Research Of Agriculture Search Visualization Platform

With the development of information technology, the requirement of the agriculture information is increasing. In this background, agriculture search engine was generated. Compared with general search engine, as vertical search engine in the agriculture, it provides users with more professional search results. But even that the search results still contain large amounts of information. To allow users to grasp the information they queried quickly, it is a good choice to display the search results by using visualization method. Information visualization combines many theories and methods of data mining, image technology and many other disciplines. It uses computer to visualize abstract information so that can search information quickly and enhance cognitive. How to combine agriculture search engine and information visualization so that can provide users precision agriculture information and make them see the visual search results at the same time? Under the support of the national science and technology supporting plan under grant N0.2006BAD10A1410 named"an agriculture search engine based on Ontology"and the national Science foundation of China under grant N0.60774096 named"an agriculture complex adaptive search engine", this paper discusses visualization for agricultural information search.At the beginning, this paper introduces some basic knowledge and methods about agricultural search engine and information visualization. Then this paper discusses agriculture text information visualization and produce price visualization. In the processing of agriculture text information visualization, according to the characteristics of text, this paper decides to establish a visual description of them with content relevant method. The characteristics that can reflect the content of text are extracted by vector space model and evaluation function. Then the relationships between the characteristics are built by word co-occurrence analysis. Finally, the visualization data composed from the characteristics and the relationship between characteristics is showed by visualization mapping called network structure view clearly.In the processing of produce price visualization, this paper establishes an index of geographic information for locating geographic information. Then the paper forecast price for cities of missing data by spatial data interpolation and divide the price level by cluster analysis. Visualization data is composed from price and price level. Finally, produce price is reflected directly by color and icon on the Google Maps.Agriculture search engine solves the problem of getting agriculture information. This paper designs and achieves agriculture search visualization platform based on agriculture search engine and describes the platform in detail from overall structure to design of the core module. The platform solves the problem that search result is not vivid enough and makes users understand information deep. Visualization service has been used in agriculture search engine and it achieves good results and meets the expected targets of design according to users'feedback.