Handling Categorical Data of Reviews from Movie Websites

  • Sonu Kumar Gupta, Santhosh Kumar K. L.


Evolution in the information’s technology makes the production or generations of the data and information increase dramatically. We can get large amount of data from the various sources like internet, including reviews about a movies, product and services. Large amount of data obtained from the various sources needed heavy system to process it. Sentiment analysis method or technique is a text processing of Natural Language Processing (NLP) that can be used to see the quality of the product, services, movies reviews, twitter sentiments and hotel reviews. This paper uses the Movies reviews data to perform the sentiment analysis obtained from the one of the famous movie website i.e. IMDB website. To get better result we worked or use word embedding to convert or transform word into vectors form and classify using different classifier algorithms. This study aims to find the comparison of the word embedding performance, while word embedding compared is Bag of Words and TF-IDF. From the experiment conducted, we conclude that the TF-IDF technique gives the highest accuracy as compare to Bag of Words and TF-IDF word embedding is the good choice for the reviews data.

How to Cite
Sonu Kumar Gupta, Santhosh Kumar K. L. (2020). Handling Categorical Data of Reviews from Movie Websites. International Journal of Advanced Science and Technology, 29(3), 13964 - 13969. Retrieved from https://sersc.org/journals/index.php/IJAST/article/view/31752