Machine Learning Approach Based on Sentiment Analysis for Effective Managerial Decision on Product Quality Management and Customer Satisfaction

  • Saumya Chaturvedi, Vimal Mishra, Nitin Mishra

Abstract

 This paper narrates the application of Machine Learning Approach based on Sentiment Analysis for product quality management and enhanced business process decision.  It is based on customer opinion analysis which is built over customer review posted on the internet in natural language. We have used R studio to implement machine learning algorithms for sentiment analysis. We have contributed two algorithms for natural language processing. We have also proposed  a framework   for sentiment  analysis. Machine Learning, web data extraction  and natural language processing are used for the understanding of customer review analysis. In this paper, specific  internet  resources such as imdb.com,  amazon.com, and yelp.com are used for customer review sentiment classification. The attributes are text sentences extracted from reviews of products, movies, and restaurants. The review analysis is utilized in making a managerial decision  on product quality in accordance  with customer satisfaction.  The satisfaction is measured over two scales such as sentiment  towards the product and aspect based sentiment.  We have evaluated the efficiency of machine learning methods for sentiment classification and analysis. An intelligent decision system is proposed for quality management  based on customer sentiment indicating satisfaction. The proposed approach is seen as a prototype of the decision system which will help in taking an effective managerial decision on product quality management according to customer feedback.

Published
2020-06-01
How to Cite
Saumya Chaturvedi, Vimal Mishra, Nitin Mishra. (2020). Machine Learning Approach Based on Sentiment Analysis for Effective Managerial Decision on Product Quality Management and Customer Satisfaction. International Journal of Advanced Science and Technology, 29(7s), 5094 - 5102. Retrieved from https://sersc.org/journals/index.php/IJAST/article/view/25795