Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{168717,
        author = {Naman Jain and Adarsh Upadhyay and Sheenam Naaz},
        title = {Email Anomaly Detection Using Machine Learning Algorithms},
        journal = {International Journal of Innovative Research in Technology},
        year = {2024},
        volume = {11},
        number = {5},
        pages = {2377-2387},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=168717},
        abstract = {Email security is crucial as it plays a major role in current business operations. Sophisticated anomalies and new threats are frequently missed by conventional rule-based email security systems. Our work addresses this by mixing machine learning techniques including ensemble learning, Auto-ML, meta-learning, transformers, and anomaly detection approaches into an innovative way to improve email anomaly identification. While auto-ML frameworks streamline the machine learning workflow by automating feature engineering, model selection, and hyperparameter tweaking, ensemble learning mixes many base models to increase prediction accuracy and robustness. Meta-learning enhances generalisation and adaptability by enabling systems to draw lessons from the past and adjust to novel circumstances. Advanced text representation skills are provided by transformers, which are well-known for their efficiency in processing sequential data. These capabilities are essential for analysing email content. Techniques for detecting anomalies in email behaviour can spot departures from the usual and indicate possible malicious activity. Our system outperforms conventional approaches by providing a comprehensive solution for email anomaly detection through the integration of different techniques. Experiments conducted on real-world email datasets demonstrate increased precision, increased detection rates, and decreased false positives, indicating the system's usefulness in enhancing email security.},
        keywords = {Machine learning, Naïve Bayes, support vector machine-nearest neighbor, random forest, bagging, boosting, neural networks.},
        month = {November},
        }