Research on Abnormal Detection of Gynecological Medical Record Coding Data Based on Improved Outlier Detection Algorithm
In response to the problem of low accuracy in analyzing gynecological medical record coding data nowadays,an improved detection algorithm combining self-organizing mapping neural network and local anomaly factors was proposed,and the friend relationship model was integrated into the improved algorithm to obtain the fusion algorithm.The performance comparison analysis of the fusion algorithm proposed in the study shows that the recognition rate and misjudgment rate of the algorithm are 97.5%and 0.18%,respectively,which are superior to the comparison algorithm.After conducting empirical analysis on the algorithm,it was found that the average time cost of the algorithm was 73 seconds when the data volume was 3000,which was significantly better than the comparison algorithm.The above results indicate that the fusion algorithm proposed in the study has good abnormal data detection performance,and its application in abnormal detection of gynecological medical record coding data can effectively improve the diagnostic accuracy of gynecological diseases.
SOMLOFGynecological medical recordsAbnormal dataFriendly relationship model
万方数据
维普
