A Machine Learning based Framework for Sentiment Classification: Indian Railways Case Study
D. Krishna Madhuri
D. Krishna Madhuri, Department of Computer Science and Engineering, GRIET, Hyderabad (Telangana), India.
Manuscript received on 05 February 2019 | Revised Manuscript received on 13 February 2019 | Manuscript published on 28 February 2019 | PP: 441-445 | Volume-8 Issue-4, February 2019 | Retrieval Number: D2733028419/19©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Machine learning in the field of computer science is the application of Artificial Intelligence (AI) that helps in making systems intelligent. It focuses on producing algorithms that may lead to AI applications in the real world. As enterprises are producing huge amount of data, it became indispensable to have machine learning techniques in place for discovering business intelligence from data for strategic decision making. However, in the contemporary era, the traditional data may be deemed inadequate for decision making. The rationale behind this is that people of all walks of life are able to exchange ideas and opinions/sentiments over social media like Facebook and Twitter. In other words, there is social feedback exists in Online Social Networks (OSNs). Collection of social media data related to business and using machine learning algorithms to extract useful knowhow from such data bestows competitive edge to enterprises. The existing literature on sentiment analysis has plenty of methods for discovering sentiments. However, it is still an open problem to have optimizations. In this paper we proposed a framework for discovering sentiments from tweets of Indian Railways. This is a domain specific framework which leverages business intelligence through different classifiers such as C4.5, Naive Bayes, SVM and Random Forest. An evaluation procedure with measures like precision, recall, F-Measure and accuracy is provided. The empirical study with Indian Railways case study revealed that the proposed framework is useful in sentiment analysis and can be tailored to suit other domains as well. By considering the atweets of Indian Railways as a case study evaluation is made in terms of precision, recall and F-Measure.
Keyword: Sentiment Classification, Machine Learning, C4.5, Naive Bayes, SVM, Random Forest.
Scope of the Article: Machine Learning