Abstract:A framework for multi-class sentiment classification is proposed.A total of 3600 comparative experiments are conducted.Performances of different feature selection/machine learning algorithms are compared.The results are valuable for further studies on multi-class sentiment classification. Multi-class sentiment classification has extensive application backgrounds, whereas studies on this issue are still relatively scarce. In this paper, a framework for multi-class sentiment classification is proposed, which includes two parts: 1) selecting important features of texts using the feature selection algorithm, and 2) training multi-class sentiment classifier using the machine learning algorithm. Then, experiments are conducted for comparing the performances of four popular feature selection algorithms (document frequency, CHI statistics, information gain and gain ratio) and five popular machine learning algorithms (decision tree, nave Bayes, support vector machine, radial basis function neural network and K-nearest neighbor) in multi-class sentiment classification. The experiments are conducted on three public datasets which include twelve data subsets, and 10-fold cross validation is used to obtain the classification accuracy concerning each combination of feature selection algorithm, machine learning algorithm, feature set size and data subset. Based on the obtained 3600 classification accuracies (4 feature selection algorithms 5 machine learning algorithms 15 feature set sizes 12 data subsets), the average classification accuracy of each algorithm is calculated, and the Wilcoxon test is used to verify the existence of significant difference between different algorithms in multi-class sentiment classification. The results show that, in terms of classification accuracy, gain ratio performs best among the four feature selection algorithms and support vector machine performs best among the five machine learning algorithms. In terms of execution time, the similar comparisons are also conducted. The obtained results would be valuable for further improving the existing multi-class sentiment classifiers and developing new multi-class sentiment classifiers.

Sentiment Analysis on News Comments Based on Supervised Learning Method

Sentiment Classification for Chinese News Using Machine Learning Methods

Chinese sentiment analysis for commodity price level fluctuation news comments

Empirical Study of Sentiment Classification for Chinese Microblog Based on Machine Learning

Chinese Micro-Blog Sentiment Classification Through a Novel Hybrid Learning Model

Experimental Study on Sentiment Classification of Chinese Review Using Machine Learning Techniques

Topic Oriented Sentimental Feature Selection Method for News Comments

Sentiment Analysis of Online Chinese Comments Based on Statistical Learning Combining with Pattern Matching.

Feature Selection For Chinese Online Reviews Sentiment Classification

Chinese Sentiment Classification Using A Neural Network Tool-Word2vec

A Multi-dimension and Multi-granularity Feature Fusion Method for Chinese Microblog Sentiment Classification

Sentiment Classification Method of Hierarchical Retrieval Based on Comment Data

Multi-class sentiment classification: The experimental comparisons of feature selection and machine learning algorithms.

Comparison of Text Sentiment Analysis Based on Machine Learning.

Research on sentiment classification of Uighur reviews

Combining Convolutional Neural Networks and Word Sentiment Sequence Features for Chinese Text Sentiment Classification

Sentiment Analysis of Comment Texts Based on BiLSTM.

Research on Feature Analysis Oriented Text Sentiment Identification

Sentiment Classification in Chinese Microblogs: Lexicon-based and Learning-based Approaches

Unsupervised Sentiment-Bearing Feature Selection for Document-Level Sentiment Classification

Sentiment analysis of Chinese Micro-blog based on semi-supervised learning