Instance-Wise Weighted Nonnegative Matrix Factorization for Aggregating Partitions with Locally Reliable Clusters
Abstract
We address an ensemble clustering problem, where reliable clusters are locally embedded in given multiple partitions. We propose a new nonnegative matrix factorization (NMF)-based method, in which locally reliable clusters are explicitly considered by using instance-wise weights over clusters. Our method factorizes the input cluster assignment matrix into two matrices H and W, which are optimized by iteratively 1) updating H and W while keeping the weight matrix constant and 2) updating the weight matrix while keeping H and W constant, alternatively. The weights in the second step were updated by solving a convex problem, which makes our algorithm significantly faster than existing NMF-based ensemble clustering methods. We empirically proved that our method outperformed a lot of cutting-edge ensemble clustering methods by using a variety of datasets.
Cite
Text
Zheng et al. "Instance-Wise Weighted Nonnegative Matrix Factorization for Aggregating Partitions with Locally Reliable Clusters." International Joint Conference on Artificial Intelligence, 2015.Markdown
[Zheng et al. "Instance-Wise Weighted Nonnegative Matrix Factorization for Aggregating Partitions with Locally Reliable Clusters." International Joint Conference on Artificial Intelligence, 2015.](https://mlanthology.org/ijcai/2015/zheng2015ijcai-instance/)BibTeX
@inproceedings{zheng2015ijcai-instance,
title = {{Instance-Wise Weighted Nonnegative Matrix Factorization for Aggregating Partitions with Locally Reliable Clusters}},
author = {Zheng, Xiaodong and Zhu, Shanfeng and Gao, Junning and Mamitsuka, Hiroshi},
booktitle = {International Joint Conference on Artificial Intelligence},
year = {2015},
pages = {4091-4097},
url = {https://mlanthology.org/ijcai/2015/zheng2015ijcai-instance/}
}