本文首先从信息源范围、信息源结构类型、信息存在状态、信息交流渠道等方面分析了信息源的特点,论述了通过网络信息源结构、内容、访问流量的挖掘实现对采集信息源进行评估与选择的策略。在此基础上,重点探讨了采集信息源的集成策略,包括集成角度、集成层次、集成模式与集成方案,并进一步结合竞争情报特性和数据挖掘功能,从集成对象、集成模式、集成层次、集成能力、对挖掘引擎效率与性能的影响等角度比较了各种集成方案的应用特性。
In this paper, the authors first analyze the characters of information sources from the scope, structure type, existence state, communication channel, and then probe into strategies about assessment and selection of them by mining the structure, content and access to the information sources. Then, the paper makes a detailed discussion on the strategies about integration, including the points of integration, levels of integration, patterns and methods of integration. At last, the authors make a comparison about the application of patterns and methods of integration from the points of view of integration object, model, level, capability, and effect on the mining engine efficiency and performance while taking the feature of competitive intelligence and performance of data mining into account.