Resource-Efficient Hybrid Machine Learning Model for IoT SMS Spam Detection

The project titled “Resource-Efficient Hybrid Machine Learning Model for IoT SMS Spam Detection” aims to develop an automated system for classifying text messages into spam and non-spam categories. The dataset used for this project is the SMS Spam Collection Dataset, which contains labeled messages suitable for supervised learning tasks. The system combines classical machine learning models such as Naive Bayes, Support Vector Machine, and Logistic Regression with deep learning architectures like LSTM, CNN, and GRU to enhance prediction accuracy while maintaining resource efficiency. Preprocessing methods including text cleaning, tokenization, vectorization, and sequence padding are applied to prepare input for the respective models. The project is implemented using a web-based interface with modules for Home, Register, Login, Dashboard, Model, Prediction, and Logout, allowing users to interact with the system seamlessly. Evaluation metrics including accuracy, precision, recall, F1-score, and ROC-AUC are utilized to assess model performance. The hybrid approach ensures effective spam detection with optimized computational requirements.

Keywords: IoT, SMS spam detection, classification, hybrid model, Naive Bayes, SVM, Logistic Regression, LSTM, CNN, GRU