Study on rough set based reduction technique for case attributes

With the purpose of improving the efficiency of case retrieval, the rough set based on reduction technique for case attributes is studied. The basic theory of rough set is introduced first. Then starting from the core, we propose an improved algorithm based on differentiating matrix for obtaining minimal attributes reduction. For continuous attributes, the supervisory discrete algorithm based on sensitivity of approximation precision or SDA/BOSOAP is presented. The technique is applied in practical dynamic scheduling problem of some iron & steel works, the computation experiment shows that it can get rid of redundant information so as to improve the efficiency of case retrieval.