Detecting Fake Companies on LinkedIn: A Machine Learning Approach

Abstract

The proliferation of fake companies on professional networking platforms like LinkedIn poses significant risks to users, including scams, identity theft, and data breaches. To address this issue, I propose a novel approach utilizing machine learning techniques to detect the legitimacy of companies on LinkedIn. I conducted extensive research to identify key factors influencing company legitimacy and incorporated them into a comprehensive dataset. Employing popular machine learning algorithms such as Support Vector Machine, Decision Tree, and K-Nearest Neighbor, I trained models to predict whether a company is real or fake. Additionally, I developed a risk labeling system to assess the level of risk associated with each company. My approach includes the integration of a scrapper component to extract essential information from resumes while ensuring user privacy and security. Through rigorous analysis and evaluation, I demonstrate the effectiveness of My methodology in accurately identifying fake companies and reducing risks for LinkedIn users. This research contributes valuable insights and practical solutions to enhance trust and security in professional networking environments.