Department of Industrial Engineering and Management, Ben-Gurion University of the Negev, Beersheba, Israel.
Pagaya Technologies, Tel Aviv, Israel.
Sci Rep. 2022 Jul 25;12(1):12650. doi: 10.1038/s41598-022-16108-3.
In the current study, we set out to examine the viability of a novel approach to modeling human personality. Research in psychology suggests that people's personalities can be effectively described using five broad dimensions (the Five-Factor Model; FFM); however, the FFM potentially leaves room for improved predictive accuracy. We propose a novel approach to modeling human personality that is based on the maximization of the model's predictive accuracy. Unlike the FFM, which performs unsupervised dimensionality reduction, we utilized a supervised machine learning technique for dimensionality reduction of questionnaire data, using numerous psychologically meaningful outcomes as data labels (e.g., intelligence, well-being, sociability). The results showed that our five-dimensional personality summary, which we term the "Predictive Five" (PF), provides predictive performance that is better than the FFM on two independent validation datasets, and on a new set of outcome variables selected by an independent group of psychologists. The approach described herein has the promise of eventually providing an interpretable, low-dimensional personality representation, which is also highly predictive of behavior.
在当前的研究中，我们着手研究一种新颖的人类个性建模方法。心理学研究表明，人们的个性可以用五个广泛的维度（五因素模型；FFM）来有效地描述；然而，FFM 可能还有提高预测准确性的空间。我们提出了一种新颖的人类个性建模方法，该方法基于模型预测准确性的最大化。与执行无监督维度减少的 FFM 不同，我们利用监督机器学习技术对问卷数据进行维度减少，使用许多有意义的心理结果作为数据标签（例如，智力、幸福感、社交能力）。结果表明，我们的五维个性总结，我们称之为“预测五”（PF），在两个独立的验证数据集上提供了比 FFM 更好的预测性能，并且在一组由独立心理学家选择的新的结果变量上也提供了更好的预测性能。本文所描述的方法有望最终提供一种可解释的、低维的个性表示，同时也能高度预测行为。
