对特定区域搜索引擎的自动分类系统的研究

随着因特网的飞速发展，特定区域搜索引擎（Domain-spedfic search engines）正变的越来越重要，因为这种搜索引擎通常能提供更精确的结果和一些一般的搜索引擎所不能提供的信息。然而特定区域搜索引擎通常需要花很多的时间来组建和维持。文中提出一个基于机器学习的方法来自动完成和维持这种特定区域搜索引擎，即运用最大加权依赖树分类方法改进以往的方法进行自动分类，使分类结果更为精确。运用此技术可以组建一个新的特定区域搜索引擎，将给人们的生活、学习提供方便。

Research on Automatic Classification System to Build Domain-Specific Search Engines

As the amount of information on World Wide Web grows, search engines become increasing difficult to find what we want.Domain-specific search engines are becoming increasingly popular because they often increase accuracy and extra features.Unfortunately, they are also difficult and time-consuming to maintain.This paper proposes the use of machine learning techniques to automate the creation and maintenance of domain-specific search engines greatly. Describe new research in information learning and text classification.Especially use the method of Maximum Weight Dependence Trees to improve the Naive Bayes method.Using these techniques,can build a domain-specific search engine that is convenient to people's study,life,and etc.