DETECTION OF AI-GENERATED TWEETS USING FAST TEXT EMBEDDINGS WITH DEEP LEARNING TECHNIQUES

Read full text Summarise with VIRAT
Authors: IJERST
Cited by: 0
References: 21

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

The rapid growth of social media platforms has significantly increased the amount of textual content generated and shared online. With the advancement of Artificial Intelligence (AI) and Natural Language Processing (NLP), machines are now capable of generating human-like text that closely resembles genuine user content. While these technologies provide many benefits, they also introduce serious challenges such as the spread of misinformation, manipulation of public opinion, and automated bot-driven campaigns. Detecting AIgenerated tweets has therefore become an important task in maintaining the authenticity and reliability of online communication platforms. This project proposes an efficient system for detecting AI-generated tweets using FastText embeddings combined with deep learning techniques. The system begins with data preprocessing, where tweet text is cleaned, normalized, and tokenized to remove noise such as special characters, URLs, and stop words. After preprocessing, FastText embeddings are used to convert textual data into meaningful vector representations that capture semantic relationships and subword information. These feature representations are then fed into a deep learning classification model capable of distinguishing between human-written tweets and AI-generated content. The performance of the proposed model is evaluated using standard evaluation metrics including accuracy, precision, recall, and F1-score to ensure reliability and effectiveness. Experimental results indicate that the model can successfully identify AI-generated tweets with high accuracy and efficiency. The proposed system can be integrated into real-time social media monitoring tools to detect automated content, reduce the spread of misinformation, and enhance the transparency of online platforms. Overall, this research contributes to improving digital trust by providing an intelligent and scalable solution for identifying machine-generated social media content.

Fields of Study & Subjects

Authorship Attribution and Profiling Misinformation and Its Impacts Spam and Phishing Detection

Publication details

Journal / source
International Journal of Engineering Research and Science & Technology
Publication date
21 March 2026
Language
EN