Private traits and attributes are predictable from digital records of human behavior

Kosinski, Michal; Stillwell, David; Graepel, Thore

doi:10.1073/pnas.1218772110

Cited by 2,232 publications

(1,589 citation statements)

References 30 publications

Supporting

Mentioning

1,401

Contrasting

Unclassified

Order By: Relevance

“…Few would be willing to cover their faces while in public. As this and other 661 studies show (e.g., Kosinski et al, 2013), such willingly shared digital footprints can be used to 662 reveal intimate traits. Thus, while we certainly need to urgently work towards policies and 663 technologies aimed at protecting privacy, in the long term the further erosion of privacy seems 664 inevitable.…”

Section: General Discussion 519mentioning

confidence: 78%

Deep neural networks are more accurate than humans at detecting sexual orientation from facial images.

Wang

Kosinski

2018

Journal of Personality and Social Psychology

Self Cite

487

262

View full text Add to dashboard Cite

29We show that faces contain much more information about sexual orientation than can be 30 perceived and interpreted by the human brain. We used deep neural networks to extract features 31 from 35,326 facial images. These features were entered into a logistic regression aimed at 32 classifying sexual orientation. Given a single facial image, a classifier could correctly distinguish 33 between gay and heterosexual men in 81% of cases, and in 71% of cases for women. Human 34 judges achieved much lower accuracy: 61% for men and 54% for women. The accuracy of the 35 algorithm increased to 91% and 83%, respectively, given five facial images per person. Facial 36 features employed by the classifier included both fixed (e.g., nose shape) and transient facial 37features (e.g., grooming style). Consistent with the prenatal hormone theory of sexual 38 orientation, gay men and women tended to have gender-atypical facial morphology, expression, 39 and grooming styles. Prediction models aimed at gender alone allowed for detecting gay males 40 with 57% accuracy and gay females with 58% accuracy. Those findings advance our 41 understanding of the origins of sexual orientation and the limits of human perception. 42Additionally, given that companies and governments are increasingly using computer vision 43 algorithms to detect people's intimate traits, our findings expose a threat to the privacy and safety 44 of gay men and women. 45

show abstract

Section: General Discussion 519mentioning

confidence: 78%

Deep neural networks are more accurate than humans at detecting sexual orientation from facial images.

Wang

Kosinski

2018

Journal of Personality and Social Psychology

Self Cite

487

262

View full text Add to dashboard Cite

show abstract

“…Additional socioeconomic and environmental covariates as well as information pertaining to psychological and personality attributes [42][43][44] may address relevant dimensions of consumer decision-making and could increase predictive performance tremendously.…”

Section: Resultsmentioning

confidence: 99%

Customer Lifetime Value Prediction in Non-Contractual Freemium Settings: Chasing High-Value Users Using Deep Neural Networks and SMOTE

Sifa¹,

Runge²,

Bauckhage³

et al. 2018

Proceedings of the Annual Hawaii International Conference on System Sciences

View full text Add to dashboard Cite

In non-contractual freemium and sharing economy settings, a small share of users often drives the largest part of revenue for firms and co-finances the free provision of the product or service to a large number of users. Successfully retaining and upselling such high-value users can be crucial to firms' survival. Predictions of customers' Lifetime Value (LTV) are a much used tool to identify high-value users and inform marketing initiatives. This paper frames the related prediction problem and applies a number of common machine learning methods for the prediction of individual-level LTV. As only a small subset of users ever makes a purchase, data are highly imbalanced. The study therefore combines said methods with synthetic minority oversampling (SMOTE) in an attempt to achieve better prediction performance. Results indicate that data augmentation with SMOTE improves prediction performance for premium and high-value users, especially when used in combination with deep neural networks.

show abstract

“…from likes and Facebook profiles [7]. Furthermore, online expressions of aesthetic preferences convey an impression in terms of characteristics like prestige, differentiation or authenticity [8].…”

mentioning

confidence: 99%

The Pictures We Like Are Our Image: Continuous Mapping of Favorite Pictures into Self-Assessed and Attributed Personality Traits

Segalin

Perina

Cristani

et al. 2017

IEEE Trans. Affective Comput.

View full text Add to dashboard Cite

Abstract-Flickr allows its users to tag the pictures they like as "favorite". As a result, many users of the popular photo-sharing platform produce galleries of favorite pictures. This article proposes new approaches, based on Computational Aesthetics, capable to infer the personality traits of Flickr users from the galleries above. In particular, the approaches map low-level features extracted from the pictures into numerical scores corresponding to the Big-Five Traits, both self-assessed and attributed. The experiments were performed over 60,000 pictures tagged as favorite by 300 users (the PsychoFlickr Corpus). The results show that it is possible to predict beyond chance both self-assessed and attributed traits. In line with the state-of-theart of Personality Computing, these latter are predicted with higher effectiveness (correlation up to 0.68 between actual and predicted traits).

show abstract

Private traits and attributes are predictable from digital records of human behavior

Cited by 2,232 publications

References 30 publications

Deep neural networks are more accurate than humans at detecting sexual orientation from facial images.

Deep neural networks are more accurate than humans at detecting sexual orientation from facial images.

Customer Lifetime Value Prediction in Non-Contractual Freemium Settings: Chasing High-Value Users Using Deep Neural Networks and SMOTE

The Pictures We Like Are Our Image: Continuous Mapping of Favorite Pictures into Self-Assessed and Attributed Personality Traits

Contact Info

Product

Resources

About