阅读量：

50

作者：

John S. Whissell，Charles L. A. Clarke

展开

摘要：

weighting is heavily dependent on both the dataset being clustered and the algorithm used. In addition, binary weighting is shown to be consistently inferior to both weighting. We investigate clustering using both BM25 term saturation in isolation and BM25 term saturation with , confirming that both are superior to their non-BM25 counterparts under several common clustering quality measures. Finally, we investigate estimation of the 1 BM25 parameter when clustering. Our results indicate that typical values of 1 from other IR tasks are not appropriate for clustering; 1 needs to be higher.

展开

关键词：

document clustering feature weighting okapi BM25

DOI：

10.1007/s10791-011-9163-y

被引量：

29

年份：

2011