RESTAURANT REVIEWS: ASPECT-BASED SENTIMENT ANALYSIS USING DEEP LEARNING AND NLP ALGORITHM

Authors

  • Vishnu Suryawanshi*, Deepali Ujalambkar, Dhammjyoti Dhawase, Yuvraj Nikam & Aman Kamble

Keywords:

Convolutional neural networks (CNN), sentiment analysis, machine learning, long short-term memory (LSTM), and bidirectional LSTM.

Abstract

Natural Language Processing is a component of artificial intelligence and machine learning that teaches computers how to interact with human (natural) languages. It is crucial to examine these evaluations to gain user feedback and raise consumer satisfaction. Restaurant services may be improved utilizing customer feedback if user thoughts or sentiments are determined from the user reviews. The preprocessing methods used for the review in the proposed study include lemmatization, tokenizing, eliminating numbers and punctuation, stopping words, and making all words lowercase. Then, for generating the final analysis, use frequency-inverse document frequency (TF-IDF) and GRID search. There were 4500 reviews included in the data that we processed. In our proposed study, we conducted a split test with 80% training data and 20% testing data. The work focuses on opinions that are both generic and those that depend on a variety of factors, including cuisine, service, atmosphere, quality, and price. The words "terrible," "good," "average," and "best," as well as "place," "love," "order," "food," and "try," as well as "staff," "menu," and "waiters," as well as "price," "dish," and "beautiful," all influence the final score. The experimental result of 75.65% was attained using the suggested algorithm.  

Published

2022-09-14

How to Cite

Vishnu Suryawanshi*, Deepali Ujalambkar, Dhammjyoti Dhawase, Yuvraj Nikam & Aman Kamble. (2022). RESTAURANT REVIEWS: ASPECT-BASED SENTIMENT ANALYSIS USING DEEP LEARNING AND NLP ALGORITHM. Harbin Gongye Daxue Xuebao/Journal of Harbin Institute of Technology, 54(9), 207–216. Retrieved from http://hebgydxxb.periodicales.com/index.php/JHIT/article/view/1371

Issue

Section

Articles