A Comparative Survey of Personality Trait Prediction Using Digital Footprints with Emphasis on Spending Behaviour

Abstract

The rapid growth of digital footprints has enabled large-scale analysis of human psychological characteristics using computational techniques. Among various digital traces, spending behaviour captured through transaction data provides an objective and continuous representation of real-world decision-making. This paper presents a survey and comparative analysis of personality prediction using digital footprints, with a primary focus on expenditure-based behavioural data. Existing studies leveraging questionnaire responses, social media activity, smartphone sensing, and multimodal data sources are systematically reviewed and compared with transaction-based approaches. Machine learning and deep learning models are analysed for their effectiveness in predicting Big Five personality traits along with additional traits such as self-control and materialism. The findings indicate that while predictive accuracy for broad personality dimensions remains moderate, more specific traits exhibit stronger and more stable associations with spending patterns. Furthermore, spending-based personality predictions demonstrate relative robustness across different socioeconomic groups when compared to other digital behavioural sources. The survey highlights the potential of transaction data as a scalable and ecologically valid alternative for personality assessment, while also emphasizing the ethical and privacy considerations associated with large-scale psychological profiling.

Introduction

Personality is a combination of behaviors, emotions, and traits shaped by biological and environmental factors. It varies across individuals and is influenced by thinking, feelings, and actions. The Big Five personality traits—Openness, Conscientiousness, Extraversion, Agreeableness, and Neuroticism—along with Self-Control and Materialism, provide a framework for understanding human behavior. Each trait reflects distinct tendencies, such as creativity (Openness), discipline (Conscientiousness), sociability (Extraversion), empathy (Agreeableness), emotional stability (Neuroticism), impulse regulation (Self-Control), and value placed on possessions (Materialism).

Personality prediction has become increasingly relevant due to large-scale digital data availability. Researchers utilize multiple data sources for personality inference: traditional questionnaires, social media activity, smartphone sensor data, and multimodal behavioral data, including spending patterns. Machine learning and deep learning methods applied to these datasets enable accurate prediction of personality traits, with multimodal approaches providing the most robust results. Transactional and spending data, in particular, can effectively predict traits like Self-Control and Materialism.

Methodologically, personality prediction frameworks involve data collection, preprocessing, feature extraction, feature selection, and predictive modeling using algorithms such as Random Forest, SVM, Gradient Boosting, and neural networks. These approaches support continuous, data-driven personality assessment for applications in personalized services, behavioral analytics, and human–computer interaction.

Conclusion

This paper presented a comprehensive survey and comparative analysis of human psychological personality prediction using digital footprints, with a particular focus on spending behaviour derived from transaction data. By examining existing research across questionnaire-based methods, social media data, smartphone sensing, and multimodal approaches, the study highlighted the growing role of machine learning and deep learning techniques in large-scale personality inference. The analysis demonstrated that spending behaviour serves as a reliable and objective digital footprint for predicting both Big Five personality traits and additional traits such as self-control and materialism. Compared to other digital sources, spending-based predictions exhibit relatively stable performance across different socioeconomic groups and over time. While predictive accuracies remain moderate for broader personality dimensions, more focused traits show stronger associations with expenditure patterns. Overall, this survey underscores the potential of transaction data as a scalable alternative for psychological assessment, while also emphasizing the need for responsible use and ethical safeguards as digital personality prediction technologies continue to evolve.

References

[1] S. Shenavi, R. K. Rao, and P. R. Kumar, “Personality trait classification using machine learning techniques based on questionnaire data,” International Journal of Advanced Computer Science and Applications, vol. 10, no. 6, pp. 412–419, 2019. [2] L. Chi, J. Li, and X. Fu, “Large-scale personality trait analysis using Big Five questionnaire and neural networks,” IEEE Access, vol. 7, pp. 176589–176600, 2019. [3] A. Karanatsiou, M. Thelwall, and K. Kappas, “Predicting Big Five personality traits from Twitter data using multi-output regression,” Computers in Human Behavior, vol. 86, pp. 1–10, 2018. [4] J. Billy, S. Guntuku, and L. Ungar, “Personality prediction from social media text using contextual word embeddings,” Proceedings of the AAAI Conference on Artificial Intelligence, vol. 34, no. 5, pp. 697–704, 2020. [5] N. D. Lane, E. Miluzzo, H. Lu, D. Peebles, T. Choudhury, and A. T. Campbell, “A survey of mobile phone sensing,” IEEE Communications Magazine, vol. 48, no. 9, pp. 140–150, 2010. [6] R. Wang, F. Chen, Z. Chen, T. Li, G. Harari, S. T. Campbell, and A. T. Campbell, “StudentLife: Assessing mental health, academic performance and behavioral trends of college students using smartphones,” Proceedings of the ACM International Joint Conference on Pervasive and Ubiquitous Computing, pp. 3–14, 2014. [7] Y. Wang, J. Li, and S. Zhao, “Multimodal personality prediction based on questionnaire and behavioral data fusion,” Expert Systems with Applications, vol. 113, pp. 456–465, 2018. [8] S. C. Matz, M. Kosinski, G. Nave, and D. J. Stillwell, “Psychological targeting as an effective approach to digital mass persuasion,” Proceedings of the National Academy of Sciences (PNAS), vol. 114, no. 48, pp. 12714–12719, 2017. [9] Pooja Sharma, Sachin Kumar, Analyzing Customer Behavior on Online Products Using Decision Tree Machine Learning 2022 International Conference on Advances in Computing, Communication and Materials (ICACCM), ISBN:978-1-6654- 7439-9, Dehradun,India,10-11 Nov. 2022 [10] R. Valanarasu, , Comparative Analysis for Personality Prediction by Digital Footprints in Social Media, Journal of Information Technologyand Digital World (2021) Vol. 03/ No. 02 Pages: 77-91 [11] Manil Vaidhya; Bikash Shrestha; Bishal Sainju; Kiran Khaniya; Aman Shakya, Personality Traits Analysis from Facebook Data,,21st International Computer Science and Engineering Conference (ICSEC) ,DOI: 10.1109/ICSEC41384.2017,15-18 Nov. 2017 [12] Kaan Sonmezoz; Ozgur Ugur; Banu Diri,MBTI Personality Prediction With Machine Learning,2020 28th Signal Process- ing and Communications Applications Conference (SIU),DOI: 10.1109/SIU49456.20205-7 Oct. 2020 [13] Avishka Jaysundara; Dimanthinie De Silva; Priyantha Kumarawadu, Personality Prediction of Social Network Users using LSTM based Sentiment Analysis,2022 International Conference on Smart Tech- nologies and Systems for Next Generation Computing (ICSTSN), DOI: 10.1109/ICSTSN53084.2022. [14] Natkamon Tovanich, Simone Centellegher, Nacera Bennacer Seghouani, Joe Gladstone, Sandra Matz and Bruno Lepri, Inferring psychological traits from spending categories and dynamic consumption patterns, EPJ Data Science volume 10, Article number: 24 (2021) [15] Sanjit Kumar R, Shrivatson R G, Rishi Priyan S, Padmavathy T, Personality Prediction using Twitter Data, InternationalResearch Journal of Engineering and Technology (IRJET) e-ISSN: 2395- 0056, Volume: 07, Issue: 07 — July 2020, ISSN: 2395-007. [16] Gladstone, J. J., Matz, S. C., Lemaire, A. Can psychological traitsbe inferred from spending? Evidence from transaction data.Evidence from transaction data. Psychological Science, 30(7), 1087–1096,2019

Copyright

Copyright © 2025 Mrs. Arpitha C. N., Dr. Sunitha M. R.. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.