An Efficient Email Spam Detection Utilizing Machine Learning Approaches

In this paper, email information was categorized as ham email and spam email by utilizing directed learning calculations. The issues of email spam discovery framework are related with high dimensionality in element determination cycle and low exactness of spam email characterization. On the other hand, feature extraction is emerging as a globalized enhancement with the implication of artificial intelligence [AI] techniques to reduce the superfluous and repetitive data and generates satisfactory outcomes with high precision. With reference to these technologies, this paper presents an element determination calculation depending on the particle swarm optimization (PSO) with three ML approaches such as decision tree, support vector machine [SVM], and Naive Bayes. The proposed PSO-based component choice calculation looks through the element space for obtaining the best element subsets. The classifier execution and length of the selected highlight vector as a classifier input are considered for executing the assessment by utilizing UCI spambase dataset. Exploratory outcomes show that the PSO-based element choice calculation was introduced to create incredible component determination results with the insignificant arrangement of selected features, which has to be brought about by a high exactness of spam email order depending on three ML classifiers. The outcomes show that the SVM-PSO-based element determination technique can obtain the most elevated precision from 93.62 to 95.97%.