Optimized Ensemble Model for Sentiment Analysis incorporating Grey Wolf and Genetic Algorithm with Voting Classifier

Abstract

The sentiment analysis is the opinion mining approach which use NLP for the categorization. The sentiment analysis can be performed on various social media data. The machine learning is the approach which are applied for the classification. In the previous research work, tokenization technique is applied for the sentiment analysis which is less efficient as compared to machine learning.

In this research work machine learning algorithms are applied for the sentiment analysis on live twitter data.  Diverse stages are executed to analyze the sentiments in which the data is pre-processed, attributes are extracted and the data is classified. The hybrid optimization algorithm is proposed in this research work for the feature extraction.

The hybrid optimization algorithm is the combination of genetic and gray wolf algorithm. The voting classification model is the used for the classification which is the combination of KNN, SVM and decision tree. The performance of proposed model is tested in terms of accuracy, precision and recall. It is analysed that proposed model achieves 98.89 percent accuracy for the sentiment analysis.