Overcoming key weaknesses of distance-based neighbourhood methods using a data dependent dissimilarity measure

Title: Overcoming key weaknesses of distance-based neighbourhood methods using a data dependent dissimilarity measure
Creator: Ting, Kaiming; Zhu, Ye; Carman, Mark; Zhu, Yue; Zhi-Hua, Zhou
Date: 2016
Type: Text; Conference paper
Identifier: http://researchonline.federation.edu.au/vital/access/HandleResolver/1959.17/161653
Identifier: vital:12527
Identifier: https://doi.org/10.1145/2939672.2939779
Identifier: ISBN:9781450342322
Abstract: This paper introduces the first generic version of data dependent dissimilarity and shows that it provides a better closest match than distance measures for three existing algorithms in clustering, anomaly detection and multi-label classification. For each algorithm, we show that by simply replacing the distance measure with the data dependent dissimilarity measure, it overcomes a key weakness of the otherwise unchanged algorithm.
Relation: 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco; August 13th-17th, 2016 p. 1205-1214
Rights: Copyright ACM
Rights: This metadata is freely available under a CCO license
Reviewed

		Thumbnail	File	Description	Size	Format