K-means algorithm is introduced. An improved algorithm is proposed for the disadvantages of randomly selecting the initial center of clustering and the vulnerability to the effects of outliers. The Distance Mean method (DM) was used to remove the outliers, then high-density and max-distance (HDMD) was used to improve the selection of the initial center of clustering. The comparison experiment before and after the improvement was carried out. Experimental results show that the improved algorithm is stable and accurate. The improved algorithm was applied to computer language teaching and achieved good classification effects. The improved algorithm is used to research and analyse mobile customer records, the effect is in conformity with the actual situation.
展开▼
