Spam Detection System Combining Cellular Automata and Naive Bayes Classifier
Fatiha Barigou, Naouel Barigou, Baghdad Atmani · 2012
In this study, we focus on the problem of spam detection. Based on a cellular automaton approach and naive Bayes technique which are built as individual classifiers we evaluate a novel method combining multiple classifiers diversified both by feature selection and different classifiers to determine whether we can more accurately detect Spam. This approach combines decisions from three cellular automata diversified by feature selection with that of naive Bayes classifier. Experimental results show that the proposed combination increases the classification performance as measured on LingSpam dataset.