Enhancing Spam Filter Using Naive Bayes and Count Vectorizer
Jiachen Liang · Advances in computer science research · 2023
This study delves into advancements in the realm of email spam filtration, a critical pillar in augmenting email security infrastructure.Given the unceasing challenges presented by unwarranted spam, the deployment of efficacious spam filtration methodologies remains imperative.Contemporary strategies encompass IP address filtering, rule-based filtering, and the employment of Naive Bayes algorithms.However, these methodologies often succumb to the continuously evolving spamming techniques.To counter these drawbacks, the current study proposes an enhanced spam filtering architecture anchored in machine learning techniques, which involves augmented spam data procurement, data processing, feature extraction via Term Frequency-Inverse Document Frequency (TF-IDF), and the implementation of machine learning models such as Naive Bayes and Support Vector Machines (SVM).This research conducts a comparative analysis of these machine learning classifiers and underscores the superior performance of the SVM Linear model in spam detection, achieving elevated accuracy levels while ensuring balanced precision and recall for both spam and non-spam emails.These findings underscore promising strides in the arena of email security.The study culminates by advocating for persistent research and the incorporation of advanced techniques to augment the accuracy and user-friendly nature of spam filtering systems.