Data Sampling method for Comprehensive Classification to Blood Test Data

Abstract

In this paper, we propose a new sampling method in medical data analysis to predict the occurrence of clinical events. In the previous method, the data before the occurrence of a clinical event was sampled as a positive case, and the past data farthest from the clinical event was sampled as a negative case so that there was a one-to-one correspondence between the positive and negative cases for each patient. However, in some patients, only one of the data was present. Therefore, in this paper, in order to make effective use of these patient data, the data of patient data where only one of the data exists is sampled so that the positive and negative cases are one-to-one.