TCM Data Mining Platform And The Services

Data mining technology became more and more mature in the past twenty years, and achieved great successes in many applications like business data processing and internet data analyzing etc. In the meanwhile, it is used by more and more applications from different domains. Meanwhile, TCM has been developed for thousands of years, and it gathered rich data with multiple types. These data contains a large number of unknown rules and knowledge. And the knowledge will play important roles in the development of modern Chinese medicine theory. In the last ten years, the CCNT laboratory of Zhejiang University did a lot of work to informationize the TCM data in co-operation with the Institute of Information on Traditional Chinese Medicine. They built multiple large scale TCM databases, and laid a good foundation for the data mining research of this domain. But there are many problems in real research and application: solutions have few standards to follow; each mining platform only possess relatively weak functions, while sharing and reusing of resources can hardly be efficient; most of these mining systems lack the ability to process large scale domain-concerned mining task.Facing these requirements from traditional Chinese medicine data mining, the data mining group of CCNT laboratory designed and implemented the DartSpora2.0, a data mining service platform. This platform draws on the advantages and principles of open source mining software, and provides rich domain-concerned mining service to TCM researchers. In this project, I accomplished the design and most of the implementation work of the 2.0 version. I also take part in the building of distributed parallel data mining algorithm library. Besides, I co-operated with other classmates and staff of Institute of Information on Traditional Chinese Medicine on the domain-concerned cases formulating, which are used as the solution of corresponding problems and courses of the platform operation.