[Paper Review] My tweets bring all the traits to the yard: Predicting personality and relational traits in Online Social Networks

Dimitra Karanatsiou, Pavlos Sermpezis+4 more|arXiv (Cornell University)|Sep 22, 2020

Complex Network Analysis Techniques4 citations

TL;DR

This paper proposes a machine learning model that predicts both individual personality traits and relational traits (e.g., attachment orientations) from social media behavior, using a feature engineering approach combining linguistic, behavioral, and emotional features. The model, inspired by psychological theory and leveraging interrelations among traits, achieves higher accuracy than state-of-the-art methods and successfully distinguishes between random users and organizational leaders based solely on psychological profiles.

ABSTRACT

Users in Online Social Networks (OSN) leaves traces that reflect their personality characteristics. The study of these traces is important for a number of fields, such as a social science, psychology, OSN, marketing, and others. Despite a marked increase on research in personality prediction on based on online behavior the focus has been heavily on individual personality traits largely neglecting relational facets of personality. This study aims to address this gap by providing a prediction model for a holistic personality profiling in OSNs that included socio-relational traits (attachment orientations) in combination with standard personality traits. Specifically, we first designed a feature engineering methodology that extracts a wide range of features (accounting for behavior, language, and emotions) from OSN accounts of users. Then, we designed a machine learning model that predicts scores for the psychological traits of the users based on the extracted features. The proposed model architecture is inspired by characteristics embedded in psychological theory, i.e, utilizing interrelations among personality facets, and leads to increased accuracy in comparison with the state of the art approaches. To demonstrate the usefulness of this approach, we applied our model to two datasets, one of random OSN users and one of organizational leaders, and compared their psychological profiles. Our findings demonstrate that the two groups can be clearly separated by only using their psychological profiles, which opens a promising direction for future research on OSN user characterization and classification.

Motivation & Objective

• To address the gap in personality prediction research that largely neglects relational traits such as attachment orientations.
• To develop a holistic personality profiling model that integrates standard personality traits with socio-relational dimensions in online social networks.
• To design a feature engineering methodology capturing behavior, language, and emotional patterns from OSN user accounts.
• To create a machine learning model informed by psychological theory, leveraging interrelations among personality facets for improved accuracy.
• To evaluate the model on diverse user groups, including random OSN users and organizational leaders, to assess its discriminative power.

Proposed method

• A comprehensive feature engineering pipeline extracts linguistic, behavioral, and emotional features from user tweets.
• Features include lexical diversity, sentiment expression, pronoun usage, and syntactic complexity to represent personality and relational traits.
• The model architecture is informed by psychological theory, explicitly modeling interrelations among personality facets to improve predictive performance.
• A supervised machine learning approach is trained on labeled user data to predict scores on standardized personality and attachment orientation scales.
• The model is evaluated on two datasets: one of random OSN users and one of organizational leaders, using cross-validation and classification metrics.
• Psychological profiles are compared between groups to assess model utility in user characterization and classification.

Experimental results

Research questions

• RQ1 Can a machine learning model effectively predict both individual personality traits and relational traits such as attachment orientations from social media content?
• RQ2 How does incorporating interrelations among personality facets, as informed by psychological theory, improve prediction accuracy compared to standard approaches?
• RQ3 To what extent can psychological profiles derived from social media behavior distinguish between different user types, such as random users and organizational leaders?
• RQ4 What specific linguistic, behavioral, and emotional features are most predictive of personality and relational traits in online social networks?

Key findings

• The proposed model achieves higher prediction accuracy than state-of-the-art approaches by leveraging interrelations among personality facets.
• The model successfully distinguishes between random OSN users and organizational leaders based solely on their psychological profiles.
• Relational traits such as attachment orientations contribute significantly to the discriminative power of the psychological profiles.
• Linguistic features such as pronoun usage, sentiment expression, and lexical diversity are strong predictors of both personality and relational traits.
• The integration of socio-relational dimensions into personality profiling enhances the model’s ability to characterize and classify user types in online social networks.