International Research Journal of Education and Technology

The rapid expansion of digital communication platforms such as email, SMS, and social networking sites has significantly increased the volume of spam messages. Spam messages not only cause inconvenience but also lead to serious cybersecurity threats including phishing attacks, identity theft, malware distribution, and financial fraud. Traditional rule-based filtering systems are no longer sufficient to detect modern and dynamically evolving spam patterns. Therefore, intelligent machine learning techniques are required to enhance detection accuracy and adaptability.

This research proposes a machine learning-based spam detection system using the Multinomial Naive Bayes classification algorithm. The system is implemented as a web-based application using the Django framework. Text preprocessing techniques including tokenization, stop-word removal, normalization, and TF-IDF vectorization are applied to convert raw textual data into structured numerical features. The model is trained on a labeled dataset and evaluated using performance metrics such as Accuracy, Precision, Recall, and F1-Score. Experimental results show that the proposed model achieves high classification accuracy and performs efficiently in real-time prediction scenarios.

Keywords: Spam Detection, Email Filtering, SMS Classification, Machine Learning, Supervised Learning, Naive Bayes Classifier, Multinomial Naive Bayes, Text Mining, Text Classification, Natural Language Processing (NLP), TF-IDF, Feature Extraction, Data Preprocessing, Tokenization, Stop Word Removal, Probabilistic Model, Bayesian Theorem, Classification Algorithm, Predictive Modeling, Information Retrieval, Data Mining, Django Framework, Web Application, Real-Time Prediction, Model Evaluation, Accuracy, Precision, Recall, F1-Score, Confusion Matrix, Cybersecurity, Phishing Detection, Content Filtering, Intelligent Filtering System.