantic analysis
1230:User Navigation Patterns
866:Federal Trade Commission
499:implicit data collection
478:Asking a user to search.
453:(k-NN) approach and the
96:Implicit data collection
91:Dimensionality reduction
5201:Proceedings of the IEEE
5007:10.1145/2645710.2645746
4916:10.1145/2532508.2532513
4875:10.1145/2043932.2043958
4821:10.1145/3267471.3267480
4739:10.1145/3383313.3412489
4733:. ACM. pp. 23–32.
4701:10.1145/3383313.3412488
4634:10.1145/3298689.3347058
4289:IEEE Internet Computing
4076:10.1145/3209978.3210014
4030:10.1145/2532508.2532511
3807:10.1145/3137597.3137601
3534:New York City, New York
3402:10.1145/3292500.3330668
3307:10.1145/3219819.3219950
3262:10.1145/3132847.3132926
3186:10.1145/3269206.3271761
3130:10.1145/3394486.3403278
2928:Knowledge-Based Systems
2812:10.1145/3079628.3079681
2575:10.1145/2532508.2532512
2510:10.1145/2532508.2532511
2363:10.1023/A:1022850703159
2039:Computer Science Review
2005:Knowledge-Based Systems
1687:10.1145/1998076.1998121
1585:10.1145/2488388.2488433
1376:Collective intelligence
1371:Collaborative filtering
1159:Collaborative filtering
1133:Artificial intelligence
961:Recommender persistence
902:root mean squared error
675:collaborative filtering
574:content-based filtering
568:Content-based filtering
556:'s recommender system.
501:include the following:
439:collaborative filtering
425:Collaborative filtering
419:Collaborative filtering
316:knowledge-based systems
312:collaborative filtering
142:Decision support system
86:Collaborative filtering
50:Collective intelligence
3723:"Netflix Prize Update"
1721:. pp. 1692–1699.
1426:Preference elicitation
1416:Personalized marketing
1386:Enterprise bookmarking
1236:External Social Trends
1038:reproducibility crisis
842:
490:Rocchio classification
434:
111:Preference elicitation
73:Methods and challenges
5643:Mass media monitoring
5572:Seaver, Nick (2022).
5543:on September 1, 2010.
4815:. ACM. pp. 1–6.
4781:. ACM. Archived from
4295:(6). Piscataway, NJ:
4113:Information Retrieval
3911:10.1145/383952.383992
2466:10.1145/963770.963772
1829:"Recommender Systems"
1401:Information explosion
1274:search tools such as
1264:Specific applications
1189:Identifying Neighbors
834:
795:generality problems.
784:Further information:
651:information retrieval
590:information filtering
586:information retrieval
432:
229:(sometimes replacing
227:recommendation system
5297:on November 17, 2014
5050:. pp. 669–673.
4684:. pp. 240–248.
4242:. pp. 585–592.
3980:. pp. 390–398.
3779:. September 6, 2013.
3729:on November 27, 2011
2897:. Springer. p.
2749:Allen, R.B. (1990).
2444:ACM Trans. Inf. Syst
2420:10.1109/TKDE.2005.99
1224:Time and Seasonality
1183:statistical distance
1179:Statistical Distance
1098:improve this article
906:precision and recall
877:Performance measures
618:Bayesian Classifiers
339:Music Genome Project
245:), is a subclass of
147:Music Genome Project
106:Matrix factorization
5648:Recommender systems
5638:Information systems
5581:Scientific articles
5568:on August 31, 2015.
5356:. pp. 1, 14–28
4512:. pp. 304–305.
4434:2011rsh..book.....R
4319:10.1109/4236.968832
3777:"MovieLens dataset"
3765:. pp. 225–231.
3703:. December 17, 2009
3457:10.1109/mis.2011.33
2712:10.1109/MC.2009.263
1620:(11): 14077–14091.
1536:Journal of Big Data
1421:Personalized search
1411:Pattern recognition
1343:internet television
1327:deliberative groups
1309:. Examples include
1173:Data Representation
1166:K-nearest neighbors
455:Pearson Correlation
233:with terms such as
36:Recommender systems
5494:The New Face of TV
5353:Harvard University
4534:. pp. 395–399
3613:The New York Times
3540:. pp. 899–908
3052:2014-09-12 at the
2848:2015-03-16 at the
2569:. pp. 15–22.
2322:2018-12-21 at the
2260:Karlgren, Jussi. "
2247:2024-05-25 at the
2240:Karlgren, Jussi. "
2135:(519): 1344–1353.
1025:click-through rate
931:click-through rate
898:mean squared error
655:sentiment analysis
451:k-nearest neighbor
435:
273:is an implemented
223:recommender system
168:GroupLens Research
5615:Raymond J. Mooney
5561:978-0-521-49336-9
5536:978-0-07-068067-8
5351:Belfer Center at
5065:978-1-7281-8337-4
5024:978-1-4503-2668-1
4925:978-1-4503-2465-6
4884:978-1-4503-0683-6
4830:978-1-4503-6586-4
4788:on March 25, 2021
4748:978-1-4503-7583-2
4711:978-1-4503-7583-2
4651:978-1-4503-6243-6
4428:. pp. 1–35.
4328:978-1-58113-561-9
4187:978-1-4899-7637-6
4150:. pp. 22–32.
4085:on April 14, 2021
4039:978-1-4503-2465-6
3995:978-1-61197-497-3
3962:978-3-319-24591-1
3920:978-1-58113-331-8
3866:978-3-642-38843-9
3492:on June 30, 2014.
3411:978-1-4503-6201-6
3316:978-1-4503-5552-0
3271:978-1-4503-4918-5
3195:978-1-4503-6014-2
3140:978-1-4503-7998-4
2908:978-3-540-72078-2
2887:Peter Brusilovsky
2872:978-3-319-29657-9
2821:978-1-4503-4635-1
2584:978-1-4503-2465-6
2537:on April 17, 2016
2519:978-1-4503-2465-6
2504:. pp. 7–14.
