165:
have shown that "easily accessible digital records of behavior, Facebook Likes, can be used to automatically and accurately predict a range of highly sensitive personal attributes including: sexual orientation, ethnicity, religious and political views, personality traits, intelligence, happiness, use
91:
of 0.340. In order to build the model, Facebook data about individuals was collected in various
European countries with prior permission from the individuals. This data was then combined with the actual loan payment information for the same people and the scoring models were built using the same
406:
283:
301:
636:
437:
599:
536:
370:
631:
197:
exhibition start-up ALPHA program. In March 2013, Big Data
Scoring was selected as one finalists of the Code_n competition, which is part of the
100:
Big Data
Scoring collects vast amounts of data from publicly available online sources and uses it to predict individuals’ behavior by applying
486:
511:
585:
413:
287:
144:
This results in more people receiving access to credit with a better interest rate thanks to increase of scoring model accuracy.
309:
244:
used for credit scoring is done legally. According to the company, their solution requires a permission from the users of
108:. Based on client feedback, their solution delivers an improvement of up to 25% in scoring accuracy when combined with
302:"Case study about a Central European lender : Big Data Scoring | The Leader in Big Data Credit Scoring Solutions"
120:
data is lacking, the added benefit can be even greater to people with little or even no credit history, for example:
83:
On April 9, 2013, the company announced that they have built a credit scoring model based purely on information from
252:
have cited invasion of privacy as an additional concern regarding using social media information in credit scoring.
603:
444:
381:
209:
presented their solutions live on stage. The company has been featured in many on-line magazines, including
646:
174:
355:
331:
461:
248:
to access their data and nothing is collected without the prior permission. Other sources such as
641:
162:
26:
service that lets consumer lenders improve loan quality and acceptance rates through the use of
342:
101:
59:
8:
332:"Private traits and attributes are predictable from digital records of human behavior"
182:
88:
158:
23:
109:
31:
625:
157:
The company is not the first to show the predictive powers of
Facebook data.
117:
512:"Should your Facebook profile influence your credit score? Startups say yes"
237:
178:
63:
241:
210:
133:
113:
55:
225:
194:
138:
330:
Kosinski, Michal; David
Stillwell; Thore Graepel (February 12, 2013).
193:
In
October 2013, Big Data Scoring was selected as one finalist of the
561:
371:"Big Data and Credit Unions: Machine Learning in Member Transactions"
105:
43:
329:
537:"CeBIT Code_n Exhibit Shows Why Useful Innovation Is the Best Kind"
249:
245:
214:
129:
84:
27:
147:
202:
112:. This also robustly translates to an equivalent increase in the
67:
35:
269:
177:
published a paper showing clear patterns in transactional data,
166:
of addictive substances, parental separation, age, and gender.
47:
218:
198:
124:
39:
205:, Germany. During Finovate Fall 2015 conference the CEO of
51:
284:"First Ever Generic European Social Media Scorecard Ready"
92:
tools used in building traditional credit scoring models.
73:
188:
95:
30:. The company was founded in 2013 and has offices in
600:"Rumor: Facebook 'likes' can hurt your credit score"
586:"We Are Not Data Mining From Social Media Illegally"
87:. According to the company, the scoring model has a
50:. The company's services are aimed at all lenders –
462:"FinovateFall 2015 - Big Data Scoring - Finovate"
623:
286:. Company web page. 9 April 2013. Archived from
78:
148:Predictive powers of big data in credit scoring
487:"When Facebook is bad for one's credit rating"
181:and external factors like the recent price of
637:Information technology companies of Estonia
368:
224:Big Data Scoring is working together with
161:, David Stillwell, and Thore Graepel from
632:Technology companies established in 2013
152:
624:
588:. Baltic Business News. May 8, 2013.
484:
74:Big data based credit scoring models
189:Press coverage and acknowledgements
96:Based on publicly available sources
13:
14:
658:
169:
592:
578:
554:
529:
504:
407:"WebSummit ALPHA Finalist List"
116:. In markets where traditional
478:
454:
430:
399:
362:
323:
294:
276:
262:
60:peer-to-peer lending platforms
1:
255:
228:in their Start Path program.
79:Based on Facebook information
369:Kallerhoff, Philipp (2013).
240:raised the question whether
231:
110:traditional in-house methods
104:data processing and scoring
7:
10:
663:
602:. MSN News. Archived from
438:"List of CODE_n finalists"
378:Filene Research Institute
175:Filene Research Institute
562:"Portfolio | Start Path"
236:Estonian business daily
163:University of Cambridge
350:Cite journal requires
306:www.bigdatascoring.com
64:microfinance providers
485:Pimentel, Benjamin.
153:Facebook information
272:. Company web page.
