Treffer: Multiscale Models for Unified Filtering Across Email, Social Media, and Mobile Networks.

The rapid increase of unsolicited content, or spam, across email, social media, and SMS platforms presents significant cybersecurity challenges. The existing approaches for spam detection can barely be adopted by different platforms due to their limitations. Prompting this study to develop a unified spam detection framework leveraging machine learning (ML) and deep learning (DL) models, including Naive Bayes, Support Vector Machines (SVM), Long Short-Term Memory (LSTM), and BERT. Advanced preprocessing techniques were proposed together with some ensemble methods to achieve scalability over multiple platforms. Results on all three diverse datasets confirm that the BERT and ensemble models consistently outperform traditional approaches, yielding accuracies as high as 96.2% for email, 92.9% for SMS, and 89.7% for social media. Such models perform well in extracting contextual and sequential subtleties from spam content and can hence be considered robust to continuously changing spamming tactics. This unified approach simplifies the detection of spam across various communication platforms while enhancing security and user experience. Practical implementation is supported with a Python-based interface that does classification in real time. Future directions may include extending this model for multilingual support and adversarial resilience. [ABSTRACT FROM AUTHOR]

Copyright of Anchor University Journal of Science & Technology (AUJST) is the property of Anchor University, Lagos, Nigeria and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)