2388:Adomavicius, G.;
2161:Stack, Charles. "
1989:978-1-4899-7637-6
1853:978-0-387-30164-8
1485:978-1-0716-2196-7
1288:statistical model
1130:
1129:
1122:
992:User demographics
815:The Netflix Prize
355:search algorithms
219:
218:
116:Similarity search
16:(Redirected from
5660:
5610:
5608:
5606:
5597:. Archived from
5569:
5564:. Archived from
5544:
5539:. Archived from
5496:
5491:
5485:
5484:
5482:
5480:
5475:. pp. 21–23
5464:
5458:
5457:
5455:
5453:
5438:
5432:
5431:
5429:
5427:
5412:
5406:
5405:
5399:
5391:
5389:
5387:
5372:
5366:
5365:
5363:
5361:
5342:
5333:
5332:
5330:
5328:
5313:
5307:
5306:
5304:
5302:
5293:. Archived from
5287:
5281:
5280:
5278:
5276:
5260:
5254:
5253:
5235:
5229:
5228:
5226:
5216:
5192:
5186:
5185:
5175:
5166:(5): 4425–4445.
5155:
5149:
5148:
5146:
5122:
5113:
5112:
5095:(4): 2635–2664.
5089:Educ Inf Technol
5084:
5078:
5077:
5043:
5037:
5036:
4994:
4988:
4987:
4947:
4938:
4937:
4903:
4897:
4896:
4862:
4856:
4849:
4843:
4842:
4804:
4798:
4797:
4795:
4793:
4787:
4776:
4767:
4761:
4760:
4722:
4716:
4715:
4703:
4693:
4677:
4671:
4670:
4668:
4666:
4627:
4607:
4601:
4600:
4574:
4550:
4544:
4543:
4541:
4539:
4529:
4520:
4514:
4513:
4501:
4495:
4494:
4476:
4444:
4438:
4437:
4421:
4415:
4414:
4404:
4386:
4377:
4371:
4370:
4368:
4366:
4356:
4347:
4341:
4340:
4312:
4280:
4274:
4273:
4267:
4258:
4252:
4251:
4237:
4228:
4222:
4221:
4219:
4217:
4207:
4198:
4192:
4191:
4161:
4152:
4151:
4143:
4137:
4136:
4110:
4101:
4095:
4094:
4092:
4090:
4084:
4078:. Archived from
4069:
4058:
4052:
4051:
4023:
4006:
4000:
3999:
3973:
3967:
3966:
3939:
3933:
3932:
3904:
3880:
3871:
3870:
3850:
3836:
3825:
3819:
3818:
3790:
3781:
3780:
3773:
3767:
3766:
3758:
3752:
3745:
3739:
3738:
3736:
3734:
3719:
3713:
3712:
3710:
3708:
3693:
3687:
3675:
3669:
3668:
3656:
3650:
3649:
3647:
3645:
3640:on March 4, 2012
3639:
3633:. Archived from
3632:
3623:
3617:
3616:
3604:
3595:
3594:
3576:
3556:
3550:
3549:
3547:
3545:
3531:
3520:
3511:
3510:
3500:
3494:
3493:
3491:
3485:. Archived from
3484:
3475:
3469:
3468:
3450:
3430:
3424:
3423:
3395:
3375:
3369:
3368:
3356:
3350:
3349:
3347:
3335:
3329:
3328:
3290:
3284:
3283:
3255:
3235:
3229:
3228:
3226:
3214:
3208:
3207:
3179:
3159:
3153:
3152:
3132:
3116:
3107:
3106:
3104:
3092:
3081:
3080:
3078:
3066:
3057:
3043:
3037:
3036:
3034:
3010:
3004:
2999:Rinke Hoekstra,
2997:
2991:
2990:, 21 (5): e12957
2978:
2972:
2971:
2969:
2967:
2952:
2946:
2945:
2943:
2919:
2913:
2912:
2896:
2893:The Adaptive Web
2883:
2877:
2876:
2858:
2852:
2840:
2834:
2833:
2805:
2790:
2784:
2777:
2771:
2769:
2761:
2755:
2754:
2746:
2740:
2738:
2730:
2724:
2723:
2705:
2685:
2679:
2678:
2676:
2667:
2661:
2660:
2658:
2642:
2636:
2635:
2603:
2597:
2596:
2562:
2553:
2547:
2546:
2544:
2542:
2536:
2530:. Archived from
2499:
2490:
2479:
2477:
2459:
2439:
2433:
2431:
2413:
2385:
2376:
2374:
2346:
2340:
2333:
2327:
2313:
2307:
2300:
2294:
2293:
2291:
2289:
2278:
2272:
2258:
2252:
2238:
2232:
2229:
2223:
2220:
2214:
2207:
2201:
2194:
2188:
2181:
2175:
2172:
2166:
2159:
2153:
2152:
2120:
2111:
2110:
2108:
2106:
2061:
2055:
2054:
2030:
2021:
2020:
2011:(8): 1310–1316.
