Evolutionary and Incremental Text Document Classifier using Deep Learning

Abstract

Currently most of the generated data are stored in text format so text mining is believed to have a high commercial potential value. There are many sources for knowledge extraction, still unstructured text considered as the largest readily available source of knowledge. These data which is mostly available in digital format need to manage. Text document classification is basically a task to automatically categorize the text documents in some pre-defined classes.

A lot of research work is already done in this domain and many classifiers have been developed including some ensemble classifiers. The classification accuracy of the classifier depends on many parameters such as preprocessing, features selection, data set, methodology etc. Any of these mentioned parameters can be significantly influential in increasing the accuracy of the classifier