从智能空间决策支持系统的功能需求出发,分析了森林防火是复杂的空间决策问题,它具有动态性、分布性、数据量大、数据性质复杂等特点,决策者面对突如其来的灾害及海量的、复杂的、易变的和分布的数据和信息,很难做出及时、科学和准确的决策。因此,建立集成空间信息分析处理的、具有知识发现能力的智能决策支持系统是十分必要的。而空间在线分析处理(Spatial OLAP)和空间在线分析挖掘(Spatial OLAM,Spatial OLAP mining)是集成空间数据处理的OLAP和OLAM,能够在空间数据仓库上发现隐含的知识和规则,并在不同的维、度上进行查询和分析。从系统需求分析、功能设计、系统实现和系统运行实例方面,论述融合了GIS、空间OLAP和空间OLAM技术的空间智能决策支持系统的设计思想和关键技术的实现。提出基于空间OLAP和OLAM知识发现机制,建立一种集成GIS、空间数据仓库、空间OLAP/OLAM和决策支持系统的新型决策分析工具。
This paper analyzes a complicated space decision question to fire prevention in the forest from function demand for intelligent spatial decision support system (ISDSS). Its data quantity is big, data character is complex, and it has dynamic and distributed characteristics and so on. Facing the unexpected disaster and massive, complex, volatile and distribution of data and information, it is difficult for policy-maker to make in time, accurate and scientific decision-making. Therefore, it is essential to set up integrated spatial information analysis processing system, having knowledge to find intellectual DSS of ability. And spatial on-line analytical processing (Spatial OLAP) and spatial on-line analytical processing mining (Spatial OLAM, Spatial OLAP mining) are on daty of the integrated space data processing, which can find implicit knowledge and rules on the space data warehouse, and inquire and analyze in different dimension and degree. From system requirement analysis, function design, system implementation and system operation example, this article expounds the fact that has merged the design philosophy of GIS, space OLAP and intellectual DSS of the space OLAM technology have emerged and key technology. This article proposed knowledge discovery mechanism based on spatial OLAP and the OLAM, establishes one kind to integrate GIS, the spatial data warehouse, spatial OLAP/OLAM and the decision support system—a new decision analysis tool.
[1] 万鲁河, 李一军. 集成"3S"技术的森林防火决策支持系统研究[J]. 系统工程理论与实践. 2004, (7): 88~93.
[2] 万鲁河, 刘万宇, 臧淑英. 森林防火辅助决策支持系统的设计与实现[J]. 管理科学. 2003, 16(3): 21~24
[3] Han J, Stefanovic N, Koperski K. Selective Materialization: An Efficient Method for Spatial Data Cube Construction. In Proceeings of the Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD’98), 1998, 4(1): 144~158.
[4] Papadias Dimitris, Kalnis Panos, Tao Yufei. Efficient OLAP Operations in Spatial Data Warehouses. HKUST-CS01-01, 2001
[5] Stefanovic N,Han J,Koperski K. Object-Based Selective Materialization for Efficient Implementation of Spatial Data Cubes [J]. IEEE Transactions on Knowledge and Data Engineering, 2000, 12(6): 938~ 958
[6] 傅明. 基于Web的空间数据挖掘. 中南大学博士论文. 2004.5: 11~14
[7] 樊博. 面向客户智能的空间数据挖掘技术研究. 哈尔滨工业大学博士论文. 2004: 40~72
[8] 万鲁河. 集成空间信息分析处理的智能决策支持系统研究. 哈尔滨工业大学博士论文. 2005.10:96~97
[9] K. Koperski, J. Adhikary and J.Han. Spatial Data Mining: Progress and Challenges. In Workshop on Research Issues on Data Mining and Knowledge Discovery (DMKD’96). 1996: 1~10
[10] S. Chawla, S. Shekhar, W.–L Wu, and U. Ozesmi. Modeling Spatial Dependencies for Mining Geospatial data: An introduction. In: Harvey Miller and Jiawei Han, Editors, Geographic Data Mining and Knowledge Discovery (GKD). London and New York. 2001: 32~159
[11] Lin Wen-yang, Kuo I-chung. OLAP data cubes configuration with genetic algorithms [J]. IEEE Transactionson Knowledge and Data Engineering. 2000, 2(5): 132-143
