Table of Content

Open Access iconOpen Access

ARTICLE

Enhancing the Classification Accuracy in Sentiment Analysis with Computational Intelligence Using Joint Sentiment Topic Detection with MEDLDA

PCD Kalaivaani1,*, Dr. R Thangarajan2

1 Assistant Professor (Selection Grade), Department of CSE, Kongu Engineering College, Perundurai, Erode, Tamil Nadu 638052, India
2 Professor, Department of CSE, Kongu Engineering College, Perundurai, Erode, Tamil Nadu 638052, India

* Corresponding Author: PCD. Kalaivaani, email

Intelligent Automation & Soft Computing 2020, 26(1), 71-79. https://doi.org/10.31209/2019.100000152

Abstract

Web mining is the process of integrating the information from web by traditional data mining methodologies and techniques. Opinion mining is an application of natural language processing to extract subjective information from web. Online reviews require efficient classification algorithms for analysing the sentiments, which does not perform an in–depth analysis in current methods. Sentiment classification is done at document level in combination with topics and sentiments. It is based on weakly supervised Joint Sentiment-Topic mode which extends the topic model Maximum Entropy Discrimination Latent Dirichlet Allocation by constructing an additional sentiment layer. It is assumed that topics generated are dependent on sentiment distributions and the words generated are conditioned on the sentiment topic pairs. MEDLDA is used to increase the accuracy of topic modeling.

Keywords


Cite This Article

P. Kalaivaani and D. R. Thangarajan, "Enhancing the classification accuracy in sentiment analysis with computational intelligence using joint sentiment topic detection with medlda," Intelligent Automation & Soft Computing, vol. 26, no.1, pp. 71–79, 2020.



cc This work is licensed under a Creative Commons Attribution 4.0 International License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
  • 1497

    View

  • 953

    Download

  • 0

    Like

Share Link