An Enhanced Ensemble Learning Method for Sentiment Analysis based on Q-learning

Abstract

Ensemble learning is a powerful technique for combining multiple classifiers to achieve improved performance. However, the challenge of applying ensemble learning to dynamic and diverse data, such as text in sentiment analysis, has limited its effectiveness. In this paper, we propose a novel reinforcement learning-based method for integrating base learners in sentiment analysis. Our method modifies the influence of base learners on the ensemble output based on the problem space, without requiring prior knowledge of the input domain. This approach effectively manages the dynamic behavior of data to achieve greater efficiency and accuracy. Unlike similar methods, our approach eliminates the need for basic knowledge about the input domain. Our experimental results demonstrate the robust performance of the proposed method compared to traditional methods of base learner integration. The significant improvement in various evaluation criteria highlights the effectiveness of our method in handling diverse data behavior. Overall, our work contributes a novel reinforcement learning-based approach to improve the effectiveness of ensemble learning in sentiment analysis.