TEXT CLASSIFICATION IN MARATHI LANGUAGE

Text Classification In Marathi Language

Mrs. Anudnya Suyog Sawant,

Computer Department

Pillai college of engineering, New panvel

Mumbai University

sanudnya23mtechcomp@student.mes.ac.in

Dr.Sharvari Govilker

Computer Department

Pillai college of engineering, New panvel

Mumbai University

sgovilkar@mes.ac.in

Abstract— In today’s world, several digitized multiple language text documents are generated daily at the Government sites, news portals, and public and private sectors, which are required to be classified effectively into various mutually exclusive pre-defined categories. Many multiple language text-based processing systems exist in application domains of information retrieval, machine translation, text summarization, simplification, keyword extraction, and other related parsing and linguistic perspectives, but still, there is a wide scope to classify the extracted text of Marathi documents into predefined categories using a classifier. Text classification works for 10 Indian languages and Arabic language are reported in the literature. A Marathi Text Classification model is proposed, which accepts a set of known Marathi documents, preprocesses them at document, sentence and word levels, extracts features, and trains different classifiers, which further classifies a set of Marathi unknown documents. This research aims to understand and reduce the language challenges of Marathi. We want to make it easier to use text classification (sorting and organizing text) in Marathi by improving the methods specifically designed for this language. The goal is to advance and enhance the ways we categorize and analyze Marathi text more effectively.

Keywords— Natural Language Processing, SVM, KNN, Naïve Bayes (Supervised learning methods)