16:Cloud-based service
647:Big data companies
606:on August 29, 2013
387:on 8 December 2015
270:"Big Data Scoring"
566:www.startpath.com
139:recent immigrants
68:leasing companies
654:
616:
615:
613:
611:
596:
590:
589:
582:
576:
575:
573:
572:
558:
552:
551:
549:
547:
533:
527:
526:
524:
522:
508:
502:
501:
499:
497:
482:
476:
475:
473:
472:
458:
452:
451:
449:
443:. Archived from
442:
434:
428:
427:
425:
424:
418:
412:. Archived from
411:
403:
397:
396:
394:
392:
386:
380:. Archived from
375:
366:
360:
359:
353:
348:
346:
338:
336:
327:
321:
320:
318:
317:
308:. Archived from
298:
292:
291:
280:
274:
273:
266:
207:Big Data Scoring
89:Gini coefficient
20:Big Data Scoring
662:
661:
657:
656:
655:
653:
652:
651:
622:
621:
620:
619:
609:
607:
598:
597:
593:
584:
583:
579:
570:
568:
560:
559:
555:
545:
543:
535:
534:
530:
520:
518:
510:
509:
505:
495:
493:
483:
479:
470:
468:
460:
459:
455:
447:
440:
436:
435:
431:
422:
420:
416:
409:
405:
404:
400:
390:
388:
384:
373:
367:
363:
351:
349:
340:
339:
334:
328:
324:
315:
313:
300:
299:
295:
282:
281:
277:
268:
267:
263:
258:
234:
191:
172:
159:Michal Kosinski
155:
150:
98:
81:
76:
17:
12:
11:
5:
660:
650:
649:
644:
642:Credit scoring
639:
634:
618:
617:
591:
577:
553:
528:
503:
477:
453:
450:on 2014-05-27.
429:
398:
361:
352:|journal=
322:
293:
290:on 2014-05-29.
275:
260:
259:
257:
254:
233:
230:
201:exhibition in
190:
187:
171:
170:Public sources
168:
154:
151:
149:
146:
142:
141:
136:
127:
97:
94:
80:
77:
75:
72:
56:payday lenders
15:
9:
6:
4:
3:
2:
659:
648:
645:
643:
640:
638:
635:
633:
630:
629:
627:
605:
601:
595:
587:
581:
567:
563:
557:
542:
538:
532:
517:
513:
507:
492:
488:
481:
467:
463:
457:
446:
439:
433:
419:on 2013-11-02
415:
408:
402:
383:
379:
372:
365:
357:
344:
333:
326:
312:on 2015-10-22
311:
307:
303:
297:
289:
285:
279:
271:
265:
261:
253:
251:
247:
243:
239:
229:
227:
222:
220:
216:
212:
208:
204:
200:
196:
186:
184:
180:
176:
167:
164:
160:
145:
140:
137:
135:
131:
128:
126:
123:
122:
121:
119:
118:credit bureau
115:
111:
107:
103:
93:
90:
86:
71:
69:
65:
61:
57:
53:
49:
45:
41:
37:
33:
29:
25:
21:
608:. Retrieved
604:the original
594:
580:
569:. Retrieved
565:
556:
544:. Retrieved
540:
531:
519:. Retrieved
515:
506:
494:. Retrieved
490:
480:
469:. Retrieved
465:
456:
445:the original
432:
421:. Retrieved
414:the original
401:
389:. Retrieved
382:the original
377:
364:
343:cite journal
325:
314:. Retrieved
310:the original
305:
296:
288:the original
278:
264:
235:
223:
206:
192:
179:credit score
173:
156:
143:
125:young people
99:
82:
19:
18:
491:MarketWatch
391:25 November
242:data mining
211:MarketWatch
183:S&P 500
134:underbanked
114:bottom line
102:proprietary
24:cloud-based
626:Categories
610:August 27,
571:2015-11-27
471:2015-11-27
423:2014-04-15
316:2015-11-27
256:References
226:MasterCard
106:algorithms
546:March 13,
521:March 11,
496:March 13,
232:Criticism
195:Websummit
44:Indonesia
466:Finovate
250:MSN News
246:Facebook
203:Hannover
130:unbanked
85:Facebook
28:big data
516:PCWorld
238:Äripäev
215:PCWorld
36:Finland
48:Poland
541:eWeek
448:(PDF)
441:(PDF)
417:(PDF)
410:(PDF)
385:(PDF)
374:(PDF)
335:(PDF)
219:eWeek
199:CeBIT
52:banks
40:Chile
22:is a
612:2013
548:2014
523:2014
498:2014
393:2015
356:help
337:: 4.
217:and
132:and
66:and
46:and
628::
564:.
539:.
514:.
489:.
464:.
376:.
347::
345:}}
341:{{
304:.
221:.
213:,
185:.
70:.
62:,
58:,
54:,
42:,
38:,
34:,
32:UK
614:.
574:.
550:.
525:.
500:.
474:.
426:.
395:.
358:)
354:(
319:.
Text is available under the Creative Commons Attribution-ShareAlike License. Additional terms may apply.