INTEGRATING NLP-BASED SENTIMENT ANALYSIS INTO SUPPLY CHAIN MANAGEMENT: A CASE STUDY OF INDIAN PRODUCTS ON AMAZON

Abstract

A critical component of modern business operations is Supply chain management (SCM). It involves the effective coordination of activities from the procurement of raw materials to the delivery of finished products to customers. With the rapid technology advancements, the integration of Machine Learning (ML) and Artificial Intelligence (AI) techniques has emerged as a powerful tool in enhancing supply chain management practices. One particular area that has gained significant attention in recent years is sentiment analysis using Natural Language Processing (NLP). Sentiment analysis (SA) or Opinion Mining (OM) is a special branch of NLP that focuses on extracting subjective information, sentiments and opinions from text data, such as customer reviews. By analyzing the sentiment expressed in these reviews, businesses can gain valuable insights into customer preferences, pain points and satisfaction levels. Applying sentiment analysis techniques to supply chain management allows organizations to better understand customer feedback and reviews, leading to informed decision-making and improved overall performance. The goal of this study is to look into how NLP-based sentiment analysis could be used to enhance supply chain management, specifically through utilizing consumer feedback for Indian items. The study's goal is to examine the use of sentiment analysis to learn more about how customers feel about various elements of the supply chain, including product availability, delivery time, packaging, and customer service. The study attempts to identify patterns, trends, and customer feelings by analyzing a dataset of over 1000 customer reviews gathered from various categories, including clothing, hair and skin care goods, and technological devices on the Amazon platform. The research's goal in conducting this case study is to draw attention to the useful uses of sentiment analysis in supply chain management. The study's results will help us comprehend sentiment analysis as a decision-making tool better and how it can be utilized to enhance different supply chain components. By giving businesses insights into client preferences and empowering them to make data-driven decisions to improve product offerings, inventory management, logistics optimisation, and overall customer pleasure, this research has the potential to be helpful to businesses. 9 We used a theoretical framework that is based on supply chain management, sentiment analysis, and NLP literature that has already been published. For the purpose of our study, we analyzed a corpus of research publications on supply chain management, sentiment analysis, and related subjects. Our study strategy included exploratory data analysis (EDA), the use of the VADER and RoBERTa models for sentiment analysis, and a qualitative analysis of the outcomes to pinpoint the most important conclusions and their ramifications. The results imply that NLP-based sentiment analysis can offer useful supply chain management insights. Our research of the Amazon dataset showed that sentiment analysis may pinpoint a product's advantages and disadvantages, point out potential areas for development, and reveal consumer preferences and expectations. Additionally, the study found that sentiment analysis can offer helpful data for a variety of supply chain management functions, such as product design, manufacturing, inventory control, and customer service. By illustrating the importance of sentiment analysis in enhancing decision-making in several supply chain management sectors, the research adds to the body of knowledge on NLP-based sentiment analysis and supply chain management. The paper also emphasizes the need for additional investigation to examine sentiment analysis's potential in other supply chain management domains, such as supplier management and logistics. The study reveals how sentiment analysis can help with decision-making and customer happiness, which has applications for supply chain managers. Additionally, our study offers a methodology that may be used in various contexts to undertake sentiment analysis in supply chain management. The study also emphasizes how critical it is to incorporate NLP-based sentiment analysis into supply chain management systems in order to monitor customer comments and reviews in real-time and create supply chains that are more responsive and centered on the needs of their customers. As a result, our research shows how NLP-based sentiment analysis may enhance supply chain management. The significance of sentiment analysis in determining consumer preferences and expectations, enhancing product design and quality, and improving inventory management is highlighted in our work as a contribution to the literature. Because it offers a methodology for doing 10 sentiment analysis and emphasizes the value of incorporating NLP-based sentiment analysis into supply chain management systems, our research also has applications for supply chain managers.