Abstract
The Adaptive Mean Shift (AMS) algorithm is a popular and simple non-parametric clustering approach based on Kernel Density Estimation. In this paper the AMS is reformulated in a Bayesian framework, which permits a natural generalization in several directions and is shown to improve performance. The Bayesian framework considers the AMS to be a method of obtaining a posterior mode. This allows the algorithm to be generalized with three components which are not considered in the conventional approach: node weights, a prior for a particular location, and a posterior distribution for the bandwidth. Practical methods of building the three different components are considered.
Original language | English |
---|---|
Pages (from-to) | 3583-3592 |
Number of pages | 10 |
Journal | Journal of Intelligent and Fuzzy Systems |
Volume | 30 |
Issue number | 6 |
DOIs | |
Publication status | Published - 2016 Apr 30 |
Bibliographical note
Publisher Copyright:© 2016 - IOS Press and the authors. All rights reserved.
Keywords
- Adaptive mean shift algorithm
- kernel density estimation
ASJC Scopus subject areas
- Statistics and Probability
- General Engineering
- Artificial Intelligence