Real Estate-oriented Vertical Search Engine Research And Implementation

With the rapid development of the Internet, the network information is increasing exponentially. In such a vast amount of information, we need to use search engine to locate information. Although the common search engine can solve the problems of resource positioning partly, but its effect is not obvious. For the information retrieval in the professional field, it is difficult to fulfill the users’ demand. The vertical search engine can solve the problem well.In this paper do in-depth research in the key technologies of vertical search engine from both theoretical and practical aspects.In this paper, first introduce the background and significance of the research, the classification of the search engine and the domestic and international development of vertical search engine. Then paper introduce the basic working principle, and introduce the system components of vertical search engine, and the key technologies of vertical engine. Later, paper introduce the topic feature of the web, construct theme feature vector, and introduce the distribution feature of the web. Paper in-depth research the theme correlation degree judgment based on content and link structure, and analyze the drawback of the two judgments. Paper import the web importance degree on the basic of theme correlation degree judgment based on the content, and design a theme crawler algorithm based on the web’s content and link structure. Theme crawler usually exist the problems of web theme island, it will cause the low cover of the theme information. In this paper we design a tunnel algorithm with the dynamic regulation of its max depth. This algorithm remits the problems of island in some degree. Then paper design a vertical search engine based on the real estate, introduce the design of the overall framework of the system, introduce the design and the implementation of every functional module, and do performance analysis and function testing for this system. At last, summarize works in this paper, and raise the future work of my study.