期刊
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
卷 33, 期 4, 页码 1389-1400出版社
IEEE COMPUTER SOC
DOI: 10.1109/TKDE.2019.2942029
关键词
Optimization; Gene expression; Fuses; Runtime; Bipartite graph; Minimization; Robustness; Co-clustering; co-clustering ensemble; multiple relevance measures; trace minimization; robustness
类别
资金
- Natural Science Foundation of China [61872300, 61873214]
- Fundamental Research Funds for the Central Universities [XDJK2019B024]
- Natural Science Foundation of CQ CSTC [cstc2018jcyjAX0228]
This paper proposes a co-clustering ensemble approach based on multiple relevance measures to achieve good consensus solutions. The method evaluates the quality of base co-clusters and measures feature-to-object relevance, combining feature-to-feature and object-to-object relevance information. Experimental results demonstrate that the approach outperforms other related co-clustering ensembles and has reduced runtime cost.
Co-clustering aims at discovering groups of both objects and features from a given data matrix. Co-clustering ensembles can produce robust co-clusters by combining multiple base co-clusterings. However, current co-clustering ensemble solutions either ignore the constraints resulting from feature-to-feature and object-to-object relevance information, or ignore feature-to-object relevance information. In this paper, we advocate that all three information sources contribute to the achievement of good consensus solutions, and propose a co-clustering ensemble (CoCE) approach based on multiple relevance measures. CoCE first evaluates the quality of base co-clusters and consequently measures feature-to-object relevance. The latter, along with feature-to-feature and object-to-object relevance measures, contribute to the definition of a hybrid graph. The consensus process uses the resulting hybrid graph; it's formulated as a trace minimization problem and introduces a block-wise matrix multiplication technique to perform the optimization. Experimental results on various datasets show that CoCE not only frequently outperforms other related co-clustering ensembles, but also has reduced runtime cost and is more robust to poor base co-clusterings.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据