Name
Title | ||
|---|---|---|
A <italic>Group</italic>-Based Distance Learning Method for Semisupervised Fuzzy Clustering |
Abstract | ||
|---|---|---|
Learning a proper distance for clustering from prior knowledge falls into the realm of semisupervised fuzzy clustering. Although most existing learning methods take prior knowledge (e.g., pairwise constraints) into account, they pay little attention to local knowledge of data, which, however, can be utilized to optimize the distance. In this article, we propose a novel distance learning method, which learns from the
<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Group</i>
-level information, for semisupervised fuzzing clustering. We first present a new format of constraint information, called
<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Group</i>
-level constraints, by elevating the pairwise constraints (must-links and cannot-links) from
<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">point</i>
level to
<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Group</i>
level. The
<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Groups</i>
, generated around data points contained in the pairwise constraints, carry not only the local information of data (the relation between close data points) but also more background information under some given limited prior knowledge. Then, we propose a novel method to learn a distance by using the
<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Group</i>
-level constraints, namely,
<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Group</i>
-based distance learning, in order to optimize the performance of fuzzy clustering. The distance learning process aims to pull must-link
<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Groups</i>
as close as possible while pushing cannot-link
<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Groups</i>
as far as possible. We formulate the learning process with the weights of constraints by invoking some linear and nonlinear transformations. The linear
<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Group</i>
-based distance learning method is realized by means of semidefinite programming, and the nonlinear learning method is realized by using the neural network, which can explicitly provide nonlinear mappings. Experimental results based on both synthetic and real-world datasets show that the proposed methods yield much better performance compared to other distance learning methods using pairwise constraints. |
Year | DOI | Venue |
|---|---|---|
2022 | 10.1109/TCYB.2020.3023373 | IEEE Transactions on Cybernetics |
Keywords | DocType | Volume |
Constraint weight,distance learning,Mahalanobis distance,neural networks,pairwise constraints,semisupervised fuzzy clustering | Journal | 52 |
Issue | ISSN | Citations |
5 | 2168-2267 | 0 |
PageRank | References | Authors |
0.34 | 30 | 5 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Xuyang Jing | 1 | 2 | 1.37 |
| Zheng Yan | 2 | 923 | 67.53 |
| Yinghua Shen | 3 | 138 | 6.12 |
| W. Pedrycz | 4 | 13966 | 1005.85 |
| Yang Ji | 5 | 127 | 27.38 |