Random Forest: A Hybrid Implementation for Sarcasm Detection in Public Opinion Mining
Ashwini M Joshi1, Sameer Prabhune2
1Ashwini M Joshi, Department of CSE, SGBAU, Amaravati.
2Sameer Prabhune, Department of CSE, SGBAU, Amaravati.
Manuscript received on September 16, 2019. | Revised Manuscript received on 24 September, 2019. | Manuscript published on October 10, 2019. | PP: 5022-5025 | Volume-8 Issue-12, October 2019. | Retrieval Number: L37581081219/2019©BEIESP | DOI: 10.35940/ijitee.L3758.1081219
Open Access | Ethics and 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: Modelling the sentiment with context is one of the most important part in Sentiment analysis. There are various classifiers which helps in detecting and classifying it. Detection of sentiment with consideration of sarcasm would make it more accurate. But detection of sarcasm in people review is a challenging task and it may lead to wrong decision making or classification if not detected. This paper uses Decision Tree and Random forest classifiers and compares the performance of both. Here we consider the random forest as hybrid decision tree classifier. We propose that performance of random forest classifier is better than any other normal decision tree classifier with appropriate reasoning.
Keywords: Random Forest, Decision Tree, Pruning, Diverse, Hybrid.
Scope of the Article: Bio-Science and Bio-Technology