Using Group Deep Learning and Data Augmentation in Persian Sentiment Analysis

Sentiment analysis is one of the most important areas in natural language processing, which investigates people’s opinions on products, services, and various events. The number of papers published on Persian sentiment analysis is much less than in other languages such as English, and this made our research in Persian use different approaches to boost the performance of Persian sentiment analysis. In this research, Group Deep Learning and Data Augmentation approach has been used to classify emotions in Persian, which has not been used in any previous research to analyze Persian emotions. After applying the Data Augmentation technique and using Group Deep Learning approach for classification, we achieved 96.5% accuracy in Persian sentiment analysis.