Sentiment analysis based on transfer learning for Chinese ancient literature

With the rise of computational social science, analyzing sociological problems using computational methods has attracted widespread attention. In recent years, sentiment analysis has become a research hotspot in computational social science. But existing researches concentrate on modern text, such as product reviews and microblogging, and hardly involve the analysis of ancient literature. In this paper, we propose a model TL-PCO based on transfer learning to classify ancient literature, then through sentiments in ancient literature we can understand social and cultural development in that era. Our model utilizes the knowledge of ancient literature itself and the corresponding modern translation. Through two proposed functions based on transfer learning we can get two kinds of features. With the addition of features from ancient literature itself, three classifiers can be trained and then vote for the final category. Experiments demonstrate the effectiveness of the proposed method on the dataset of Chinese poems in Tang Dynasty. Moreover, the different periods of Tang Dynasty and different genres are analyzed in detail. Compared with the analysis of social history, the results confirmed the effectiveness of this method.