Tweet Sentiment Classification by Semantic and Frequency Base Features Using Hybrid Classifier

  • Hemant Kumar MenariaEmail author
  • Pritesh Nagar
  • Mayank Patel
Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 1045)


The technique of sentiment analysis is considered as the most powerful tools in the field of natural language processing as it comes up with large number of possibilities to perceive the sentiments of people’s on several distinct topics. The concept of aspect-based sentiment analysis aims to figure out it further and determines what an individual is talking about, and explains whether she/he likes it or not. A practical example of an ideal realm in context to this topic discussed represents millions of possible schemes of Indian welfare planning. The government has launched such type at all the possible levels in schools, center and state level. These schemes work with an association of both the state and the central government. Such welfare-based schemes are generally introduced for various distinct levels on the basis of peoples (individuals) and their behavior or lifestyle. The schemes are launched for the purpose of developing the minority and the weaker section of the society. Whereas some of the welfare schemes are mainly introduced for girls and women only and it helps in empowering the status of the women by providing financial help as well as the basic need and requirements. There are several distinct ways to handle this major issue by the mechanism of machine learning process. In this paper, a labeled form of data is mainly used based on the polarity, preprocessing of the Tweets, which further extracts unigram features after the process of Tweet-based preprocessing methodology. In case of preprocessing, the data with huge noise is removed with the help of tokenization process; stop word removal process and stemming (deriving) these processes to clear the redundant data such as repeat emoji, words, and hashtags. These label and features are usually learned by SVM, KNN and a hybrid of KNN. In proposed experiment Hybrid approach shows improvement in precision and accuracy than other.


Machine learning Support vector machine Twitter sentiment analysis 


  1. 1.
    Krishna, R.: Fuzzy aspects in sentiment analysis and, pp. 7750–7755 (2016)Google Scholar
  2. 2.
    Shidaganti, G., Hulkund, R.G., Prakash, S.: Analysis and exploitation of twitter data using machine learning techniques. In: International Proceedings on Advances in Soft Computing, Intelligent Systems and Applications, pp. 135–146. Springer, Singapore (2018)Google Scholar
  3. 3.
    Rout, J.K., Choo, K.K.R., Dash, A.K., Bakshi, S., Jena, S.K., Williams, K.L.: A model for sentiment and emotion analysis of unstructured social media text. Electron. Commer. Res. 18(1), 181–199 (2018)CrossRefGoogle Scholar
  4. 4.
    Mumtaz, D., Ahuja, B.: A Lexical and Machine Learning-Based Hybrid System for Sentiment Analysis, vol. 713. Springer, Singapore (2018)Google Scholar
  5. 5.
    Al-Smadi, M., Al-Ayyoub, M., Al-Sarhan, H., Jararwell, Y.: An aspect-based sentiment analysis approach to evaluating Arabic news affect on readers. J. Univers. Comput. Sci. 22(5), 630–649 (2016)MathSciNetGoogle Scholar
  6. 6.
    Chen, P.T., Sessions, T.: PacificVAST Program PacificVis Program (2018)Google Scholar
  7. 7.
    Fouad, M.M., Gharib, T.F., Mashat, A.S.: The International Conference on Advanced Machine Learning Technologies and Applications (AMLTA2018), vol. 723, no. January (2018)Google Scholar
  8. 8.
    Bifet, D.S., Khan, F.H., View, S.A.: TOM : twitter opinion mining framework using hybrid classification scheme, no. January (2014)Google Scholar
  9. 9.
    Zahrotun, L.: Comparison jaccard similarity, cosine similarity and combined both of the data clustering with shared nearest neighbor method. Comput. Eng. Appl. 5(11), 2252–4274 (2016)Google Scholar
  10. 10.
    Agarwal, N., Rawat, M., Vijay, M.: Comparative analysis of jaccard coefficient and cosine similarity for web document similarity measure. Int. J. Adv. Res. Eng. Technol. 2(5), 18–21 (2014)Google Scholar
  11. 11.
    Medhat, W., Hassan, A., Korashy, H.: Sentiment analysis algorithms and applications : a survey. Ain Shams Eng. J. 5(4), 1093–1113 (2014)CrossRefGoogle Scholar
  12. 12.
    Virmani, D., Malhotra, V., Tyagi, R.: Sentiment analysis using collaborated opinion mining. arXiv Prepr. arXiv1401.2618, no. January (2014)Google Scholar
  13. 13.
    Balachandran, L., Kirupananda, A.: Online reviews evaluation system for higher education institution: an aspect based sentiment analysis tool. In: 2017 11th International Conference on Software, Knowledge, Information Management and Applications, pp. 1–7 (2017)Google Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2020

Authors and Affiliations

  • Hemant Kumar Menaria
    • 1
    Email author
  • Pritesh Nagar
    • 1
  • Mayank Patel
    • 1
  1. 1.Computer Science and EngineeringGeetanajli Institute of Technical StudiesUdaipurIndia

Personalised recommendations