TY - JOUR
T1 - Balanced Graph Cut With Exponential Inter-Cluster Compactness
AU - Wu, Danyang
AU - Nie, Feiping
AU - Lu, Jitao
AU - Wang, Rong
AU - Li, Xuelong
N1 - Publisher Copyright:
© 2020 IEEE.
PY - 2022/8/1
Y1 - 2022/8/1
N2 - Recently, balanced graph-based clustering has been a hot issue in clustering domain, but the balanced theoretical guarantees of previous models are either qualitative or based on a probabilistic random graph, which may fail to various real data. To make up this vital flaw, this letter explores a novel balanced graph-based clustering model, named exponential-cut (Exp-Cut), via redesigning the intercluster compactness based on the exponential transformation exp μ x. It is worth noting that exponential transformation not only provides a bounded balanced tendency for Exp-Cut, but also helps Exp-Cut to achieve balanced results on an arbitrary graph via adjusting its curvature μ. To solve the optimization problem involved in Exp-Cut model, an efficient heuristic solver is proposed and the computational complexity is O(n2) per iteration. Experimental results demonstrate that our proposals outperform competitors on all benchmarks with respect to clustering performance, balanced property, and efficiency.
AB - Recently, balanced graph-based clustering has been a hot issue in clustering domain, but the balanced theoretical guarantees of previous models are either qualitative or based on a probabilistic random graph, which may fail to various real data. To make up this vital flaw, this letter explores a novel balanced graph-based clustering model, named exponential-cut (Exp-Cut), via redesigning the intercluster compactness based on the exponential transformation exp μ x. It is worth noting that exponential transformation not only provides a bounded balanced tendency for Exp-Cut, but also helps Exp-Cut to achieve balanced results on an arbitrary graph via adjusting its curvature μ. To solve the optimization problem involved in Exp-Cut model, an efficient heuristic solver is proposed and the computational complexity is O(n2) per iteration. Experimental results demonstrate that our proposals outperform competitors on all benchmarks with respect to clustering performance, balanced property, and efficiency.
KW - Clustering
KW - machine learning
KW - unsupervised learning
UR - http://www.scopus.com/inward/record.url?scp=85135337686&partnerID=8YFLogxK
U2 - 10.1109/TAI.2021.3123126
DO - 10.1109/TAI.2021.3123126
M3 - 文章
AN - SCOPUS:85135337686
SN - 2691-4581
VL - 3
SP - 498
EP - 505
JO - IEEE Transactions on Artificial Intelligence
JF - IEEE Transactions on Artificial Intelligence
IS - 4
ER -