基于改进CFSFDP算法的文本聚类方法及其应用*
|
詹春霞, 王荣波, 黄孝喜, 谌志群
|
Application of Text Clustering Method Based on Improved CFSFDP Algorithm
|
Zhan Chunxia,Wang Rongbo,Huang Xiaoxi,Chen Zhiqun
|
|
表2 4种算法的Accuracy、Precision、Recall、F-Measure值比较 |
|
|
算法 | 数据集 | Accuracy | Precision | Recall | F-Measure | Agglomerative | data1050 | 0.7305 | 0.7743 | 0.7969 | 0.7854 | data3100 | 0.7077 | 0.6976 | 0.7811 | 0.7370 | data5000 | 0.6808 | 0.6598 | 0.6627 | 0.6612 | DBSCAN | data1050 | 0.6486 | 0.6795 | 0.7332 | 0.7052 | data3100 | 0.6797 | 0.6761 | 0.7880 | 0.7278 | data5000 | 0.6006 | 0.6270 | 0.6500 | 0.6643 | CFSFDP | data1050 | 0.8171 | 0.8050 | 0.8090 | 0.8070 | data3100 | 0.750 | 0.7375 | 0.6617 | 0.6975 | data5000 | 0.7425 | 0.7438 | 0.6189 | 0.6756 | 本文算法 | data1050 | 0.8333 | 0.7171 | 0.9098 | 0.8609 | data3100 | 0.7574 | 0.7421 | 0.7676 | 0.7546 | data5000 | 0.7712 | 0.7340 | 0.7450 | 0.7395 |
|
|
|