An RBF Neural Network Clustering Algorithm Based on K-Nearest Neighbor

Table 1

Algorithm process description.

---

Algorithm1 RBF-KNN clustering

---

Input: Dataset X, k, K-nearest neighbor-value

Output: Cluster labels c_i of X

(1): Run K-means to pre-divide X int k clusters

(2): Transform pre-divide labels to pseudo-labels

(3): Repeat

(4):  Training RBF network by minimizing loss

(5): Repeat

(6):  Get correction labels by KNN graph

(7):  Update network parameters by minimizing loss

(8): Data_mark = RBF_KNN(X)

(9): Run K-means to divide Data_mark into k clusters

---