Improving Case-Based Reasoning Systems by Combining K-Nearest Neighbour Algorithm with Logistic Regression in the Prediction of Patients’ Registration on the Renal Transplant Waiting List
Figure 1
During the learning phase, a training set is used to compute the parameters of a logistic regression model. These parameters enable the computations of the weights of attributes as well as patients’ weights. Then a setting set is used to evaluate an optimal K value for the K-NN algorithm. Finally all these estimates are exploited to evaluate five decision making algorithms referred to by the indexes (i) to (v).