Protein Function Prediction using a Double Weighted K-Nearest Neighbor Method

摘要

In this paper, we present an improved KNN (K-Nearest Neighbor) classification method for protein function prediction, which is called DWKNN (Double Weighted K-Nearest Neighbor). In order to estimate its performance, we compare it with the original KNN method and WKNN (Weighted K-Nearest Neighbor) method using the model organism yeast (Sacchyromyces cerevisiae). The experiment results show that DWKNN method is superior to KNN and WKNN in general.