Saved in:
Bibliographic Details
Main Authors: Qin, Xiaoyu, Ting, Kai Ming, Zhu, Ye, Lee, Vincent CS
Format: Preprint
Published: 2019
Subjects:
Online Access:https://arxiv.org/abs/1907.00378
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929225365192704
author Qin, Xiaoyu
Ting, Kai Ming
Zhu, Ye
Lee, Vincent CS
author_facet Qin, Xiaoyu
Ting, Kai Ming
Zhu, Ye
Lee, Vincent CS
contents A recent proposal of data dependent similarity called Isolation Kernel/Similarity has enabled SVM to produce better classification accuracy. We identify shortcomings of using a tree method to implement Isolation Similarity; and propose a nearest neighbour method instead. We formally prove the characteristic of Isolation Similarity with the use of the proposed method. The impact of Isolation Similarity on density-based clustering is studied here. We show for the first time that the clustering performance of the classic density-based clustering algorithm DBSCAN can be significantly uplifted to surpass that of the recent density-peak clustering algorithm DP. This is achieved by simply replacing the distance measure with the proposed nearest-neighbour-induced Isolation Similarity in DBSCAN, leaving the rest of the procedure unchanged. A new type of clusters called mass-connected clusters is formally defined. We show that DBSCAN, which detects density-connected clusters, becomes one which detects mass-connected clusters, when the distance measure is replaced with the proposed similarity. We also provide the condition under which mass-connected clusters can be detected, while density-connected clusters cannot.
format Preprint
id arxiv_https___arxiv_org_abs_1907_00378
institution arXiv
publishDate 2019
record_format arxiv
spellingShingle Nearest-Neighbour-Induced Isolation Similarity and its Impact on Density-Based Clustering
Qin, Xiaoyu
Ting, Kai Ming
Zhu, Ye
Lee, Vincent CS
Machine Learning
A recent proposal of data dependent similarity called Isolation Kernel/Similarity has enabled SVM to produce better classification accuracy. We identify shortcomings of using a tree method to implement Isolation Similarity; and propose a nearest neighbour method instead. We formally prove the characteristic of Isolation Similarity with the use of the proposed method. The impact of Isolation Similarity on density-based clustering is studied here. We show for the first time that the clustering performance of the classic density-based clustering algorithm DBSCAN can be significantly uplifted to surpass that of the recent density-peak clustering algorithm DP. This is achieved by simply replacing the distance measure with the proposed nearest-neighbour-induced Isolation Similarity in DBSCAN, leaving the rest of the procedure unchanged. A new type of clusters called mass-connected clusters is formally defined. We show that DBSCAN, which detects density-connected clusters, becomes one which detects mass-connected clusters, when the distance measure is replaced with the proposed similarity. We also provide the condition under which mass-connected clusters can be detected, while density-connected clusters cannot.
title Nearest-Neighbour-Induced Isolation Similarity and its Impact on Density-Based Clustering
topic Machine Learning
url https://arxiv.org/abs/1907.00378