Classification of Food Review Sentiment in Bangla Language Using NLP, Machine Learning and LSTM

Abstract

In this era of Internet technology in Bangladesh the online food supply industry was still flourishing. Most individuals in Bangladesh's city region may easily order their food from several food suppliers during the epidemic for Covid-19. New consumer comfort doors are opened by a growing trend in food supply. People try to make meal choices based on ratings and feedback. The quality of meals is not only ranking. Rating is a mix of menu, delivery surveillance program and delivery attitude. Users must thus read every food remark. However, this procedure takes time. Because every remark on different meals is tough to read. We sought to create an intelligent meal evaluation system for this purpose. For various online meal delivery applications, we have gathered around 840 Bangla phrases. In our study we have utilized sentiment analyses and some classification algorithms such as KNN, Decision Tree, Support Vector Machine (SVM), and deep learning based LSTM algorithm. We utilized an algorithm to forecast which produced the highest precision and F1 score.