A Hybrid Approach for Content Filtering and Blacklisting in OSN

Online Social Networks (OSNs) are becoming increasingly popular due to their extensive communication capabilities, enabling numerous social interactions and content sharing opportunities. However, content posting on these platforms often faces challenges such as unwanted comments and vulgar language. The proposed paper introduces a hybrid approach to enhance content filtering and blacklisting mechanisms in OSNs, along with user image capturing. The rising occurrence of undesirable content, including spam, offensive material, and malicious links, underscores the need for a more robust and adaptive system. The hybrid approach merges rule-based filtering with machine learning techniques to improve the accuracy and efficiency of content moderation. Predefined criteria are used in rule-based filtering and semantic text classification algorithms to identify and categorize unwanted content. Simultaneously, machine learning models, trained on extensive and varied datasets, offer the ability to dynamically adapt to evolving patterns and subtleties in user-generated content. Furthermore, a blacklisting mechanism is included to proactively detect and block content from known malicious sources. The proposed hybrid approach aims to create a safer and more enjoyable online experience for users by reducing the presence of harmful content.