The classic algorithm of k-means is discussed,that is one of the most widespread methods in clustering,including both strongpoint's and shortages.Not only is it sensitive to the original clustering center,but also it may be affected by the k.Given these shortages,an improved algorithm is discussed,which makes improvements in k and selection of original clustering center.To select original clustering center based on the max-min distance.This paper presents the application which all show that the improved algorithm can lead to better and more stable solutions than k means algorithm.The experiment and application affection by the outliers is down to a much low figure.The improved algorithm was used to the students' score analysis in universities and had a good closer.