Automatic Generation of Pan Concept Tree on Numerical Data
Rong Jiang · Chinese Journal of Computers · 2000
Concept hierarchy plays a fundamentally important role in data mining. Through automatically generating the concept hierarchies, the mining efficacy is improved, and the knowledge is discovered at different abstraction levels. The method to represent numerical concept with cloud models is introduced. Qualitative concepts can be represented effectively with three digital parameters of cloud models: expected value Ex, entropy En and hype entropy He. Cloud transform is realized to automatically produce the basic numerical concepts as the leaf nodes in pan concept tree. Automatic generation of pan concept tree based on cloud models is also provided. Lastly, climbing up and jumping up on the pan concept tree is studied, as the basis of discovering all kinds of knowledge in different levels.