AIFeb 13, 2017
On Seeking Consensus Between Document Similarity Measures
arXiv:1702.03724v11 citations
Originality Synthesis-oriented
AI Analysis
This is an incremental study on consensus clustering methods for document analysis, with limited practical impact.
The paper tackled the problem of selecting a consensus partition from multiple document similarity measures, finding that using a complement of the Rand Index leads to the trivial total-separation partition where each element is in its own set.
This paper investigates the application of consensus clustering and meta-clustering to the set of all possible partitions of a data set. We show that when using a "complement" of Rand Index as a measure of cluster similarity, the total-separation partition, putting each element in a separate set, is chosen.