ItemConference Paper

SentimentViz: Leveraging RoBERTa in Python for Advanced Sentiment Analysis and Decision-Making for a Famous Indian FMCG (Ayurvedic) Brand

Title
SentimentViz: Leveraging RoBERTa in Python for Advanced Sentiment Analysis and Decision-Making for a Famous Indian FMCG (Ayurvedic) Brand
Creator
Kundu, Seeboli Ghosh; Kundu, Avisek; Sahu, Santosh Kumar; Kalsi, Suchi; Siddhapura, Akshita; Badgayan, Nitesh Dhar
Description
Indian consumer preferences for Ayurvedic brands increasingly turn to the marketplace for well-being. Ayurveda has a deep-rooted history in emerging economies like India, and its increasing role in health, wellness, and exports contributes to Indias economic development. The consumption and changing lifestyle patterns significantly contribute to achieving the United Nations Sustainable Development Goals (SDGs). The primary objective of this study is to explore consumer sentiment that includes perceptions, feelings, and attitudes toward these natural healthcare products contributing to specific SDG targets, leading to good health and well-being. In a data-driven world where governments and businesses seek insights from vast amounts of unstructured text data, sentiment analysis plays a pivotal role in decision-making. Sentiment analysis helps analyze different aspects of unstructured data, including customer experience and insights generated in terms of usage, challenges, and preferences and ultimately helps manage customer engagement. The sentiment analysis requires understanding the context, grouping similar words, removing unrelated content, and then gauging the sentiment of the text. There has always been a challenge to contextualize and gauze the deeper sentiments and create scalable solutions. To build on this need for deeper sentiment understanding and scalable solutions, SentimentViz is a proposed accelerator as part of this paper that leverages Python and chooses the best methodology for text-mining problems. It enables real-time analysis with robust visualization capabilities: In this study, the SentimentViz accelerator is leveraged to estimate the sentiment of 9 products using a robust data science framework and best-of-the-class ML techniques. The detailed consumer sentiment analysis helped to develop a deeper understanding of the value of FMCG (Ayurvedic products) for an emerging economy like India. This will help marketers build targeted marketing campaigns, brand health monitoring, and customer retention strategies through informed decisions. The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.
Source
Smart Innovation, Systems and Technologies;Volume;454 SIST;pp.517-529
Date
01-01-2026
Publisher
Springer Science and Business Media Deutschland GmbH
Subject
Decision-making; Marketing analytics; Natural language processing; Real-time analysis; Sentiment analysis
Coverage
Kundu S.G., Symbiosis Centre for Management Studies, Symbiosis International (Deemed University), Bengaluru Campus, Pune, India; Kundu A., School of Business Management, Christ University, Bengaluru, India; Sahu S.K., School of Mechanical Engineering, VIT-AP University, Besides A.P. Secretariat, Andhra Pradesh, Amaravati, India; Kalsi S., Symbiosis Centre for Management Studies, Symbiosis International (Deemed University), Bengaluru Campus, Pune, India; Siddhapura A., Symbiosis Centre for Management Studies, Symbiosis International (Deemed University), Bengaluru Campus, Pune, India; Badgayan N.D., KPMG, Mumbai, India
Rights
Restricted Access; Hardcopy may be available in the library
Relation
ISSN: 21903018; ISBN: 978-303207836-0;
Format
online
Language
English
Type
Conference paper

Citation

Kundu, Seeboli Ghosh; Kundu, Avisek; Sahu, Santosh Kumar; Kalsi, Suchi; Siddhapura, Akshita; Badgayan, Nitesh Dhar, “SentimentViz: Leveraging RoBERTa in Python for Advanced Sentiment Analysis and Decision-Making for a Famous Indian FMCG (Ayurvedic) Brand,” CHRIST (Deemed To Be University) Institutional Repository, accessed June 19, 2026, https://archives.christuniversity.in/items/show/25370.

Output Formats