These articles are all based on western language systems, so I was curious about how it can work in Chinese. I am suprised to find that there are also very many uses of text mining based on Chinese. For example, the keyword composition of Chinese Tang poems and Song lyrics mapped by Zero E-lab research lab some time ago is very interesting, capturing all the high frequency or more trendy words of ancient poems. And the connection between each keyword is clearly shown by using the method of network analysis diagram, and even some readers are able to deduce some classic poems by themselves according to that network diagram. However, it is not so simple to deal with the subtext of ancient Chinese (literary), especially poetry, because single-word words account for more than 80% of the statistical information of ancient Chinese vocabulary, plus the fact that ancient Chinese has a great deal of meaning in small words, so the subtext technique for modern Chinese is often not applicable to it. In the analysis process, the results are only the “introduction” and “clues”, the most important thing is to rely on the human brain to analyze the results, with the help of the background/business knowledge and analytical models, from the surface of the text to its depths, to discover the deeper meaning that can not be grasped by the shallow reading, to explore its value.