The Research Of Web-Based Spatial Data Mining

Spatial Data Mining is a new kind of technology that is developed in 1990th. And recently more and more scholars have paid great attention to it. With the rapid development of network technologies, relevant researchers are more interested in how to mine spatial data in Internet or Intranet, that is, how to do Web Based Spatial Data Mining (WBSDM).The writer has also taken over a project--Web Based Spatial DataMining, which was proposed in 1999 by the writer and authorized to be subsidized by the Natural Science foundation of Hunan Province of China, from 2000 to 2003, and probed into some relative aspects. The dissertation focused on the technologies, algorithm and application of WBSDM, and the main studies and new contributions are described as follows:1. Technologies of spatial data mining and web-based data mining are summarized and then the concept of WBSDM is put forward and its main functions and meaning of research are disserted in detail.2. The framework of web-based spatial data mining is presented. And its background, definition, features, functions, system structure, implementation strategy, key technologies and so on are illustrated systematically.3. According to the operation requirements of WBSDM, based on the generalization of WebGIS, combined with multi-agent system, a distributed WebGIS integrated model is designed. The clear technological solution is following: the uniform dataaccess interface protocol based on XML/J2EE--Replys andRequest and applying JAFMAS to extend remote distributed data mining component. And the design goals, system structure and key technologies are described in detail so that the introduced model can fit the requirements of WBSDM.4. MAS technology is introduced into WBSDM and the pure Javacomponent of mobile Agent--JAFMAS, that is provided byothers, is analysed detailedly. In order to apply JAFMAS to WBSDM, painstaking research work is done and the component is extended. Multi-agent cooperating algorithm is also produced to fit requirements of WBSDM.5. The knowledge model based on rough information granules is built and a positive field equal algorithm based on fundamental information granules is identified. Then the characteristic of indistinguishable array is explained and an attribute reductionalgorithm based on rough information granules is proposed and realized in order that the rough set theory can be adapted to the mining in great spatial database.6. In order to solve the fusion problem of knowledge derived from different spatial database in WBSDM, a new knowledge discovery model based on knowledge base is given out. It is specified that it is effective in processing the redundancy, circulation and conflict of rule knowledge.7. A web-based spatial data mining prototype system, WBSDMS, is devised and carried out partially. The city waterworks database is mined by it, with the combination of the practical item. The system can do mining by selecting many algorithms and is open and extensible.