首先提出了一种优化初始中心点方法用以解决聚类的局部最优问题.同时通过样本的模糊加权减少边缘噪音数据对聚类效率的影响.文本聚类试验表明,该模糊文本聚类算法取得较好的聚类效果.
This paper proposes a new way that selects initial cluster center in order to solve the partially most superior phenomenon.By using the fuzzy weighting on the samples,this improved method decreases the influence that the k-means algorithm is very sensitive to the isolated point.Lastly,we have a test about text clustering and the result shows that this method obtains good clustering effect.