Cyberus – Measuring Risk

Abstract : Cyberus is an advanced software tool designed to assess the legitimacy and potential risks associated with messages. By leveraging machine learning algorithms and datasets, Cyberus provides users with a comprehensive risk analysis, enabling them to make informed decisions in their digital communication. The tool incorporates lexical analysis and URL analysis to evaluate the content and embedded links in messages, offering accurate and timely risk assessments. The development of Cyberus involved the collection and analysis of various datasets, including spam mail, spam SMS, and malicious URL datasets. Machine learning models, such as Support Vector Classifier and Decision Tree, were trained on these datasets to achieve high accuracy in identifying spam messages and malicious URLs. The tool takes user input, divides it into text and URLs, and applies the trained models to calculate an overall Cyberus Risk Index, indicating the level of illegitimacy and potential risks associated with the message. The implementation of Cyberus provides an abstract view of its design and functionality. The detailed implementation can be explored on the official GitHub repository https://github.com/s-m quadri/cyberus, as Cyberus is an open-source project. Performance analysis demonstrates the effectiveness of Cyberus in terms of accuracy and efficiency. As a versatile and adaptable tool, Cyberus holds future potential in improving message analysis and risk assessment, further enhancing user protection in the ever-evolving digital landscape. Overall, Cyberus serves as a valuable resource for individuals and organizations seeking to evaluate the legitimacy of messages, protect themselves from cyber threats, and make confident decisions in their digital interactions.

Keyword Spam Detection, Machine Learning, Spam Email, Spam Filtering, Opinion Spam.