ISSN No:2250-3676 ----- Crossref DOI Prefix: 10.64771 ----- Impact Factor: 9.625
   Email: editor@ijesat.com , ijesatj@gmail.com,   

(Peer Reviewed, Referred & Indexed Journal)


    AI-Driven Online Abuse Detection And Automated Social Media Content Moderation

    Ponnaboina Vinayak Yadav, Palakurthi Sai Kiran, S Akhil, Chinthakindi Aditya, Dr D MURALI

    Author

    ID: 4057

    DOI:

    Abstract :

    The Rapid Expansion Of Social Networking Platforms Has Increased The Prevalence Of Cyberbullying, Creating Significant Social And Psychological Challenges For Online Users. To Address This Issue, This Study Presents An Intelligent Cyberbullying Detection And Content Moderation Framework That Combines Natural Language Processing, Machine Learning, And Image-based Text Analysis. The Proposed System Employs Term Frequency–Inverse Document Frequency (TF-IDF) For Textual Feature Extraction And A Class-balanced Logistic Regression Model For Identifying Harmful Content In Social Media Posts. To Extend Moderation Beyond Conventional Text Inputs, An Optical Character Recognition (OCR) Module Powered By EasyOCR Extracts And Analyzes Text Embedded Within Images, Memes, And Screenshots. The Framework Further Incorporates A Hybrid Decision Mechanism That Integrates Rule-based Validation With Probabilistic Machine Learning Predictions To Improve Detection Reliability. Unlike Traditional Moderation Systems That Only Classify Content, The Proposed Solution Actively Promotes Positive Communication By Generating Constructive Alternative Suggestions For Flagged Messages. Experimental Evaluation On The Davidson Twitter Dataset Demonstrates A Classification Accuracy Of 94.2%, While Maintaining Low Computational Overhead And Real-time Response Capability. The Image-analysis Pipeline Also Achieves Effective Performance In Detecting Abusive Content Contained Within Visual Media. Implemented As A Lightweight Flask-based Web Application, The System Operates Efficiently On Standard Hardware Without Requiring Cloud Resources, Making It Suitable For Educational Institutions, Social Media Platforms, And Community Moderation Environments. The Results Highlight The Effectiveness Of Combining Interpretable Machine Learning Techniques With OCR-enabled Multimodal Analysis For Practical And Scalable Cyberbullying Prevention.

    Published:

    18-3-2024

    Issue:

    Vol. 24 No. 3 (2024)


    Page Nos:

    230-236


    Section:

    Articles

    License:

    This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

    How to Cite

    Ponnaboina Vinayak Yadav, Palakurthi Sai Kiran, S Akhil, Chinthakindi Aditya, Dr D MURALI, AI-Driven Online Abuse Detection and Automated Social Media Content Moderation , 2024, International Journal of Engineering Sciences and Advanced Technology, 24(3), Page 230-236, ISSN No: 2250-3676.

    DOI: