论文部分内容阅读
针对互联网开放性、层次性、演化性、巨量性等本质特性,从复杂自适应系统这一全新的角度,以农业垂直搜索为应用背景,提出一种复杂自适应搜索模型.该搜索模型的主要特点是通过建立信息采集、分类、清洗与服务智能体联盟,组成多智能体实验环境.通过建立模型的学习机制与进化机制,改善搜索模型对网络环境的动态适应能力.经过与现有主流搜索引擎的比较实验发现,它在查准率方面具有明显优势.同时,由于该搜索模型具备通用的结构体系,因而在建立其它行业的垂直搜索模型时,可被方便地移植使用.
Aiming at the nature of open, hierarchical, evolutive and huge nature of Internet, a complex and adaptive search model is proposed from the perspective of complex adaptive system with the vertical search of agriculture as an application background. The main feature is to establish a multi-agent experiment environment by establishing a coalition of information collection, classification, cleaning and service agents, and to improve the dynamic adaptability of the search model to the network environment by establishing a learning mechanism and an evolutionary mechanism of the model. Compared with the search engines, it is found that the search engine has the obvious advantage in the accuracy of the search, meanwhile, the search model can be conveniently used when establishing the vertical search model in other industries because of its universal structure.