VIKI: Systematic Cross-Platform Profile Inference of Online Users

TL;DR

VIKI introduces a systematic cross-platform profiling pipeline that links user identities, extracts multi-faceted across-platform personas, and integrates them into unified cross-platform profiles. Using an LLM-based inference for $T_i(u,p)$ across platforms, NLP category classifiers for professional/personal interests, and PerspectiveAPI for offensiveness, the approach reveals substantial cross-platform persona changes, with neuroticism showing the largest shift and significant links between offense and personality traits. Experiments on 1.2K tech users across GitHub, LinkedIn, and X demonstrate that about 78% change at least one trait across platforms, and KS tests confirm platform-driven differences in OCEAN distributions. The work also uncovers cross-platform patterns in interests and offense, identifies meaningful user clusters, and discusses ethical implications and limitations, offering a foundation for more nuanced cross-platform user profiling and its potential applications and policy considerations.

Abstract

What can we learn about online users by comparing their profiles across different platforms? We use the term profile to represent displayed personality traits, interests, and behavioral patterns (e.g., offensiveness). We also use the term {\it displayed personas} to refer to the personas that users manifest on a platform. Though individuals have a single real persona, it is not difficult to imagine that people can behave differently in different ``contexts'' as it happens in real life (e.g., behavior in office, bar, football game). The vast majority of previous studies have focused on profiling users on a single platform. Here, we propose VIKI, a systematic methodology for extracting and integrating the displayed personas of users across different social platforms. First, we extract multiple types of information, including displayed personality traits, interests, and offensiveness. Second, we evaluate, combine, and introduce methods to summarize and visualize cross-platform profiles. Finally, we evaluate VIKI on a dataset that spans three platforms -- GitHub, LinkedIn, and X. Our experiments show that displayed personas change significantly across platforms, with over 78% of users exhibiting a significant change. For instance, we find that neuroticism exhibits the largest absolute change. We also identify significant correlations between offensive behavior and displayed personality traits. Overall, we consider VIKI as an essential building block for systematic and nuanced profiling of users across platforms.

Figures (3)

• Figure 1: Some users display significantly different personalities in different platforms. We show the change in the openness and neuroticism traits (on a 1-5 scale) of users across LinkedIn and X using a Complementary Cumulative Distribution Function (CCDF). Over 60% of users display a change in neuroticism by at least one point and 20% by at least two points between the two platforms. The distribution of change is substantial but less pronounced for openness.
• Figure 2: The VIKI methodology for cross-platform user profiling.
• Figure 3: Magnitude of change in multiple OCEAN personality traits of users in DPop across LinkedIn and X.