Sentiment Analysis : Using Different Models for Monitoring and Analyzing Customer Reviews

Purpose : Social media platforms offer a variety of useful information to people, resulting in the generation of enormous amounts of data, including images, videos, music, text, etc. Many methods have been developed to extract insights from this data. Accurate polarity identification of customer reviews remains a persistent and fascinating issue. This study explored the efficacy of various sentiment analysis models in monitoring and analyzing customer reviews, focusing on their ability to provide actionable insights for businesses. Design/Methodology/Approach : The research employed a comparative analysis of different sentiment analysis techniques. Sentiment analysis uses lexicon-based approaches, machine learning, and deep learning models for evaluating data. A dataset of customer reviews collected via Google Forms was utilized to test these models. The study involved data pre-processing and evaluation to determine the accuracy and precision of the customer reviews. Findings : The study revealed how sentiment analysis can shed light on consumer sentiment and reveal shifting tendencies. This can enable marketers to make better decisions and improve their marketing, product development, and customer service efforts. One important finding of the study is the usage of SA for deriving real-time actionable insights for companies. Practical Implications : The findings suggested that businesses can enhance their customer experience management by adopting sentiment analysis, particularly in scenarios with diverse datasets. Originality/Value : This research contributed to the growing field of sentiment analysis, offering valuable insights for both academic researchers and industry practitioners and aimed at improving customer sentiment monitoring practices.