一種基于SOM改進的PCM聚類方法
打開文本圖片集
摘 要:針對PCM算法在聚類計算過程中存在的初始聚類中心隨機選取,聚類結(jié)果可能陷入局部最優(yōu)解等問題,提出一種改進策略。利用SOM網(wǎng)絡(luò)對數(shù)據(jù)進行初步處理,得到PCM算法的初始聚類中心,使得算法聚類效果得到明顯提升。
關(guān)鍵詞:可能性c均值聚類;自組織映射;聚類
中圖分類號:TP183 文獻標志碼:A 文章編號:2095-2945(2022)03-0133-03
Abstract: In order to solve the problem that the initial clustering center is randomly selected and the clustering result may fall into the local optimal solution in the clustering calculation process of PCM (possibilistic c-means clustering) algorithm, an improved strategy is proposed. The initial clustering center of the PCM algorithm is obtained by using the SOM (self-organizing mapping) network to deal with the data, which improves the clustering effect of the algorithm obviously.
Keywords: PCM (possibilistic c-means clustering); SOM (self-organizing mapping); clustering
在大數(shù)據(jù)背景下,數(shù)據(jù)挖掘技術(shù)逐漸成為當(dāng)今熱門的研究課題。(剩余3377字)