Email Spam Filteration with Machine Learning
Manmohan Singh Sajwan, Abhinav Singh Mahar, Priyanshu Rawat, Khushi Sharma, Purushottam Das · 2024
Email spam has become a significant contemporary issue, mainly propelled by the rapid expansion of internet users. Correspondingly, the incidence of email spam is on the rise. These unsolicited communications serve as vehicles for nefarious and unethical purposes, including phishing and fraudulent activities. Using sophisticated techniques spammers create malicious links capable of compromising system security and infiltrating personal data. Crafting counterfeit profiles and email accounts is effortlessly achieved by spammers, enabling them to masquerade as genuine entities in their deceptive communications. Especially vulnerable individuals are unaware of such schemes, thus emphasizing the critical need to identify and mitigate such deceptive emails. This project aims to detect fraud email spam through the application of various machine learning algorithms. Numerous machine learning algorithms such as Naïve Bayes, SVMs, KNN & Decision Trees were used with specific datasets for this project. Integration has been done for the selection of the most effective algorithm for email spam detection as per definitive metric standards. Thus, our key findings indicate that Multinomial Naïve Bayes Algorithm is most suitable for spam filtration.