IDENTIFYING MACHINE GENERATED TWEETS USING CNN-LSTM ARCHITECTURE
Alt Text: Identifying Machine-Generated Tweets Using CNN-LSTM Architecture
Title: Identifying Machine-Generated Tweets Using CNN-LSTM Architecture
Caption: Developing deep learning techniques to detect machine-generated tweets and social media deepfakes.
Description: This study proposes a CNN-LSTM-based deep learning framework for identifying machine-generated tweets using the publicly available Tweepfake dataset. Through comparative analysis against other deep learning models, the approach demonstrates superior accuracy in distinguishing human-generated versus bot-generated social media posts.
Keywords: Deepfake Detection, Machine Learning, Social Media Bots, CNN-LSTM, Tweet Classification
International Journal of Engineering and Techniques – Volume 10 Issue 3, June 2024
A. Ramesh1, E. Chitti Babu2
1Assistant Professor, Department of Computer Science & Engineering, Geethanjali Institute of Science and Technology, Gangavaram, Andhra Pradesh, India.
2Assistant Professor, Department of Computer Science & Engineering, Geethanjali Institute of Science and Technology, Gangavaram, Andhra Pradesh, India.
Abstract
Recent advancements in natural language generation have strengthened deep neural models, making them powerful tools for social media manipulation. Text-generative models enable adversaries to create deepfake posts that influence online discourse. This research introduces a CNN-LSTM-based deep learning framework for detecting machine-generated tweets, leveraging word embeddings and the Tweepfake dataset. The study conducts comparative analysis with baseline machine learning models and alternative deep learning architectures, highlighting the superior performance of the proposed method in identifying bot-generated tweets.
Keywords
Deepfake Detection, Machine Learning, Social Media Bots, CNN-LSTM, Tweet Classification
How to Cite
Ramesh, A., Chitti Babu, E., “Identifying Machine-Generated Tweets Using CNN-LSTM Architecture,” International Journal of Engineering and Techniques, Volume 10, Issue 3, June 2024. ISSN 2395-1303
Post Comment