Integration of data clustering to intelligent email classifier

Abstract

In recent years there are more than billion of email users .People are dependent on electronic mails. ft has become an urgent and crucial component of communication .One of the cheapest and fastest communication mean. Emails can grow at a large scale. Sorting the emails according to their content can be a useful and time saving. Cluster analysis can come in action while classifying the email documents .Cluster analysis is a sub-field in machine learning a sub field of artificial intelligence, that refers to a group of algorithms that try to find a natural grouping of objects based on some objective metric. In general this problem is hard because a good grouping might be subjective. The k-means algorithm is one of the simplest and predominantly used algorithms for unsupervised learning procedure clustering. Existing email clusters are supervised clusters. In this thesis, we inherit the simplest machine learning strategies like k-means, k nearest neighbors and represent an intelligent email classifier combined with both supervised and unsupervised algorithm that categorizes email documents based on the body contents.