Performance Modeling of Classification Techniques on Movie Sentiments
Pooja Rani1, Jaswinder Singh2
1Pooja Rani, M. Tech Scholar, Department of Computer Science & Engineering, Guru Jambheshwar University of Science & Technology, Hisar, India.
2Jaswinder Singh, Assistant Professor Department of Computer Science & Engineering, Guru Jambheshwar University of Science & Technology, Hisar, India.
Manuscript received on November 24, 2019. | Revised Manuscript received on December 08, 2019. | Manuscript published on December 30, 2019. | PP: 1125-1131 | Volume-9 Issue-2, December, 2019. | Retrieval Number: B2805129219/2020©BEIESP | DOI: 10.35940/ijeat.B2805.129219
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: The sentiment-based social media represents a goldmine approach for analyzing the performance of the products, hotels, movies, politics, etc. Large opinions of the people are found over movie comments that are honest, informative, and casual as compared to the formal type of data-survey modeling using magazines or reports. The work proposed is based on the rating of movies. This paper analyzes the performance of classifiers for the prediction of sentiment class i.e., positive and negative by using artificial neural network, k-nearest neighbor and hybrid approach. The success of these classification techniques depends mainly on the appropriate extraction of the set of characteristics used to detect sentiments. Hybrid of two or more classifiers is mainly used to enhance the results. In the proposed experiment Hybrid of ANN and KNN shows improvement in precision and accuracy than other classifiers.
Keywords: Sentiment Analysis, Artificial Neural Network, Nearest Neighbor, Hybrid, Sentiments.