2000:
1994:
1993:
1963:
1954:
1953:
1925:
1916:
1915:
1879:
1873:
1872:
1864:
1858:
1857:
1833:
1824:
1818:
1817:
1815:
1813:
1801:
1795:
1794:
1776:
1752:
1737:
1736:
1716:
1707:
1701:
1700:
1676:
1667:
1661:
1654:
1648:
1647:
1629:
1605:
1599:
1598:
1572:
1566:
1563:
1554:
1553:
1551:
1527:
1521:
1520:
1518:
1516:
1507:. Archived from
1496:
1490:
1489:
1459:
1140:machine learning
1125:
1118:
1114:
1111:
1105:
1082:
1074:
1047:RecSys Challenge
850:RecSys community
622:cluster analysis
299:academic journal
211:
204:
197:
32:
31:
21:
5668:
5667:
5663:
5662:
5661:
5659:
5658:
5657:
5628:
5627:
5613:Prem Melville,
5604:
5602:
5601:on May 11, 2009
5562:
5537:
5504:
5502:Further reading
5499:
5492:
5488:
5478:
5476:
5465:
5461:
5451:
5449:
5439:
5435:
5425:
5423:
5413:
5409:
5393:
5392:
5385:
5383:
5373:
5369:
5359:
5357:
5343:
5336:
5326:
5324:
5314:
5310:
5300:
5298:
5289:
5288:
5284:
5274:
5272:
5261:
5257:
5250:
5236:
5232:
5193:
5189:
5156:
5152:
5123:
5116:
5085:
5081:
5066:
5044:
5040:
5025:
4995:
4991:
4948:
4941:
4926:
4904:
4900:
4885:
4863:
4859:
4850:
4846:
4831:
4805:
4801:
4791:
4789:
4785:
4774:
4768:
4764:
4749:
4723:
4719:
4712:
4678:
4674:
4664:
4662:
4652:
4608:
4604:
4581:10.1145/3434185
4551:
4547:
4537:
4535:
4527:
4521:
4517:
4502:
4498:
4445:
4441:
4422:
4418:
4384:
4378:
4374:
4364:
4362:
4354:
4348:
4344:
4329:
4281:
4277:
4265:
4259:
4255:
4235:
4229:
4225:
4215:
4213:
4205:
4199:
4195:
4188:
4162:
4155:
4144:
4140:
4108:
4102:
4098:
4088:
4086:
4082:
4067:
4059:
4055:
4040:
4021:10.1.1.1031.973
4007:
4003:
3996:
3974:
3970:
3963:
3940:
3936:
3921:
3902:10.1.1.165.5800
3881:
3874:
3867:
3826:
3822:
3791:
3784:
3775:
3774:
3770:
3759:
3755:
3746:
3742:
3732:
3730:
3721:
3720:
3716:
3706:
3704:
3695:
3694:
3690:
3685:Wayback Machine
3676:
3672:
3657:
3653:
3643:
3641:
3637:
3630:
3624:
3620:
3605:
3598:
3557:
3553:
3543:
3541:
3529:
3521:
3514:
3501:
3497:
3489:
3482:
3476:
3472:
3448:10.1.1.476.6726
3431:
3427:
3412:
3376:
3372:
3357:
3353:
3336:
3332:
3317:
3291:
3287:
3272:
3236:
3232:
3215:
3211:
3196:
3160:
3156:
3141:
3117:
3110:
3093:
3084:
3067:
3060:
3054:Wayback Machine
3044:
3040:
3032:10.1145/2843948
3011:
3007:
2998:
2994:
2979:
2975:
2965:
2963:
2953:
2949:
2920:
2916:
2909:
2884:
2880:
2873:
2859:
2855:
2850:Wayback Machine
2841:
2837:
2822:
2803:
2791:
2787:
2778:
2774:
2762:
2758:
2747:
2743:
2731:
2727:
2703:10.1.1.147.8295
2686:
2682:
2674:
2668:
2664:
2643:
2639:
2604:
2600:
2585:
2560:
2554:
2550:
2540:
2538:
2534:
2520:
2497:
2491:
2482:
2440:
2436:
2411:10.1.1.107.2790
2386:
2379:
2347:
2343:
2334:
2330:
2324:Wayback Machine
2314:
2310:
2301:
2297:
2287:
2285:
2279:
2275:
2269:Wayback Machine
2259:
2255:
2249:Wayback Machine
2239:
2235:
2230:
2226:
2221:
2217:
2208:
2204:
2195:
2191:
2182:
2178:
2173:
2169:
2160:
2156:
2121:
2114:
2104:
2102:
2100:
2062:
2058:
2031:
2024:
2001:
1997:
1990:
1964:
1957:
1942:10.1145/3285954
1926:
1919:
1880:
1876:
1865:
1861:
1854:
1831:
1825:
1821:
1811:
1809:
1802:
1798:
1774:10.1038/513129a
1753:
1740:
1729:
1714:
1708:
1704:
1697:
1674:
1668:
1664:
1655:
1651:
1606:
1602:
1595:
1573:
1569:
1564:
1557:
1528:
1524:
1514:
1512:
1511:on May 30, 2010
1497:
1493:
1486:
1460:
1449:
1445:
1440:
1351:
1335:
1319:community notes
1303:
1297:
1295:Decision-making
1271:
1266:
1245:
1204:
1202:Neural networks
1156:
1126:
1115:
1109:
1106:
1095:
1083:
1072:
1034:
1032:Reproducibility
946:
944:Beyond accuracy
927:conversion rate
879:
874:
846:Gravity R&D
823:
817:
788:
782:
765:
756:
747:
734:
729:
679:knowledge-based
671:
570:
561:social networks
466:data collection
427:
421:
416:
366:
308:
277:recommendation
215:
124:Implementations
28:
23:
22:
15:
12:
11:
5:
5666:
5656:
5655:
5650:
5645:
5640:
5626:
5625:
5611:
5583:
5582:
5578:
5577:
5570:
5560:
5545:
5535:
5520:
5509:
5508:
5503:
5500:
5498:
5497:
5486:
5459:
5433:
5407:
5367:
5334:
5308:
5282:
5255:
5248:
5230:
5207:(3): 247–278.
5187:
5150:
5114:
5079:
5064:
5038:
5023:
4989:
4939:
4924:
4898:
4883:
4857:
4844:
4829:
4799:
4762:
4747:
4717:
4710:
4672:
4650:
4602:
4545:
4515:
4496:
4459:(7): 959–977.
4439:
4416:
4372:
4342:
4327:
4275:
4253:
4223:
4193:
4186:
4153:
4138:
4096:
4053:
4038:
4001:
3994:
3968:
3961:
3934:
3919:
3872:
3865:
3820:
3782:
3768:
3753:
3740:
3714:
3688:
3670:
3651:
3618:
3596:
3567:(3): 387–397.
3551:
3512:
3495:
3470:
3425:
3410:
3370:
3351:
3330:
3315:
3285:
3270:
3230:
3209:
3194:
3154:
3139:
3108:
3082:
3058:
3038:
3005:
2992:
2973:
2947:
2914:
2907:
2878:
2871:
2853:
2835:
2820:
2785:
2772:
2756:
2741:
2725:
2680:
2662:
2637:
2618:(4): 305–338.
2598:
2583:
2548:
2518:
2480:
2457:10.1.1.78.8384
2434:
2404:(6): 734–749.
2377:
2357:(4): 285–330.
2341:
2328:
2308:
2295:
2273:
2253:
2233:
2224:
2215:
2202:
2189:
2176:
2167:
2154:
2112:
2098:
2056:
2022:
1995:
1988:
1955:
1917:
1890:(1–2): 23–24.
