A novel approach for the identification of writing traits on email database

Abstract

Trait is a particular characteristic that can produce a particular type of behavior. The email which is written communication medium among the people is the source, to identif1y the writing traits of the person. This paper proposes a novel approach to identify the writing traits of person from their email communication. The proposed technique is combination of an unsupervised k-means clustering algorithm and rule based system to classify the traits of email writer. We classified the email writers into three categories as perfect writer, average writer and casual writer. The experiment was carried out on Enron's email data set. This experiment is helpful to predict the writing behavior of a person which will be helpful in identifying an email writer or improve his/her writing style.