Rough Knowledge Mining Algorithm Based on Extension Decision Rule Lattice for Electronic Patient Record

Electronic patient record mining deals with the implicit and useful medical information stored in the electronic patient record database. By this technology the useful knowledge is extracted and the scientific and auxiliary decision-making is proved for the diagnosis and treatment of disease. In this paper, a rough knowledge mining algorithm (named RKMA) based on extension decision rule lattice is presented. Firstly the electronic patient records are converted into different recordings in order to be further reduced and classified in the medical database. Secondly both variable rough thresholding and concept lattice are related. A novel extension decision rule lattice is constructed. This extension lattice can rough reduce the mutative knowledge attributions and well mine all disease diagnosis rules without more redundancy by avoiding finding frequent item sets. The experimental results show this RKMA algorithm well can mine consistent decision rules, and it is better robustness than other algorithms on the capability of attribution reduction and rule mining for electronic patient records.