Hybrid Feature-Based Comparative Analysis of Supervised and Unsupervised ML-Based Intelligent Bot Detection Model for Social Media Data

DOI

10.2991/978-94-6239-791-0_10How to use a DOI?

Keywords

Bot Detection; Machine Learning; Supervised Learning; Unsupervised Learning; Random Forest; Support Vector Machine; Naive Bayes; Logistic Regression; K-Means; TF-IDF; Confusion Matrix; ROC-Curve; F1-Score

Abstract

Social media has become an important part of how people communicate and share information. However, the increasing use of it has led to rise in fake and automated bot accounts. These accounts create miscommunication and often spread fake news, spam, misleading information. Since bots are becoming difficult to identify, detecting them has become a necessary problem. In this study, we have developed a bot detection model - approach by combining text-based end user behaviour features. TF-IDF is used to analyse the text, along with features such as account information, engagement ratio, retweet activity, and mention patterns. 5 machine learning models were tested: Random Forest, Support Vector Machine (SVM), Logistic Regression, Naive Bayes, and K-Means clustering. Among these models, Random Forest gave the best result, with an accuracy of 90%, while K-Means showed lower performance. The results show that using both text and user behavioural features can help improve bot detection and can support the development of safer and more reliable social media platforms which can be used by everyone.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Siddhi Kale
AU  - Vishal Shirsath
AU  - Pradip Salve
AU  - Akriti Biswas
PY  - 2026
DA  - 2026/10/09
TI  - Hybrid Feature-Based Comparative Analysis of Supervised and Unsupervised ML-Based Intelligent Bot Detection Model for Social Media Data
BT  - Proceedings of the International Conference on Artificial Intelligence and Machine Learning for Agro-Ecosystems (ICAIML-AE 2026)
PB  - Atlantis Press
SP  - 96
EP  - 106
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6239-791-0_10
DO  - 10.2991/978-94-6239-791-0_10
ID  - Kale2026
ER  -