Sentiment Analysis Using Part-of-Speech-Based Feature Extraction and Game-Theoretic Rough Sets
Author | : Yixing Chen |
Publisher | : |
Total Pages | : 0 |
Release | : 2022 |
Genre | : |
ISBN | : |
Download Sentiment Analysis Using Part-of-Speech-Based Feature Extraction and Game-Theoretic Rough Sets Book in PDF, Epub and Kindle
Sentiment analysis, one of the trending natural language processing tasks, is used to mine opinions or sentiments from a given text. There are two significant challenges in sentiment analysis. The first challenge is the complexity in data pre-processing caused by the high dimensionality of textual data. The second is the uncertainty in classifying sentiment polarities due to the ambiguity of natural languages. Existing research may lack an efficient and straightforward solution to resolve the first issue; or discuss the trade-off between accuracy and coverage regarding uncertain data. To address these issues, we propose a model using part-of-speech-based feature extraction to reduce dimensionality and game-theoretic rough sets (GTRS) to analyze the accuracy and coverage trade-off. We evaluate this model with three different datasets, Yelp reviews, IMDB movie reviews, and Amazon product reviews. The experiment results show that the proposed model outperforms Pawlak's rough set model and 0.5-probabilistic rough set model. In comparison with the sentiment analysis tool Valence Aware Dictionary for Sentiment Reasoning (VADER) and four traditional binary classification models (i.e., SVM, na ̈ıve Bayes, decision tree, and KNN), the proposed model also achieves higher accuracy. This research suggests that the proposed model has achieved higher results of both accuracy and coverage, and is promising to deal with the complexity and uncertainty in sentiment analysis tasks.