Geographic Data Clustering Research based on Improved Support Vector Machine

According to the intensive research on geographic data mining algorithm, comparing and analyzing the existing spatial data -mining algorithm, a novel geographic data clustering algorithm based on improved support vector machine is put forward. In order to reduce constraint and complexity of the problem, we propose an improved support vector machine. Besides, an adaptive factor is introduced into support vector machine to make it have feature selection function. The experiments are done based on data from Taihang mountain space geographic data of the geography insitute of the Chinese science academy. The results show that the proposed support vector machine can reduce clustering complexity and save execution time. It provides significant reference for geographic data clustering.