1874:
1859:
1852:
1819:
1796:
1738:
1727:
1702:
1695:
1662:
1649:
1600:
1593:
1567:
1555:
1522:
1491:
1484:
1446:
1444:
1441:
1439:
1438:
1433:
1431:Product finder
1428:
1423:
1418:
1413:
1408:
1403:
1398:
1393:
1388:
1383:
1378:
1373:
1368:
1363:
1358:
1352:
1350:
1347:
1334:
1331:
1296:
1293:
1276:Google Scholar
1270:
1267:
1265:
1262:
1244:
1241:
1240:
1239:
1233:
1227:
1203:
1200:
1199:
1198:
1192:
1186:
1176:
1155:
1152:
1128:
1127:
1086:
1084:
1077:
1071:
1068:
1033:
1030:
1029:
1028:
1018:
1012:
1001:
995:
988:
987:
964:
957:
956:
945:
942:
878:
875:
873:
870:
819:Main article:
816:
813:
781:
778:
774:bandit problem
764:
761:
755:
752:
746:
743:
733:
730:
728:
725:
724:
723:
717:
711:
705:
699:
670:
667:
626:decision trees
608:
607:
604:
569:
566:
550:
549:
543:
537:
519:
518:
515:
512:
509:
506:
495:
494:
485:
482:
479:
476:
423:Main article:
420:
417:
415:
412:
381:Jussi Karlgren
377:Jussi Karlgren
365:
362:
343:
342:
335:
307:
304:
291:mobile devices
217:
216:
214:
213:
206:
199:
191:
188:
187:
186:
185:
180:
175:
170:
162:
161:
157:
156:
155:
154:
152:Product finder
149:
144:
139:
134:
126:
125:
121:
120:
119:
118:
113:
108:
103:
98:
93:
88:
83:
75:
74:
70:
69:
68:
67:
62:
57:
52:
44:
43:
39:
38:
26:
9:
6:
4:
3:
2:
5665:
5654:
5651:
5649:
5646:
5644:
5641:
5639:
5636:
5635:
5633:
5623:
5620:
5616:
5612:
5600:
5596:
5595:
5594:IEEE Spectrum
5590:
5585:
5584:
5580:
5579:
5575:
5571:
5567:
5563:
5557:
5553:
5552:
5546:
5542:
5538:
5532:
5528:
5527:
5521:
5519:
5518:9781617292705
5515:
5511:
5510:
5506:
5505:
5495:
5490:
5474:
5470:
5463:
5448:
5444:
5437:
5422:
5418:
5411:
5403:
5397:
5382:
5378:
5371:
5355:
5354:
5348:
5341:
5339:
5323:
5319:
5312:
5296:
5292:
5286:
5270:
5266:
5259:
5251:
5249:9780262042840
5245:
5242:. MIT press.
5241:
5234:
5225:
5220:
5215:
5210:
5206:
5202:
5198:
5191:
5183:
5179:
5174:
5169:
5165:
5161:
5154:
5145:
5140:
5136:
5132:
5128:
5121:
5119:
5110:
5106:
5102:
5098:
5094:
5090:
5083:
5075:
5071:
5067:
5061:
5057:
5053:
5049:
5042:
5034:
5030:
5026:
5020:
5016:
5012:
5008:
5004:
5000:
4993:
4985:
4981:
4977:
4973:
4969:
4965:
4962:(1): 69–101.
4961:
4957:
4953:
4946:
4944:
4935:
4931:
4927:
4921:
4917:
4913:
4909:
4902:
4894:
4890:
4886:
4880:
4876:
4872:
4868:
4861:
4854:
4848:
4840:
4836:
4832:
4826:
4822:
4818:
4814:
4810:
4803:
4784:
4780:
4773:
4766:
4758:
4754:
4750:
4744:
4740:
4736:
4732:
4728:
4721:
4713:
4707:
4702:
4697:
4692:
4687:
4683:
4676:
4661:
4657:
4653:
4647:
4643:
4642:11311/1108996
4639:
4635:
4631:
4626:
4621:
4617:
4613:
4606:
4598:
4594:
4590:
4589:11311/1164333
4586:
4582:
4578:
4573:
4568:
4564:
4560:
4556:
4549:
4533:
4526:
4519:
4511:
4507:
4500:
4492:
4488:
4484:
4480:
4475:
4470:
4466:
4462:
4458:
4454:
4450:
4443:
4435:
4431:
4427:
4420:
4412:
4408:
4403:
4398:
4395:(1–2): 1–23.
4394:
4390:
4383:
4376:
4360:
4353:
4346:
4338:
4334:
4330:
4324:
4320:
4316:
4311:
4310:10.1.1.2.2932
4306:
4302:
4298:
4294:
4290:
4286:
4279:
4271:
4264:
4257:
4249:
4245:
4241:
4234:
4227:
4211:
4204:
4197:
4189:
4183:
4179:
4175:
4171:
4167:
4160:
4158:
4149:
4142:
4134:
4130:
4126:
4122:
4118:
4114:
4107:
4100:
4081:
4077:
4073:
4066:
4065:
4057:
4049:
4045:
4041:
4035:
4031:
4027:
4022:
4017:
4013:
4005:
3997:
3991:
3987:
3983:
3979:
3972:
3964:
3958:
3954:
3950:
3946:
3938:
3930:
3926:
3922:
3916:
3912:
3908:
3903:
3898:
3894:
3890:
3886:
3879:
3877:
3868:
3862:
3858:
3854:
3849:
3848:10.1.1.465.96
3844:
3840:
3835:
3834:
3824:
3816:
3812:
3808:
3804:
3800:
3796:
3789:
3787:
3778:
3772:
3764:
3757:
3750:
3744:
3728:
3724:
3718:
3702:
3698:
3692:
3686:
3682:
3679:
3674:
3666:
3662:
3655:
3636:
3629:
3622:
3614:
3610:
3603:
3601:
3592:
3588:
3584:
3580:
3575:
3570:
3566:
3562:
3555:
3539:
3535:
3528:
3527:
3519:
3517:
3508:
3507:
3499:
3488:
3481:
3474:
3466:
3462:
3458:
3454:
3449:
3444:
3440:
3436:
3429:
3421:
3417:
3413:
3407:
3403:
3399:
3394:
3389:
3385:
3381:
3374:
3366:
3362:
3355:
3346:
3341:
3334:
3326:
3322:
3318:
3312:
3308:
3304:
3300:
3296:
3289:
3281:
3277:
3273:
3267:
3263:
3259:
3254:
3249:
3245:
3241:
3234:
3225:
3220:
3213:
3205:
3201:
3197:
3191:
3187:
3183:
3178:
3173:
3169:
3165:
3158:
3150:
3146:
3142:
3136:
3131:
3126:
3122:
3115:
3113:
3103:
3098:
3091:
3089:
3087:
3077:
3072:
3065:
3063:
3055:
3051:
3048:
3045:Robin Burke,
3042:
3033:
3028:
3024:
3020:
3016:
3009:
3002:
2996:
2989:
2988:
2983:
2977:
2962:
2958:
2951:
2942:
2937:
2933:
2929:
2925:
2918:
2910:
2904:
2900:
2895:
2894:
2888:
2882:
2874:
2868:
2864:
2857:
2851:
2847:
2844:
2839:
2831:
2827:
2823:
2817:
2813:
2809:
2802:
2801:
2796:
2789:
2782:
2776:
2767:
2760:
2752:
2745:
2736:
2729:
2721:
2717:
2713:
2709:
2704:
2699:
2695:
2691:
2684:
2673:
2666:
2657:
2652:
2648:
2641:
2633:
2629:
2625:
2621:
2617:
2613:
2609:
2602:
2594:
2590:
2586:
2580:
2576:
2572:
2568:
2567:
2559:
2552:
2533:
2529:
2525:
2521:
2515:
2511:
2507:
2503:
2496:
2489:
2487:
2485:
2475:
2471:
2467:
2463:
2458:
2453:
2449:
2445:
2438:
2429:
2425:
2421:
2417:
2412:
2407:
2403:
2399:
2395:
2392:(June 2005).
