Research And Implementation Of Vertical Search Based On Fuel Price

According to the development of network technology,in the face of massive data information,the general search engine has been unable to meet the needs of professional search,while the vertical search technology for specialized fields,has been widely used in many fields,such as academic and shopping vertical search engines.Coal price information is an important information in the field of coal,which has an important guiding significance for the purchase decision-making of thermal power enterprises,but nowadays there is no vertical search engine for coal price.According to the characteristics of coal price information,a vertical search system for coal price is designed.The coal price information focused crawler is established,including the core content of topic determination,topic correlation calculation,link prediction and so on.The theme is described by selecting keywords manually,and each keyword is given weight by term frequency-inverse document frequency value.Using Word2 vec model to expand the topic words selected manually from the semantic point of view,and using Latent Dirichlet Allocation model to expand the topic of incremental crawling documents.The vector space model is used to describe the topic and web page,and the topic correlation of web page is calculated by using the frequency information of topic words.According to the characteristics of coal price information,coal price information is divided into index data and news data.Index data are obtained by Jsoup,regular matching and other means,stored in a structured way,and displayed in a broken line graph.News information data is obtained by web page,and its text is indexed and stored to provide users with retrieval function.Based on the above scheme,the vertical search in the field of coal price is finally realized,and index data and news data are successfully captured,and the retrieval and browsing functions are provided to users in the form of web.The test results show that the system can meet the needs of users for coal price information search,and can provide data support for coal price prediction and coal purchase decision.