The scale and complexity of health information service system has increased dramatically, and its development activities and management are difficult to control. In the field of, Traditional methods and simple mathematical statistics methods are difficult to solve the problems caused by the explosive growth of data and information, which will adversely affect health information service system management finally. So, it is particularly important to find valuable information from the source code, design documents and collected software datasets and to guide the development and maintenance of software engineering. Therefore, some experts and scholars want to use mature data mining technologies to study the large amount of data generated in software engineering projects (commonly referred to as software knowledge base), and further explore the potential and valuable information inherently hidden behind the software data. This article initially gives a brief overview of the relevant knowledge of data mining technology and computer software technology, using decision tree graph mining algorithm to mine the function adjustment graph of the software system definition class, and then source code annotations are added to the relevant calling relationships. Data mining technology and computer software technology are deeply integrated, and the decision tree algorithm in data mining is used to mine the knowledge base of computer software. Potential defect changes are listed as key maintenance objects. The historical versions of source code change files with defects are found dynamically and corrected in time, to avoid the increase of maintenance cost in the future.