2391:
2384:
2382:
2372:
2368:
2364:
2360:
2356:
2352:
2345:
2338:
2332:
2325:
2321:
2318:
2312:
2305:
2299:
2284:
2277:
2270:
2266:
2263:
2257:
2250:
2246:
2243:
2237:
2228:
2219:
2212:
2206:
2199:
2193:
2186:
2180:
2171:
2164:
2158:
2150:
2146:
2142:
2138:
2134:
2130:
2126:
2119:
2117:
2101:
2099:1-58113-561-0
2095:
2091:
2087:
2083:
2080:
2077:
2073:
2072:
2067:
2066:Lyle H. Ungar
2060:
2052:
2048:
2044:
2040:
2036:
2029:
2027:
2018:
2014:
2010:
2006:
1999:
1991:
1985:
1981:
1977:
1973:
1969:
1962:
1960:
1951:
1947:
1943:
1939:
1935:
1931:
1924:
1922:
1913:
1909:
1905:
1901:
1897:
1893:
1889:
1885:
1878:
1870:
1863:
1855:
1849:
1845:
1841:
1837:
1830:
1823:
1807:
1800:
1792:
1788:
1784:
1780:
1775:
1770:
1766:
1762:
1758:
1751:
1749:
1747:
1745:
1743:
1734:
1730:
1728:9781577353232
1724:
1720:
1713:
1706:
1698:
1696:9781450307444
1692:
1688:
1684:
1680:
1673:
1666:
1659:
1653:
1645:
1641:
1637:
1633:
1628:
1623:
1619:
1615:
1611:
1604:
1596:
1594:9781450320351
1590:
1586:
1582:
1578:
1571:
1562:
1560:
1550:
1545:
1541:
1537:
1533:
1526:
1510:
1506:
1502:
1495:
1487:
1481:
1477:
1473:
1469:
1465:
1458:
1456:
1454:
1452:
1447:
1437:
1434:
1432:
1429:
1427:
1424:
1422:
1419:
1417:
1414:
1412:
1409:
1407:
1404:
1402:
1399:
1397:
1394:
1392:
1391:Filter bubble
1389:
1387:
1384:
1382:
1379:
1377:
1374:
1372:
1369:
1367:
1364:
1362:
1359:
1357:
1354:
1353:
1346:
1344:
1340:
1330:
1328:
1324:
1320:
1316:
1312:
1308:
1302:
1292:
1289:
1284:
1281:
1277:
1261:
1259:
1255:
1251:
1237:
1234:
1231:
1228:
1225:
1222:
1221:
1220:
1216:
1213:
1209:
1196:
1193:
1190:
1187:
1184:
1180:
1177:
1174:
1171:
1170:
1169:
1167:
1162:
1160:
1151:
1149:
1145:
1144:deep learning
1141:
1136:
1134:
1124:
1121:
1113:
1103:
1099:
1093:
1092:
1087:This section
1085:
1081:
1076:
1075:
1067:
1064:
1060:
1056:
1052:
1048:
1044:
1039:
1026:
1022:
1019:
1016:
1013:
1009:
1005:
1002:
999:
996:
993:
990:
989:
984:
983:Netflix Prize
980:
976:
972:
971:user profiles
968:
965:
962:
959:
958:
954:
951:
950:
949:
941:
937:
934:
932:
928:
922:
918:
914:
911:
907:
903:
899:
894:
892:
888:
884:
883:effectiveness
869:
867:
863:
859:
853:
851:
847:
841:
839:
833:
830:
828:
827:Netflix Prize
822:
821:Netflix Prize
812:
809:
805:
800:
796:
793:
787:
777:
775:
771:
760:
751:
742:
740:
721:
718:
715:
712:
709:
706:
703:
700:
697:
694:
693:
692:
689:
686:
682:
680:
676:
666:
664:
663:deep learning
660:
656:
652:
648:
643:
637:
633:
631:
627:
623:
619:
614:
605:
602:
601:
600:
598:
593:
591:
587:
582:
577:
575:
565:
562:
557:
555:
547:
544:
541:
538:
535:
531:
528:
527:
526:
524:
516:
513:
510:
507:
504:
503:
502:
500:
492:
491:
486:
483:
480:
477:
474:
473:
472:
469:
467:
463:
458:
456:
452:
446:
444:
440:
431:
426:
411:
409:
405:
400:
398:
394:
390:
386:
382:
378:
373:
370:
361:
358:
356:
351:
348:
340:
336:
332:
331:
330:
328:
327:Pandora Radio
324:
319:
317:
313:
303:
300:
296:
295:set-top boxes
292:
288:
284:
280:
276:
272:
267:
265:
264:online dating
260:
256:
251:
248:
244:
240:
236:
232:
228:
224:
212:
207:
205:
200:
198:
193:
192:
190:
189:
184:
181:
179:
178:Netflix Prize
176:
174:
171:
169:
166:
165:
164:
163:
159:
158:
153:
150:
148:
145:
143:
140:
138:
135:
133:
130:
129:
128:
127:
123:
122:
117:
114:
112:
109:
107:
104:
102:
99:
97:
94:
92:
89:
87:
84:
82:
79:
78:
77:
76:
72:
71:
66:
63:
61:
58:
56:
53:
51:
48:
47:
46:
45:
41:
40:
37:
34:
33:
30:
19:
5621:
5605:December 10,
5603:. Retrieved
5599:the original
5592:
5573:
5566:the original
5550:
5541:the original
5525:
5489:
5477:. Retrieved
5472:
5462:
5450:. Retrieved
5446:
5436:
5424:. Retrieved
5420:
5410:
5384:. Retrieved
5380:
5370:
5358:. Retrieved
5350:
5325:. Retrieved
5321:
5311:
5299:. Retrieved
5295:the original
5285:
5275:December 31,
