Overcoming Key Weaknesses of Distance-based Neighbourhood Methods using a Data Dependent Dissimilarity Measure

被引：46

作者：

Ting, Kai Ming ^{[1
]}

Zhu, Ye ^{[2
]}

Carman, Mark ^{[2
]}

Zhu, Yue ^{[3
]}

Zhou, Zhi-Hua ^{[3
]}

机构：

[1] Federat Univ, Churchill, Vic 3842, Australia

[2] Monash Univ, Clayton, Vic 3800, Australia

[3] Nanjing Univ, Nanjing 210023, Jiangsu, Peoples R China

来源：

KDD'16: PROCEEDINGS OF THE 22ND ACM SIGKDD INTERNATIONAL CONFERENCE ON KNOWLEDGE DISCOVERY AND DATA MINING | 2016年

关键词：

Data dependent dissimilarity; distance measure; distance-based neighbourhood; probability-mass-based neighbourhood; k nearest neighbours; SIMILARITY; MASS;

D O I：

10.1145/2939672.2939779

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

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.

引用

页码：1205 / 1214

页数：10

共 50 条

[1] Overcoming weaknesses of density peak clustering using a data-dependent similarity measure
Rasool, Zafaryab
Aryal, Sunil
Bouadjenek, Mohamed Reda
Dazeley, Richard
[J]. PATTERN RECOGNITION, 2023, 137
[2] Data-dependent dissimilarity measure: an effective alternative to geometric distance measures
Aryal, Sunil
Ting, Kai Ming
Washio, Takashi
Haffari, Gholamreza
[J]. KNOWLEDGE AND INFORMATION SYSTEMS, 2017, 53 (02) : 479 - 506
[3] Data-dependent dissimilarity measure: an effective alternative to geometric distance measures
Sunil Aryal
Kai Ming Ting
Takashi Washio
Gholamreza Haffari
[J]. Knowledge and Information Systems, 2017, 53 : 479 - 506
[4] A Euclidean distance-based measure of efficiency in data envelopment analysis
Amirteimoori, Alireza
Kordrostami, Sohrab
[J]. OPTIMIZATION, 2010, 59 (07) : 985 - 996
[5] Optimizing distance-based methods for large data sets
Scholl, Tobias
Brenner, Thomas
[J]. JOURNAL OF GEOGRAPHICAL SYSTEMS, 2015, 17 (04) : 333 - 351
[6] Optimizing distance-based methods for large data sets
Tobias Scholl
Thomas Brenner
[J]. Journal of Geographical Systems, 2015, 17 : 333 - 351
[7] Gene expression data classification: some distance-based methods
Makinde, Olusola Samuel
[J]. KUWAIT JOURNAL OF SCIENCE, 2019, 46 (03) : 31 - 39
[8] An Internal Cluster Validity Index Using a Distance-based Separability Measure
Guan, Shuyue
Loew, Murray
[J]. 2020 IEEE 32ND INTERNATIONAL CONFERENCE ON TOOLS WITH ARTIFICIAL INTELLIGENCE (ICTAI), 2020, : 827 - 834
[9] Evaluating the geographic concentration of industries using distance-based methods
Marcon, E
Puech, F
[J]. JOURNAL OF ECONOMIC GEOGRAPHY, 2003, 3 (04) : 409 - 428
[10] Benchmarking distance-based partitioning methods for mixed-type data
Efthymios Costa
Ioanna Papatsouma
Angelos Markos
[J]. Advances in Data Analysis and Classification, 2023, 17 : 701 - 724

← 1 2 3 4 5 →