Multidimensional Domain Knowledge Framework for Poet Profiling

Authorship profiling is a subtask of authorship identification. This task can be regarded as an analysis personal writing styles, which has been widely investigated. However, no previous studies have attempted to analyze the classical Chinese poetry. First, we provide approach evaluate popularity poets, and also establish public corpus containing top 20 most popular poets in Tang Dynasty for profiling. Then, novel poetry framework named multidimensional domain knowledge poet (M-DKPP) proposed, combining attribution text’s stylistic features with described by experts traditional studies. A case study Li Bai used prove validity applicability our framework. Finally, performance M-DKPP evaluated four poem datasets. On all datasets, proposed outperforms several baseline approaches attribution.