5273:. Retrieved
5268:
5258:
5239:
5233:
5204:
5200:
5190:
5163:
5159:
5153:
5134:
5130:
5092:
5088:
5082:
5047:
5041:
5015:10486/665450
4998:
4992:
4959:
4955:
4907:
4901:
4866:
4860:
4847:
4812:
4802:
4790:. Retrieved
4783:the original
4778:
4765:
4730:
4720:
4681:
4675:
4663:. Retrieved
4615:
4605:
4562:
4558:
4548:
4536:. Retrieved
4531:
4518:
4509:
4499:
4456:
4452:
4442:
4425:
4419:
4392:
4388:
4375:
4363:. Retrieved
4358:
4345:
4292:
4288:
4278:
4269:
4256:
4239:
4226:
4214:. Retrieved
4209:
4196:
4169:
4147:
4141:
4116:
4112:
4099:
4087:. Retrieved
4080:the original
4063:
4056:
4011:
4004:
3977:
3971:
3944:
3937:
3888:
3832:
3823:
3798:
3794:
3771:
3762:
3756:
3743:
3733:December 14,
3731:. Retrieved
3727:the original
3717:
3705:. Retrieved
3700:
3691:
3673:
3664:
3654:
3642:. Retrieved
3635:the original
3621:
3612:
3564:
3560:
3554:
3544:November 17,
3542:. Retrieved
3525:
3505:
3498:
3487:the original
3473:
3441:(2): 64–76.
3438:
3434:
3428:
3383:
3373:
3367:: 2592–2599.
3364:
3354:
3333:
3298:
3288:
3243:
3233:
3212:
3167:
3157:
3120:
3041:
3022:
3018:
3008:
2995:
2985:
2976:
2964:. Retrieved
2960:
2950:
2931:
2927:
2917:
2892:
2881:
2865:. Springer.
2862:
2856:
2838:
2799:
2788:
2775:
2765:
2759:
2750:
2744:
2728:
2696:(8): 30–37.
2693:
2689:
2683:
2665:
2646:
2640:
2615:
2611:
2601:
2565:
2551:
2539:. Retrieved
2532:the original
2501:
2447:
2443:
2437:
2401:
2397:
2390:Tuzhilin, A.
2354:
2350:
2344:
2331:
2311:
2298:
2286:. Retrieved
2276:
2256:
2236:
2227:
2218:
2205:
2192:
2179:
2170:
2157:
2132:
2128:
2103:. Retrieved
2070:
2059:
2042:
2038:
2008:
2004:
1998:
1971:
1933:
1929:
1887:
1883:
1877:
1868:
1862:
1835:
1822:
1812:December 31,
1810:. Retrieved
1799:
1764:
1760:
1718:
1705:
1678:
1665:
1652:
1617:
1613:
1603:
1576:
1570:
1539:
1535:
1525:
1513:. Retrieved
1509:the original
1504:
1494:
1467:
1381:Configurator
1336:
1304:
1285:
1272:
1246:
1235:
1229:
1223:
1217:
1205:
1194:
1188:
1178:
1172:
1163:
1157:
1137:
1131:
1116:
1110:October 2023
1107:
1096:Please help
1091:verification
1088:
1035:
1020:
1014:
1003:
997:
991:
975:data privacy
966:
960:
952:
947:
938:
935:
923:
919:
915:
895:
880:
854:
843:
837:
835:
831:
824:
801:
797:
789:
769:
766:
757:
748:
735:
727:Technologies
719:
713:
707:
701:
695:
690:
688:filtering).
683:
681:approaches.
672:
638:
634:
609:
597:user profile
595:To create a
594:
581:user profile
578:
573:
571:
558:
551:
545:
539:
529:
520:
497:Examples of
496:
488:
470:
459:
447:
436:
401:
393:Paul Resnick
374:
367:
359:
352:
344:
320:
309:
270:
268:
252:
242:
238:
234:
230:
226:
222:
220:
60:Star ratings
35:
29:
5301:December 8,
5137:: 439–457.
4665:October 16,
4565:(2): 1–49.
4538:December 2,
4365:November 1,
4216:November 1,
3025:(4): 1–19.
2966:October 31,
2541:October 22,
2450:(1): 5–53.
2288:October 27,
2105:February 2,
2088:. pp.
1436:Rating site
1008:Serendipity
1004:Serendipity
917:imprecise.
887:evaluations
792:smartphones
647:text mining
540:Scalability
404:Adomavicius
389:Pattie Maes
369:Elaine Rich
5632:Categories
5322:Nieman Lab
5214:2003.07631
5173:2104.13030
4691:2005.09683
4625:1907.06902
4572:1911.07698
3574:1805.02276
3393:1902.05570
3345:2006.05779
3253:1711.04725
3224:1808.09781
3177:1706.03847
3102:1812.02353
3076:1511.06939
1443:References
1366:Cold start
1333:Television
1307:polarizing
1299:See also:
998:Robustness
872:Evaluation
720:Meta-level
657:(see also
592:research.
554:Amazon.com
530:Cold start
523:cold start
414:Approaches
383:, then at
347:cold start
81:Cold start
5109:254475908
5074:232150789
4976:0924-1868
4757:221785064
4660:196831663
4597:208138060
4491:149344712
4483:1369-118X
4305:CiteSeerX
4133:213169978
4016:CiteSeerX
3897:CiteSeerX
3843:CiteSeerX
3801:: 37–45.
