Association Rule Mining And Its Applied Research, Concept Retrieval

With the rapid development of Internet technology and then the surpassing increase of all kinds of information characterized by geometric progression, The intellectualized information retrieval has become a major research topic, for the traditional information retrieval can't meet the requirements of users. Since the concept-based retrieval is one of the most important ways of intellectualized information retrieval. In order to solve the existing problem about concept-based retrieval, this paper expounds and proves how to generate the association between concepts and the concepts layers, build concept-based trees automatically and achieve the concept-based retrieval through the technology of mining association rules.Firstly, this paper briefly introduces the development regarding the concept-based retrieval and the technology of mining association rules, Based on this introduction, this paper analyses the existing problems about the concept-based retrieval. Secondly, in order to find the solution to how to derive key words/entries from documents, this paper puts forward a matrix-weighted mining association rule algorithm based on the weighted mining algorithm and the weighted value of vector space model, VSM. According to this novel algorithm, the set of frequent items can be derived from the idea of k-weight-estimate, and next, association rules can be discovery according to the matrix-weighted confidence. Experiments have proved that this novel algorithm is really works. And thirdly, to copy with the interactive mining association rules while retrieving, this paper further puts forward a mining constrained association rule algorithm, which achieves its goal through the mined association rules. This mining constrained association rule algorithm works just became it saves a great deal of time without re-mining the database. What's more, this conclusion is based on experiments. And then, to solve the retrieval expansion, this paper further puts forward a technology of concept-based association expansion retrieval, and a method of generating concept layers, which is able to submit the relevant forms expanded from the retrieval ones in line with the association rules of key words/entries to the retrieval system. Finally, On the basis of the mining association rule algorithm, coupled with the newly-raised algorithm and technology, a prototype is designed, which is a Chinese concept-based information retrieval based mining association rules.