Optimal Feature Selection for Learning-Based Algorithms for Sentiment Classification
- 32 Downloads
Sentiment classification is an important branch of cognitive computation—thus the further studies of properties of sentiment analysis is important. Sentiment classification on text data has been an active topic for the last two decades and learning-based methods are very popular and widely used in various applications. For learning-based methods, a lot of enhanced technical strategies have been used to improve the performance of the methods. Feature selection is one of these strategies and it has been studied by many researchers. However, an existing unsolved difficult problem is the choice of a suitable number of features for obtaining the best sentiment classification performance of the learning-based methods. Therefore, we investigate the relationship between the number of features selected and the sentiment classification performance of the learning-based methods. A new method for the selection of a suitable number of features is proposed in which the Chi Square feature selection algorithm is employed and the features are selected using a preset score threshold. It is discovered that there is a relationship between the logarithm of the number of features selected and the sentiment classification performance of the learning-based method, and it is also found that this relationship is independent of the learning-based method involved. The new findings in this research indicate that it is always possible for researchers to select the appropriate number of features for learning-based methods to obtain the best sentiment classification performance. This can guide researchers to select the proper features for optimizing the performance of learning-based algorithms. (A preliminary version of this paper received a Best Paper Award at the International Conference on Extreme Learning Machines 2018.)
KeywordsMachine learning Feature selection Optimal feature selection Relationship analysis Sentiment classification Social media Text analysis
The authors would like to thank Dr. Ho Seng Beng, Dr. Quek Boon Kiat, and the team of A*STAR AI program for their discussion and help. The authors would also like to thank the intern students from NTU and SUTD for the assistance in this research.
Compliance with Ethical Standards
This article does not contain any studies with human participants or animals performed by any of the authors.
Conflict of Interest
The authors declare that they have no conflict of interest.
- 4.Dashtipour K, Gogate M, Adeel A, Ieracitano C, Hussain A. Exploiting deep learning for Persian sentiment analysis. Int Conf Brain Inspired Cognit Syst. 2018:597–604.Google Scholar
- 5.Cambria E, Hussain A, Durrani T, Havasi C, Eckl C, Munro J. Sentic computing for patient centered applications. Proc IEEE ICSP. 2010:1279–82.Google Scholar
- 10.Wang Z, Chong CS, Lan L, Yang Y, Ho S, Tong JC. Fine-grained sentiment analysis of social media with emotion sensing. Future Technol Conf. 2016:1361–4.Google Scholar
- 12.Cambria E, Poria S, Hazarika D, Kwok K. SenticNet 5: discovering conceptual primitives for sentiment analysis by means of context embeddings. In: The thirty-second AAAI conference on artificial intelligence (AAAI-18); 2018. p. 1795–802.Google Scholar
- 16.Kolchyna O, Souza TTP, Treleaven P, Aste T. Twitter sentiment analysis: lexicon method, machine learning method and their combination. arXiv preprint arXiv. 2015:32.Google Scholar
- 17.Zhang L, Ghosh R, Dekhil M, Hsu M, Liu B. Combining lexicon-based and learning-based methods for twitter sentiment analysis. Int J Electron Commun Soft Comput Sci Eng. 2015;89:1–8.Google Scholar
- 18.Cambria E, Olsher D, Kwok K. Sentic activation: a two-level affective common sense reasoning framework. Proc AAAI. 2012:186–92.Google Scholar
- 19.Cambria E, Mazzocco T, Hussain A, Eckl C. Sentic medoids: organizing affective common sense knowledge in a multi-dimensional vector space. LNCS. 2011;6677:601–10.Google Scholar
- 20.Wang Z, Tong JC, Ho SB. Method and system of intelligent sentiment and emotion sensing with adaptive learning. In: Patent cooperation treaty (PCT) international application no.PCT/SG2017/050172; 2017.Google Scholar
- 22.Narayanan V, Arora I, Bhatia A. Fast and accurate sentiment classification using an enhanced Naive Bayes model. Int Conf Intell Data Eng Automated Learn. 2013:194–201.Google Scholar
- 23.Wang Z, Tong JC, Chin HC. Enhancing machine-learning methods for sentiment classification of web data. Asia Inf Retr Symp. 2014;8870:394–405.Google Scholar
- 27.Cambria E, Hussain A, Havasi C, Eckl C. SenticSpace: visualizing opinions and sentiments in a multi-dimensional vector space. Knowl-Based Intell Inf Eng Syst. 2010:385–93.Google Scholar
- 28.Tang J, Alelyani S, Liu H. Feature selection for classification: a review. Data Classif Algorithms Appl. 2014:37.Google Scholar
- 32.Prusa JD, Khoshgoftaar TM, Dittman DJ. Impact of feature selection techniques for tweet sentiment classification. Twenty-Eighth Int Flairs Conf. 2015:299–304.Google Scholar
- 33.Nigam K, Lafferty J, Mccallum A. Using maximum entropy for text classification. In: IJCAI-99 workshop on machine learning for information filtering; 1999. p. 61–7.Google Scholar
- 34.Cortes C, Vapnik V. Support-vector networks. Mach Learn. 1995;20(3):273–97.Google Scholar
- 35.Huang G, Zhu Q, Siew C. “Extreme learning machine: a new learning scheme of feedforward neural networks,” in Neural Networks, 2004. Proceedings. 2004 IEEE International Joint Conference on, 2004, vol. 2, no. August 2004, pp. 985–990.Google Scholar
- 36.Li S, Xia R, Zong C, Huang C-R. “A framework of feature selection methods for text categorization,” in Proceedings of the Joint Conference of the 47th Annual Meeting of the ACL and the 4th International Joint Conference on Natural Language Processing of the AFNLP, 2009, no August, pp. 692–700.Google Scholar
- 38.“Twitter-sentiment-analyzer,” Available from: https://github.com/ravikiranj/twitter-sentiment-analyzer/tree/master/data [Cited 4 Sep. 2013].
- 39.“Twitter-sentiment-analysis2,” Available from: https://www.kaggle.com/c/twitter-sentiment-analysis2/data [Cited 2 Dec. 2017].
- 41.Gelman A, Goodrich B, Gabry J, Ali I. R-squared for Bayesian regression models. Am Stat. 2018:1–6.Google Scholar