IJSR - International Journal of Scientific Research

Volume : VII, Issue : IV, April - 2018

Shree Vaishnavi. R, Deepika. R. C, Amritha. S

Text classication is a process of classifying documents into predened categories through different classiers learned from labelled and

unlabelled training samples. Many researchers who work on binary text classication attempt to a more effective way to separate relevant texts

from a large data set. Here we proposes a three–way decision model for dealing with the uncertain boundary to improve the binary text classication

performance based on the rough set technique and centroid solutions. Four discussion rules are proposed from the training process and applied to

the incoming document for more precise classication. However, the current text classiers cannot unambiguously describe the Decision boundary

between the positive and negative objects because of uncertainties caused by text features selection and the knowledge learning process . The

experimental results show that the usage of boundary vectors is very effective and efcient for dealing with uncertainties for the decision

boundaries, and the proposed model has signicantly improved the performance of binary text classication in term of F1 measure and AUC area

compared with six other popular