Bangladeshi Indigenous Language Classification From Text Analysis

Abstract

Bangladesh is a country having cultural and historical linguistic richness and the current pattern of language reflects the sociological characteristic of the society. Of the numerous languages spoken in Bangladesh and around the world, the indigenous languages of the country occupy a special place here as they provide an insight into the distinct culture and identity of various ethnic groups. These languages themselves, which often have their roots in traditions as old as the prehistoric periods, contain details of otherwise traditions, myths and histories of local communities. The paper names “Bangladeshi Indigenous Language Classification from Text Analysis” addresses the task of classification of indigenous languages in Bangladesh, with the target attributes being ‘Dhaka’ ‘Chattogram’ and ‘Mymensingh’ This work uses an exciting dataset collected through Google Form and employs BNB, SVM, LR Random Forest, BiLSTM and CNN to classify the indigenous language A report of the study has shown that CNN is the 66%. This induces more importance and relevance of mature deep learning techniques for proper classification of native languages which are found in the linguistic complex of Bangladesh