The main intention of this paper is to determine the sentimental analysis for movies prediction using machine learning techniques. A novel methodology along with necessary algorithms is presented here. The input dataset obtained from IMDb is chosen for prediction of box office collection. Firstly preprocessing is performed to remove the redundant features such as stop words, punctuations, numbers, whitespace, lowercase conversion, word stemming, sparse term removal etc. Preprocessing technique involves two steps namely tokenization and stop word removal. The preprocessed text is sent for transformation process. This formation involves calculating the values of TF and IDF. These are performed for estimating the sentiment analysis. Afterwards, fuzzy clustering is performed that helps in getting both positive and negative outcomes on basis of different movies datasets. In this way, a classification model is made utilizing SVM classifier for foreseeing the pattern of the box office incomes from the reviews of movies.
CITATION STYLE
Jadon, M., Sharma, I., & Sharma, A. K. (2020). Sentiment Analysis for Movies Prediction Using Machine Leaning Techniques. In Lecture Notes on Data Engineering and Communications Technologies (Vol. 38, pp. 457–465). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-030-34080-3_52
Mendeley helps you to discover research relevant for your work.