Mining Classification Knowledge Based on Cloud Models

Abstract: Data classification is an important research topic in the field of data mining and knowledge discovery. It finds the common properties among a set of objects in a database and classifies them into different classes. There have been many data classification methods studied, including decision-tree method, statistical methods, neural networks, rough sets, etc. In this paper, we present a new mathematical representation of qualitative concepts—Cloud Models. With the new models, mapping between quantities and qualities becomes much easier and interchangeable. Based on the cloud models, a novel qualitative strategy for data classification in large relational databases is proposed. Then, the algorithms for classification are developed, such as cloud generation, complexity reduction, identifying interacting attributes, etc. Finally, we perform experiments on a challenging medical diagnosis domain, acute abdominal pain. The results show the advantages of the model in the process of knowledge discovery