Telugu News Article Classification Using Deep Learning Models
Mora Sai Nitish Reddy, Sulakunta Aravind, Haripavan Reddy Bojjam · 2023
In this day and age of information, numerous Telugu Language documents have been converted into digital format and are now available. It would be easier to retrieve these electronic data records if these papers were organized into a class according to the information contained within them. Day by day the amount of digital data increases, so text classification has emerged as possibly the most important challenge facing information systems that are concerned with text records. In this Project Methods for classifying text applied to unstructured Telugu language to get order useful information and insights from the text which is not in a structured fashion. There hasn't been a lot of work done on Telugu Language because Indian languages are notoriously difficult to classify the language has such a sophisticated morphological system that it calls for specialized algorithms to undertake morphological analysis. Raw data are put through a series of preprocessing steps that are tailored to the Telugu language to generate a vector, where it contains both good and limited no of features. For the construction of an accurate classification model for Telugu text documents, a significant amount of pre-processing is necessary. In this study, we implemented various deep learning models and evaluated how well they performed on Telugu text. Models that we implemented are MLP, CNN, LSTM, RNN, and BILSTM.