欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
针对开源数据挖掘平台Weka在聚类方面只集成了少数聚类算法的缺点，对其进行二次开发，扩充其聚类算法。介绍FCM算法的基本思想和算法描述，将FCM算法嵌入到Weka平台，充分利用Weka的类和可视化功能。选取一种实例密度加权的方法对该算法进行改进，调整聚类中心位置，并将改进后的算法与原算法进行实验比较分析。实验结果表明，改进后的算法明显减少了迭代次数，并获得更好的聚类效果。
英文摘要：
Aiming at the shortcoming of the open source data mining platform Weka that in clustering it has only integrated few clustering algorithms, we conduct the secondary development on it and expand its clustering algorithms. The basic idea of the FCM algorithm and the algorithm description are introduced as well. By embedding the FCM algorithm into Weka platform, we make the full use of the class and visual functions of the Weka. We select an instance density weighted method to improve the algorithm and adjust the clustering centre position, then through experiment we compare and analyse the improved algorithm with the original one. Experimental result shows that the improved algorithm clearly reduces the times of iteration, and acts better clusterinz effect.
同期刊论文项目
同项目期刊论文
Abnormal Functional Brain Network Metrics for Machine Learning Classifier in Depression Patients Ide
期刊信息
位置：