Sentiment Analysis of Live Stream Comments using Machine Learning Algorithms

In today’s world there is always some confusion when it comes to claiming something. People often spend a lot of time deciding on the best points in their daily activities, such as "what bistro to eat" and "what movie to watch". Previously, the main source of knowledge was interviews with well-known colleagues, i.e. a rumor. With the advent of the Internet, it has often been replaced by online reviews. These reviews reflect the general consensus about the products, people or events involved. Sentiment evaluation is a subject of computational linguistics and natural language processing that specializes in emotion. Machine learning monitoring algorithms for sentiment classification were able to yield more effective results. The proposed model uses the Naive Bayes classification technique for sentiment analysis, which can differentiate negative and positive reviews. Myntra has launched India’s first scaled platform that combines live streaming with real-time selling (2-way ecommerce) — a digital shopping experience where real people with real stories sell real products. On Myntra, live-shopping influencers sell brand products in real-time. Customers then communicate with one another by asking questions and providing comments. They purchase the items using the video’s embedded link and then proceed to the checkout. They frequently receive promos that can run as long as the stream. With this study, we’ll try to determine the influencer’s engagement score based on the consumers’ comments and reactions on the live stream.