A Heuristic Approach for Text Classification with Ontology: A Review

In day to day life, mass information is readily available. So, the motive and main concern of automatic text classifiers is to segregate the relevant documents and irrelevant documents. The automated text classifier organizes the clustered data automatically. In this paper, a method for automatically classifying text has been implemented, by pre-processing the data in hand and selecting the necessary features and then classifying the data with ontology. Here, two techniques for pre-processing have been used that are stop words removal and stemming. In this paper, when the words are stemmed, it's necessary to remove the noise from the data, which has been implemented here using feature selection technique known as univariate selection and using tools named as PyDotPlus and Graphviz, followed by ontology-based classification of documents which can be implemented in future.