Track: Data Analytics and Big Data

Abstract

Reviews and tweets posted on social media and websites offer timely opinion and feedback for companies to learn about customers’ concerns. As a seller or marketer, it is important to understand what customer thinks and feel about products or services offered by the companies. This study aims to explore the sentiments of public opinion and understand the reason behind it through sentiment analysis and topic modelling approach. The first objective of this study is to analyse customer perceptions and sentiments towards multilevel marketing (MLM) company’s product. Then, the second objective is to identify the primary topics of concern regarding the products shared on social media. To achieve these objectives, customer reviews were collected from multiple platforms such as social media namely Twitter, online web review i.e., Moutshut.com, Trustpilot, sitejabber.com, and e-commerce websites such as Amazon, Lazada and Shopee. Other than that, a survey was also conducted to triangulate the data and to get first-hand feedback and review from the customers. In this study, the reviews were analysed using sentiment analysis and topic modelling techniques: Valance Aware Dictionary and Sentiment Reasoner (VADER), Natural Language Toolkit (NLTK) packages, Latent Dirichlet Allocation (LDA). The findings of the study revealed that over 50% of the reviews were positive and customers were mostly shared about the product quality such as the originality, delivery and product packaging in their reviews. This study offers some insight about public perceptions towards the products from MLM company and at the same time helps in improving the products and services globally and locally.