Font Size: a A A

Research And Application Of Interactive Mining Of Association Rules

Posted on:2007-12-11Degree:MasterType:Thesis
Country:ChinaCandidate:T BaiFull Text:PDF
GTID:2178360212958603Subject:Computer software and theory
Abstract/Summary:PDF Full Text Request
With the rapidly development of information industrialization, the conflict between explosion of data and poorness of knowledge has been more and more acute, and data mining plays more and more important roles in many areas. Of all the braches of data mining, the research on association rule mining is the one deepest explored and its application is the most widely used. At present the research on association rule mining is mostly based on the support-confidence framework. Du to the intrinsic shortcoming of the frame, users can not use the mining system interactively and this results in that few association rules are interesting to them. So the exploration of how to set up an interactive system of association rule mining in order to produce more interesting association rules has become a novel and popular task in the research of association rule mining. Several experts employed new measures to improve the satisfaction and interest of association rules without adopting support-confidence framework in association rule mining. Under such background, this paper makes a study of the interactive association rule mining and proposes an algorithm for interactive mining of association rules (IMAR).Firstly, this paper introduces the basic theory related to data mining and association rule mining. Then is a traditional algorithm (Apriori) of association rule mining based on support-confidence framework. Finally, the paper put forwards and discuses in detail an algorithm for tackling interactive association mining, which is called interactive mining of association rules (IMAR).The algorithm prescribes the rule constraints and expresses them using Boolean expression. The rule constraints include not only the minimal support and confidence threshold but also the items that the head and the body of an association rule must have or must not have. IMAR sets up the set of frequent sets using the lattice-based approach. Firstly, the...
Keywords/Search Tags:Association Rule, Interactiveness, Constraints, DataMining, Concept Lattice
PDF Full Text Request
Related items