Main Article Content
Abstract
Email has widely used for generate the intrusion over the network. NLP based techniques already
introduced to extract the features from available data set and select the important features, and build the
train model for background knowledge (BK). TF-IDF is very popular technique which has been used by
recent researches. Correlation co-occurrence, weighted term frequency and Stanford NLP features has
used to build the train module. In this research we propose email phishing detection and prevention using
various machine learning algorithms. Initially system deals with synthetic female stammer data set which
already contains some normal as well as malicious contains. Many existing systems have already done
statistical approaches to detect malicious emails. The proposed idea illustrates supervised learning
approach to detect email spam’s. This research also focused on how system works with machine learning
as well as deep learning algorithms. The estimated results for propose system should be higher than
classical detection approaches.