Abstract: 【Purpose/significance】The tacit knowledge contained in Traditional Chinese Medicine medical records is of great value. The analysis of Traditional Chinese Medicine medical records through data mining technology and the mining of tacit knowledge in clinical medical records are beneficial to the personal knowledge and experience of well-known Traditional Chinese Medicine practi⁃ tioners. Inheritance and accumulation and expansion of theoretical knowledge of Traditional Chinese Medicine.【Method/process】Col⁃ lect the chief complaints and current medical history of patients in Chinese medical records, and clean the data of the chief complaints and current medical history. Use the data mining technology based on the cart algorithm to mine tacit knowledge in traditional Chinese medical records to explore the disease and patient symptoms the relationship between various attributes.【Result/conclusion】Taking stomachache as an example, this paper finds the correlation between stomachache and various attributes of patients' symptoms, which provides a reference for data mining technology in the research of tacit knowledge mining in Traditional Chinese Medicine medical re⁃ cords.【Innovation/limitation】The cart algorithm used in this paper to judge the correlation between gastric pain and patient attributes is compared with the diagnosis results in medical records, and the correlation accuracy of this method is higher than that of ID3 algo⁃ rithm, C4.5 algorithm and SLIQ algorithm, and the correlation is the highest.