Advertisement

Opinion Mining of Bengali Review Written with English Character Using Machine Learning Approaches

  • Sabiha Sunjida Ahmed
  • Sharmin Akter Milu
  • Md. Ismail Siddiqi Emon
  • Sheikh Shahparan MahtabEmail author
  • Md. Fahad Mojumder
  • Md. Israq Azız
  • Jamal Ahmed Bhuiyan
  • M. J. Alam
Chapter
  • 32 Downloads
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 637)

Abstract

In this paper, we have done sentiment analysis for English written Bengali words given in different online shops in Bangladesh. For this work, we have chosen four latest mobile phones popular in Bangladesh. Here, the user reviews were in Bengali words written by English characters. The data was taken from online shopping sites from Bangladesh. Here, we have assumed six different features of mobiles written in the Result section. The main objective of the study was to find out the sentiment of Bengali words written with English alphabets. As it is a trend to write such reviews in Bangladesh, the data was taken and preprocessed to fit in algorithm, and they were compared whether it is positive or negative. Python was used as simulation tool, and Pursehub was used to extract the data set, and the system successfully finds out the positivity and negativity of the reviews. This result was achieved by using confusion matrix and that is making the overall performance of those mobile handsets. Out of 1201 reviews, 599 were found to be negative and 826 were found to be positive. The F1 score was 85.25%, accuracy was achieved 85.31%, and recall rate was 84.95%.

Keywords

Sentiment analysis Machine learning approaches Natural language processing Bengali Naïve Python 

References

  1. 1.
    Mohammad, S.M.: Sentiment analysis: detecting valence, emotions, and other affectual states from text. In: Emotion Measurement, pp. 201–237. Elsevier (2016)Google Scholar
  2. 2.
    Silva, J.J.D., Haddela, P.S.: A term weighting method for identifying emotions from text content. In: 2013 8th IEEE International Conference on Industrial and Information Systems (ICIIS), pp. 381–386. IEEE (2013)Google Scholar
  3. 3.
    Mehra, R., Bedi, M.K., Singh, G., Arora, R., Bala, T., Saxena, S.: Sentimental analysis using fuzzy and naive bayes. In: 2017 International Conference on Computing Methodologies and Communication (ICCMC), pp. 945–950. IEEE (2017)Google Scholar
  4. 4.
    Pang, K.B., Lee, L., Vaithyanathan, S.: Thumbs up? Sentiment classification using machine learning techniques. In: Proceedings of the ACL-02 Conference on Empirical Methods in Natural Language Processing, Vol. 10, pp. 79–86. Association for Computational Linguistics (2002) unpublishedGoogle Scholar
  5. 5.
    Chowdhury, R.S., Chowdhury, W.: Performing sentiment analysis in Bangla microblog posts. In: 2014 International Conference on Informatics, Electronics & Vision (ICIEV), pp. 1–6. IEEE (2014)Google Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2020

Authors and Affiliations

  • Sabiha Sunjida Ahmed
    • 1
  • Sharmin Akter Milu
    • 2
  • Md. Ismail Siddiqi Emon
    • 1
  • Sheikh Shahparan Mahtab
    • 3
    Email author
  • Md. Fahad Mojumder
    • 4
  • Md. Israq Azız
    • 5
  • Jamal Ahmed Bhuiyan
    • 6
  • M. J. Alam
    • 3
  1. 1.Department of CSEFeni UniversityFeniBangladesh
  2. 2.Department of CSTENoakhali Science & Technology UniversityNoakhaliBangladesh
  3. 3.Department of EEEFeni UniversityFeniBangladesh
  4. 4.Department of ECENorth South UniversityDhakaBangladesh
  5. 5.Department of EEEDhaka UniversityDhakaBangladesh
  6. 6.Department of CEFeni UniversityFeniBangladesh

Personalised recommendations