Sentiment Analysis of Netflix Multi-Genre Using Support Vector Machine

Abstract

The main objective of this research is to understand the audience's sentiment regarding various Netflix web series and their corresponding genres. This study, which contributes to sentiment analysis, collects data from the eight official Twitter handles of actors/actresses and web series from five genres: comedy, drama, Sci-fi, romance, and action. The data collection process involved extracting tweets related to these web series and classifying them into positive, negative, and neutral categories. Classification is done by using a Support Vector Machine (SVM), a popular classifier in machine learning. In addition to sentiment analysis, the study also involved measuring activity metrics and popularity metrics to assess the level of interaction between the audience and the actors/actresses. This data was then combined with the classified tweets for further analysis. The research results indicated that the Sci-fi genre is the prime choice for the audience. The conclusion was drawn based on the collective sentiment expressed in tweets and the level of engagement and popularity observed in interactions with the actors and actresses associated with action-oriented web series. Overall, this research, which is a significant contribution to the field, sheds light on the preferences of Netflix viewers with respect to different genres and highlights the significance of the Sci-fi genre in capturing the audience's attention and interest. The findings could be valuable for content creators and streaming platforms in tailoring their offerings to match the viewers' preferences and enhance user satisfaction.