Year: 2024
Type: article
Abstract: Social media platforms have become a very integral part of everyday life, enabling people to connect, communicate, and share information across the globe without time or location barriers.However, these platforms are increasingly being targeted by malicious actors who create fake profiles to deceive users, spread false information, and engage in fraudulent activities.It is vital for a social network to detect fake profiles and eliminate them as it ensures security and integrity.The conventional ways used to detect fake profiles typically depend on heuristic methods or classification based on simple rules.The need to improve accuracy toward fake profile detection has given motivation to look into new kinds of advanced computational methods for this purpose.This paper therefore presents an approach by utilizing machine learning and NLP techniques to enhance the process in detecting and classifying the fake profiles on social networks.Our method uses algorithms such as Support Vector Machines (SVM) and Naïve Bayes for efficient classification based on the analysis of user behavior, textual content, and metadata associated with profiles.NLP techniques are particularly useful for extracting linguistic patterns and identifying anomalies in user-generated content, further improving detection rates.The proposed framework aims to achieve higher accuracy and reliability compared to traditional methods.This research thus addresses the challenges associated with fake profiles and improves the overall security of social media platforms and promotes a safer online environment for users.
Language: English
Topic: Spam and Phishing Detection
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
OpenAlex ID: W4405890666