"Draw My Topics": Find Desired Topics fast from large scale of Corpus

We develop the “Draw My-Topics” Toolkit, which provides a fast way to incorporate social scientists’ concerns and interests into the standard topic model. Instead of using raw corpus with primitive processing as input, an algorithm based on Vector Space Model and Conditional Entropy are used to connect social scientists’ subjective want and the unsupervised topic models’ output. Space for users’ adjustment on specific corpus of their interest is accommodated in our algorithm. We demonstrate the toolkit’s use on the Diachronic People’s Daily Corpus in Chinese. Several interesting “central words” like “Enlai Zhou” (First PRC premier minister) and “Cultural Revolution” which may be interested of social scientists from different disciplines and the original corpus are used as input of our toolkit, then the most related topics are present efficiently for further research purpose.