Research On Mining Technology Oriented To The Agriculture Information Service Platform

With the rapid development of computer network technology and information technology, the ability to product and collect data is greatly improved. Also the database technology is promoted. Now how to extract knowledge and information from mass data is the work in data mining for human service. Data mining is a multi crossing areas with the application-oriented. It can promote the application of data mining techniques and theoretical study. One of the urgent problems with data mining to solute is how to construct information for all level users.According to the combination of the technology of data mining with modern agriculture information, research on mining technology oriented to the agriculture information service platform aims to solve the problem that the bad time-effective and incomplete information about the users demanded.This paper analyzes features of data, design the method of data mining and the knowledge acquisition modes, according to the regional and seasonal of agricultural information. Based on the wasted of dimension in traditional vector space model algorithm, the article gives the method to solve. Through analysis, different frequencies of keywords and reference library of keywords was used to mine the data. Through constructing the reference library of keywords in constructing data process using vector space model, it reduced the information lost and the dimension of vector space. Different levels of knowledge were also got through the data's volume, drill, rotating, slice etc. The mining of User access information aims to provide value-added services for the system.The experiments show that mining through different levels can get different levels knowledge. And based on the construction of library reference of keywords solve the problem of vector space dimension data and low density. The quality of clustering is improved.