A Machine Learning Approach to Classify Singular and Plural Numbers of Bengali Words

Abstract

This paper discusses a method for detecting the singular and plural number of Bengali sentences using context-sensitive grammar rules and NLP that accepts almost all types of Bangla words. The study is likely to result in the development of a system to digitize Bangladesh's vegetable classification system, allowing future generations to learn more about it. The purpose of this project is to classify the singular and plural number from Bengali words using a machine learning classification method. When the noun refers to a single item, the singular number is used, while the plural number is used when the word refers to multiple items. We have now created a model that can recognize singular and plural numbers in Bengali words using a machine learning classification technique. We encountered challenges when collecting Bangla data due to limited availability, but we eventually collected a total dataset of around 1000 words for this project. On 1000 data points from various groups, we achieved a maximum accuracy of 90.6%.