3644:April 30,
3443:CiteSeerX
3149:221191348
2698:CiteSeerX
2656:1301.7363
2632:207035184
2474:207731647
2452:CiteSeerX
2428:206742345
2406:CiteSeerX
2149:125187672
2045:: 29–50.
1912:161141937
1904:1058-8167
1636:1573-7721
1339:broadband
1212:black-box
1021:Labelling
979:profiling
953:Diversity
702:Switching
464:forms of
243:algorithm
173:MovieLens
65:Long tail
55:Relevance
5479:July 17,
5452:July 17,
5426:July 17,
5421:Engadget
5396:cite web
5386:July 17,
5360:July 17,
5327:July 17,
5033:15665277
4839:52942462
4792:April 3,
4089:March 5,
3929:18903114
3815:10651930
3681:Archived
3591:19209845
3465:16752808
3420:62903207
3325:50775765
3280:21066930
3050:Archived
2889:(2007).
2846:Archived
2720:58370896
2690:Computer
2371:16544257
2320:Archived
2265:Archived
2245:Archived
1950:59337456
1936:: 1–22.
1783:25186906
1733:ACM Copy
1644:36511631
1349:See also
1321:, which
1063:Bellogín
1051:Ekstrand
696:Weighted
642:metadata
546:Sparsity
462:implicit
306:Overview
287:websites
283:metadata
279:platform
275:software
259:playlist
235:platform
160:Research
42:Concepts
5554:. CUP.
5529:. CUP.
5447:Gizmodo
5381:Poynter
4893:2215419
4430:Bibcode
4411:8996665
4337:1977107
4272:: 1–39.
4248:8307833
4048:8202591
3893:225–231
3707:June 1,
3295:"STAMP"
3204:1159769
2934:: 1–9.
2593:4411601
2528:8202591
2090:253–260
1808:. WIRED
1791:4460749
1515:June 1,
1323:YouTube
1315:Twitter
1256:(SVD),
1252:(LSA),
1055:Konstan
967:Privacy
714:Cascade
685:Netflix
364:History
323:Last.fm
225:, or a
5558:
5533:
5516:
5246:
5107:
5072:
5062:
5031:
5021:
4984:388764
4982:
4974:
4934:333956
4932:
4922:
4891:
4881:
4837:
4827:
4755:
4745:
4708:
4658:
4648:
4595:
4489:
4481:
4409:
4335:
4325:
4307:
4246:
4184:
4131:
4046:
4036:
4018:
3992:
3959:
3927:
3917:
3899:
3863:
3845:
3813:
3589:
3463:
3445:
3418:
3408:
3323:
3313:
3278:
3268:
3202:
3192:
3147:
3137:
2905:
2869:
2830:653908
2828:
2818:
2718:
2700:
2630:
2591:
2581:
2526:
2516:
2472:
2454:
2426:
2408:
2369:
2147:
2096:
1986:
1948:
1910:
1902:
1850:
1789:
1781:
1761:Nature
1725:
1693:
1642:
1634:
1591:
1542:(59).
1482:
1280:PubMed
1146:, and
1043:RecSys
661:) and
628:, and
613:tf–idf
239:engine
231:system
5507:Books
5209:arXiv
5168:arXiv
5105:S2CID
5070:S2CID
5029:S2CID
4980:S2CID
4930:S2CID
4889:S2CID
4835:S2CID
4786:(PDF)
4775:(PDF)
4753:S2CID
4686:arXiv
4656:S2CID
4620:arXiv
4593:S2CID
4567:arXiv
4528:(PDF)
4487:S2CID
4407:S2CID
4385:(PDF)
4355:(PDF)
4333:S2CID
4301:54–62
4266:(PDF)
4244:S2CID
4236:(PDF)
4206:(PDF)
4129:S2CID
4109:(PDF)
4083:(PDF)
4068:(PDF)
4044:S2CID
3925:S2CID
3841:–37.
3811:S2CID
3701:WIRED
3665:Index
3638:(PDF)
3631:(PDF)
3587:S2CID
3569:arXiv
3530:(PDF)
3490:(PDF)
3483:(PDF)
3461:S2CID
3416:S2CID
3388:arXiv
3340:arXiv
3321:S2CID
3276:S2CID
3248:arXiv
3219:arXiv
3200:S2CID
3172:arXiv
3145:S2CID
3097:arXiv
3071:arXiv
2826:S2CID
2804:(PDF)
2716:S2CID
2675:(PDF)
2651:arXiv
2628:S2CID
2589:S2CID
2561:(PDF)
2535:(PDF)
2524:S2CID
2498:(PDF)
2470:S2CID
2424:S2CID
2367:S2CID
2145:S2CID
2079:SIGIR
1946:S2CID
1908:S2CID
1832:(PDF)
1787:S2CID
1715:(PDF)
1675:(PDF)
1640:S2CID
1311:Polis
1015:Trust
770:DRARS
708:Mixed
559:Many
241:, or
5607:2018
5556:ISBN
5531:ISBN
5514:ISBN
5481:2024
5454:2024
5428:2024
5402:link
5388:2024
5362:2024
5329:2024
5303:2014
5277:2015
5244:ISBN
5060:ISBN
5019:ISBN
4972:ISSN
4920:ISBN
4879:ISBN
4825:ISBN
4794:2021
4743:ISBN
4706:ISBN
4667:2019
4646:ISBN
4540:2013
4479:ISSN
4367:2013
4323:ISBN
4218:2013
4182:ISBN
4091:2021
4034:ISBN
3990:ISBN
3957:ISBN
3915:ISBN
3861:ISBN
3735:2011
3709:2015
3646:2009
3546:2011
3406:ISBN
3311:ISBN
3266:ISBN
3190:ISBN
3135:ISBN
2968:2016
2903:ISBN
2867:ISBN
2816:ISBN
2579:ISBN
2543:2013
2514:ISBN
2290:2017
2107:2008
2094:ISBN
1984:ISBN
1900:ISSN
1848:ISBN
1814:2015
1779:PMID
1723:ISBN
1691:ISBN
1632:ISSN
1589:ISBN
1517:2015
1505:TIME
1480:ISBN
1061:and
1059:Said
900:and
808:Lyft
806:and
804:Uber
588:and
408:Beel
385:SICS
325:and
293:and
5219:doi
5205:109
5178:doi
5139:doi
5097:doi
5052:doi
5011:hdl
5003:doi
4964:doi
4912:doi
4871:doi
4817:doi
4735:doi
4696:doi
4638:hdl
4630:doi
4585:hdl
4577:doi
4469:hdl
4461:doi
4397:doi
4315:doi
4174:doi
4121:doi
4072:doi
4026:doi
3982:doi
3949:doi
3907:doi
3853:doi
3803:doi
3579:doi
3538:ACM
3453:doi
3398:doi
3303:doi
3258:doi
3182:doi
3125:doi
3027:doi
2984:",
2936:doi
2932:157
2899:325
2808:doi
2708:doi
2620:doi
2571:doi
2506:doi
2462:doi
2416:doi
2359:doi
2137:doi
2133:112
2086:ACM
2076:ACM
2047:doi
2013:doi
1976:doi
1938:doi
1892:doi
1840:doi
1769:doi
1765:513
1683:doi
1622:doi
1581:doi
1544:doi
1472:doi
1278:or
1206:An
1100:by
929:or
910:DCG
908:or
5634::
5591:.
