EVOLUTION OF SENTIMENT ANALYSIS IN REVIEWS FOR INTELLIGENT PRODUCT RECOMMENDATIONS

Shoukat Ullah; Dr. Aurangzeb Khan; Dr. Khairullah Khan; Dr. Aman Ullah

doi:10.71146/kjmr224

Authors

Shoukat Ullah University of Science & Technology, Bannu Author
Dr. Aurangzeb Khan University of Science & Technology, Bannu, Pakistan Author
Dr. Khairullah Khan University of Science & Technology, Bannu, Pakistan Author
Dr. Aman Ullah Department of Commerce Education and Management Sciences, Higher Education, Archives and Libraries Department, Khyber Pakhtunkhwa, Pakistan Author

DOI:

https://doi.org/10.71146/kjmr224

Keywords:

Sentiment Analysis, Machine Learning, Appraisal Theory

Abstract

Consumer sentiment analysis (SA) has emerged as a popular research trend for social media applications, encompassing diverse domains like product reviews, healthcare, crime, finance, travel, and academics. Unraveling consumer perceptions and opinions from online reviews is paramount in gaining valuable insights. However, the increasing volume, subjectivity, and heterogeneity of social web data pose challenges for manual processing. To address this, researchers have turned to machine learning (ML) techniques, which offer promising solutions for real-life applications. This paper conducts a systematic literature review to assess ML techniques' efficacy, scope, and applicability in product reviews and analyze customers' attitudes from unstructured reviews specifically related to the product's quality, complexity, innovation, and impact. This research study also explores the main concepts and relevant literature of appraisal theory to compute people's attitudes because appraisal theory provides a more detailed description of the text than traditional SA.