A coarse to fine granular tree based on density peaks

DOI

10.2991/icmmct-17.2017.226How to use a DOI?

Keywords

Multi-granularity; Density peaks; Hierarchical clustering; coarse to fine

Abstract

The granular computing is a methodology aiming at simulating the process and structure of the human cognition in the world. A new promising clustering algorithm, the density peak clustering (DPC) was proposed recently, which whereas, just takes procedures at a single granularity and could be conditionally ineffective by the inaccurate judgment by the decision graph. In this paper, we expand the DPC to the multi-granularity space and construct a coarse to fine granular tree. The structure deeply simulates the cognizing framework of human from a view of global to local of conceptions, making an innovative enlightenment to hierarchical clustering and cognitive computing in fields like robotics. Experiments show that the method includes every possible conclusion of the DPC by variable peaks picked from the decision graph, thus avoiding the limitations by uncertain artificial selections, at the same time, providing a competitive granular tree framework that could be analogically transplanted to other hierarchical algorithms.

Copyright

© 2017, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Xukun Li
AU  - Jie Yang
PY  - 2017/04
DA  - 2017/04
TI  - A coarse to fine granular tree based on density peaks
BT  - Proceedings of the 2017 5th International Conference on Machinery, Materials and Computing Technology (ICMMCT 2017)
PB  - Atlantis Press
SP  - 1146
EP  - 1151
SN  - 2352-5401
UR  - https://doi.org/10.2991/icmmct-17.2017.226
DO  - 10.2991/icmmct-17.2017.226
ID  - Li2017/04
ER  -