Research On The Method Of Decision Rule Acquisition In Concept Lattice

As a kernel data structure in the theory of formal concept analysis, concept lattice is a powerful tool for data analysis. There are many advantages to build and apply concept hierarchy in the knowledge acquisition; meanwhile, Hasse diagram of concept lattice embodies such concept hierarchy structure. Therefore, studying the basic theory of concept lattice and applying it in knowledge acquisition have important significance.In this paper, algorithm of extracting decision rules based on minimal generator(MGA algorithm) and algorithm of extracting decision rules set using proper premise(PPA algorithm) are studied, some meaningful results are obtained.In the MGA algorithm, at first concept lattice and minimal generators are created according to decision context, then those minimal generators of the concepts which contain decision attributes are computed to generate decision rules. Moreover, support and confidence degree are obtained. Finally, the algorithm runs on a medicine database to construct decision rules, and the results illustrate its validity.In the PPA algorithm, all the concepts are generated from the decision context, and then subsets of intent for each concept are judged whether to be a proper premise, thus proper premises according to each concept are computed. At last, decision rules set is obtained by all the proper premises, which is a complete decision rules set.In conclusion, stability of MGA algorithm and PPA algorithm are verified in the paper. Furthermore, through comparison and analysis of the two algorithms, it can been found that the number of the decision rules would be greatly reduced by means of increasing traversing index tree of the concept lattice. This result has certain guiding significance for the extracting rules from concept lattice.