5471:.
5445:.
5419:.
5398:}}
5394:{{
5379:.
5349:.
5337:^
5320:.
5267:.
5217:.
5203:.
5199:.
5176:.
5164:35
5162:.
5133:.
5129:.
5117:^
5103:.
5093:25
5091:.
5068:.
5058:.
5027:.
5017:.
5009:.
4978:.
4970:.
4960:26
4958:.
4954:.
4942:^
4928:.
4918:.
4887:.
4877:.
4833:.
4823:.
4811:.
4777:.
4751:.
4741:.
4729:.
4704:.
4694:.
4654:.
4644:.
4636:.
4628:.
4614:.
4591:.
4583:.
4575:.
4563:39
4561:.
4557:.
4508:.
4485:.
4477:.
4467:.
4457:21
4455:.
4451:.
4405:.
4393:22
4391:.
4387:.
4331:.
4321:.
4313:.
4303:.
4299::
4291:.
4287:.
4268:.
4238:.
4180:.
4156:^
4127:.
4117:23
4115:.
4111:.
4042:.
4032:.
4024:.
3988:.
3955:.
3923:.
3913:.
3905:.
3895:.
3887:.
3875:^
3859:.
3851:.
3839:25
3809:.
3799:19
3797:.
3785:^
3699:.
3663:.
3611:.
3599:^
3585:.
3577:.
3565:45
3563:.
3536::
3515:^
3459:.
3451:.
3439:26
3437:.
3414:.
3404:.
3396:.
3382:.
3363:.
3319:.
3309:.
3297:.
3274:.
3264:.
3256:.
3242:.
3198:.
3188:.
3180:.
3166:.
3143:.
3133:.
3111:^
3085:^
3061:^
3021:.
3017:.
2959:.
2930:.
2926:.
2901:.
2824:.
2814:.
2797:.
2714:.
2706:.
2694:42
2692:.
2626:.
2616:17
2614:.
2610:.
2587:.
2577:.
2563:.
2522:.
2512:.
2500:.
2483:^
2468:.
2460:.
2448:22
2446:.
2422:.
2414:.
2402:17
2400:.
2396:.
2380:^
2365:.
2355:19
2353:.
2143:.
2131:.
2127:.
2115:^
2092:.
2084::
2043:20
2041:.
2037:.
2025:^
2009:24
2007:.
1982:.
1958:^
1944:.
1934:13
1932:.
1920:^
1906:.
1898:.
1888:12
1886:.
1846:.
1785:.
1777:.
1763:.
1759:.
1741:^
1731:.
1689:.
1677:.
1638:.
1630:.
1618:77
1616:.
1612:.
1587:.
1558:^
1538:.
1534:.
1503:.
1478:.
1450:^
1142:,
1053:,
1006:–
933:.
665:.
653:,
649:,
624:,
620:,
468:.
445:.
399:.
329:.
289:,
269:A
237:,
221:A
5609:.
5483:.
5456:.
5430:.
5404:)
5390:.
5364:.
5331:.
5305:.
5279:.
5252:.
5227:.
5221::
5211::
5184:.
5180::
5170::
5147:.
5141::
5135:7
5111:.
5099::
5076:.
5054::
5035:.
5013::
5005::
4986:.
4966::
4936:.
4914::
4895:.
4873::
4841:.
4819::
4796:.
4759:.
4737::
4714:.
4698::
4688::
4669:.
4640::
4632::
4622::
4599:.
4587::
4579::
4569::
4542:.
4493:.
4471::
4463::
4436:.
4432::
4413:.
4399::
4369:.
4339:.
4317::
4293:5
4250:.
4220:.
4190:.
4176::
4135:.
4123::
4093:.
4074::
4050:.
4028::
3998:.
3984::
3965:.
3951::
3931:.
3909::
3869:.
3855::
3817:.
3805::
3737:.
3711:.
3667:.
3648:.
3615:.
3593:.
3581::
3571::
3548:.
3467:.
3455::
3422:.
3400::
3390::
3348:.
3342::
3327:.
3305::
3282:.
3260::
3250::
3227:.
3221::
3206:.
3184::
3174::
3151:.
3127::
3105:.
3099::
3079:.
3073::
3035:.
3029::
3023:6
2970:.
2944:.
2938::
2911:.
2875:.
2832:.
2810::
2770:.
2739:,
2737:.
2722:.
2710::
2659:.
2653::
2634:.
2622::
2595:.
2573::
2545:.
2508::
2478:.
2476:.
2464::
2432:.
2430:.
2418::
2375:.
2373:.
2361::
2292:.
2151:.
2139::
2109:.
2053:.
2049::
2019:.
2015::
1992:.
1978::
1952:.
1940::
1914:.
1894::
1856:.
1842::
1816:.
1793:.
1771::
1735:.
1699:.
1685::
1646:.
1624::
1597:.
1583::
1552:.
1546::
1540:9
1519:.
1488:.
1474::
1123:)
1117:(
1112:)
1108:(
1094:.
536:.
210:e
203:t
196:v
20:)
Text is available under the Creative Commons Attribution-ShareAlike License. Additional terms may apply.