Research Article
A Design of the Group Decision Making Medical Diagnosis Expert System Based on SED-JD Algorithm
@INPROCEEDINGS{10.1007/978-3-030-21373-2_33, author={Na Zong and Wuyungerile Li and Pengyu Li and Bing Jia and Xuebin Ma}, title={A Design of the Group Decision Making Medical Diagnosis Expert System Based on SED-JD Algorithm}, proceedings={Security and Privacy in New Computing Environments. Second EAI International Conference, SPNCE 2019, Tianjin, China, April 13--14, 2019, Proceedings}, proceedings_a={SPNCE}, year={2019}, month={6}, keywords={Medical expert system Group Decision Making Similarity measurement}, doi={10.1007/978-3-030-21373-2_33} }- Na Zong
Wuyungerile Li
Pengyu Li
Bing Jia
Xuebin Ma
Year: 2019
A Design of the Group Decision Making Medical Diagnosis Expert System Based on SED-JD Algorithm
SPNCE
Springer
DOI: 10.1007/978-3-030-21373-2_33
Abstract
Medical expert system not only has a lot of medical professional knowledge, but also has inference ability. The inference engine is not only one of the cores of the expert system, but also the key to designing the expert system. We focus on inference engine. In order to improve the diagnostic accuracy of medical diagnostic expert system, we propose the Group Decision Making (GDM) medical diagnosis expert system based on the Standardized Euclidean Distance-Jaccard Distance (SED-JD) algorithm. The mainly research content of inference engine is similarity measurement algorithm (that is SED-JD) and inference engine rule scheme (that is GDM). In order to get more accurate diagnosis, data preprocessing was performed before our experiments. In the design of inference engine, the selection of the Group Decision Making Objects (GDMOs) depends on the maximum similarity distance (MaxDist). The final decision result depends on the average similarity distance of each subgroup. By comparing the similarity scheme and GDM scheme, the experimental results show that GDM scheme is more effective and accurate. By comparing the Standardized Euclidean Distance (SED) algorithm, the Jaccard Distance (JD) algorithm and SED-JD algorithm, the experimental results show that SED-JD algorithm is more accurate.
