Abstract

The rapid growth of social media has significantly influenced how individuals, especially new tech industry entrants, present themselves professionally. As these job market candidates transition into the workforce, social media behavior can provide insights into personal behavior. However, these social media platforms also pose risks, such as subtle misconduct traits (e.g., endorsements of harmful ideologies, engagement in cyberbullying) which may not be captured in traditional background checks or interviews. For professional networking and career development, a platform like LinkedIn can be a key source for analyzing formal interactions, job-seeking behaviors, and professional endorsements. Platforms like Instagram, TikTok and Threads are more informal and personal, where users often share lifestyle content, personal opinions, and entertainment-focused media, which can reveal underlying attitudes and behaviors not typically visible in professional spaces. Studies have shown that behaviors exhibited on one may differ significantly from those on others, yet all can impact professional perceptions. As a result, companies are increasingly incorporating social media screenings to identify questionable behaviors as input before making hiring decisions. While there has been plenty of research on workplace misconduct, fewer studies have focused on proactive methods to address this behavior before employees are hired. This research fills that gap by utilizing a cross-platform analysis technique that will compare user behavior across LinkedIn, Instagram, Threads and TikTok, identifying discrepancies between professional and personal personas. Natural Language Processing (NLP) will analyze text from posts and comments, searching for inappropriate language or patterns that could indicate misconduct. Sentiment analysis will assess emotional tone and consistency across platforms to gauge alignment with professional standards. Computer vision will be used to detect unprofessional or inappropriate content in images and videos, particularly on visual platforms like Instagram and TikTok. Tone, frequency of posts, and engagement in potentially harmful discussions will be collected to help determine misconduct behaviors. An AI-powered rating system will be developed based on the analysis of the social media content. This system may be a tool that recruiters from firms that are smaller with scarce resources a cost-effective tool for evaluating candidates without relying on expensive recruiting services. Findings will add to existing literature on social media tools, misconduct, and recruitment. For job candidates, this system provides an inexpensive tool to evaluate their own social media presence by reviewing their social media activity prior to entering the job market, which can add to candidate market readiness. Further, findings will contribute to both the understanding of digital professionalism and the development of practical tools for more informed hiring decision.

Download

Share

COinS