数据流挖掘技术及其在仿真中的应用

Data Streams Mining Techniques and its Application in Simulation System

随着仿真系统复杂程度的增加和规模的增大,仿真时间越来越长,仿真所产生的数据量越来越大,使得仿真数据具有数据流的特性,因此可以采用数据流挖掘技术处理仿真数据。综述了数据流和数据流挖掘技术的主要特点;提出了基于数据流挖掘技术的仿真应用框架;设计了通用数据流挖掘成员,以便能够快速将数据流挖掘算法集成到基于HLA体系结构的仿真系统中,并以导弹突防仿真系统为例介绍了所设计的通用数据流关联规则挖掘成员。

With increasing of complexity and augmenting of scale of simulation system, the simulation time becomes much longer and the simulation data becomes much huger. These make the simulation data has the characters of data streams, so that the simulation data can be processed by data streams mining techniques. We summarized the characters of data streams and data streams mining techniques, and presented simulation application framework based on data streams mining techniques, and designed general data-streams-mining federate to make it easy to integrate the various data streams mining algorithms in HLA architecture-based simulation system quickly, and introduce our general federate for mining association rule in data streams with the example of Missile-Breakthrough simulation system.