Tools
Tools
-, Pawan Kumar (2024) Sentiment Analysis Techniques for Restaurant Reviews Across Multiple Attributes. Masters thesis, Dublin, National College of Ireland.
Preview |
PDF (Master of Science)
Download (1MB) | Preview |
Preview |
PDF (Configuration Manual)
Download (1MB) | Preview |
Abstract
The spread of the Internet and social media have made the world more connected and nearer, where people share their opinions, ideas, and pictures with the public. Exploiting this information has become the most popular for companies around the world, and sentiment analysis is one technique out of it. The online restaurant business has made sentiment analysis a critical tool for understanding consumer opinions and improving decision-making. The study focuses on implementing sentiment analysis on the review data to extract the opinion of customers as positive or negative using a standard methodology, KDT (Knowledge Discovery in Text). Various machine learning models like Logistic Regression, Xg Boost, Random Forest and Naïve Bayes and deep learning models (LSTM) are used in research focusing on extracting useful information from text data using Natural Language Understanding (NLU) and NLP ( Natural Language Processing). To convert the text data into numbers, metrics like TF-IDF, Count Vectorizer and Word2vec are used where the LSTM model with Count Vectorizer attained the maximum accuracy of 95.4%. The research has showcased a comprehensive analysis of various techniques tailored for the restaurant industry, offering valuable information for academic as well as business applications.
Actions (login required)
 |
View Item |