Feature Enhancement Based Text Sentiment Classification using Deep Learning Model

摘要

Text sentiment classification is a significant task in the recent years to understand the opinions and thoughts hidden in the text to enhance more productivity in e-commerce websites and also in the social media. Here we integrate deep learning models to analyze the text sentiments. In this paper, Convolutional Recurrent Neural Network (CRNN) method for text sentiment analysis is proposed. The proposed CRNN is a combination of different layers used to extract the features from the text dataset. During training CRNN is able to learn the features set of the text sentiment dataset. The performance of the proposed approach is evaluated on text sentiments of publically available movie review (MR) dataset. Results show that the proposed method outperforms the traditional deep learning techniques