A mountain means clustering algorithm
Junnian Wang, Jianxun Liu, Lanxia Liu · 2008
A modified mountain clustering algorithm based on the hill valley function is proposed. Firstly, the grid and the mountain function are constructed in data space according to the mountain clustering method, and the mountain values of the data are computed. Secondly, the hill valley function is introduced to partition the data distributed on each peak. If the hill valley function’ value of two datum equal to 1, it means these two datum are on the same mountain and belong to thee same cluster, otherwise they are not. Finally, the means of data samples in each cluster are computed as the clustering centres. The testing of three data base indicate that the proposed mountain means clustering algorithm can categorise the clustering centres and the data numbers in each clusters exactly as well as efficiently.