Abstract:Clustering ensemble, which aims to learn a robust consensus clustering from multiple weak base clusterings, has achieved promising performance on various applications. With the development of big data, the scale and complexity of data is constantly increasing. However, most existing clustering ensemble methods typically employ shallow clustering algorithms to generate base clusterings. When confronted with high-dimensional complex data, these shallow algorithms fail to fully utilize the intricate features present in the latent data space. As a result, the quality and diversity of the generated base clusterings are insufficient, thus affecting the subsequent ensemble performance. To address this issue, we propose a novel clustering ensemble algorithm for handling deep embeddings using cluster confidence (CEDECC) to improve the robustness and performance. Instead of simply combining deep clustering with clustering ensembles, we take into consideration that the performance of existing deep clustering methods heavily relies on the quality of low-dimensional embeddings generated during the pre-training stage. The quality of embeddings is unstable due to the influence of different initialization parameters. In CEDECC, specifically, we first construct a cluster confidence measure to evaluate the quality of low-dimensional embeddings. Typically, high-quality low-dimensional embeddings yield accurate clustering results with the same model parameters. Then, we utilize multiple high-quality embeddings to generate the base partitions. In the ensemble strategy phase, we consider the cluster-wise diversity and propose a novel ensemble cluster estimation to improve the overall consensus performance of the model. Extensive experiments on three benchmark datasets and four real-world biological datasets have demonstrated that the proposed CEDECC consistently outperforms the state-of-the-art clustering ensemble methods.

A Cluster-Based Intelligence Ensemble Learning Method for Classification Problems.

Ensemble Learning Method for Classification: Integrating Data Envelopment Analysis with Machine Learning

CEIL: A General Classification-Enhanced Iterative Learning Framework for Text Clustering

A new classifier for imbalanced data with iterative learning process and ensemble operating process

A Hybrid and Ensemble Intelligent Pattern Classification Algorithm

Knowledge Based Cluster Ensemble for Cancer Discovery from Biomolecular Data

LibD3C: Ensemble Classifiers with a Clustering and Dynamic Selection Strategy

Ensemble Classification Based on Supervised Clustering for Credit Scoring

Hybrid Incremental Ensemble Learning for Noisy Real-World Data Classification.

A Novel Classifier Ensemble Method Based on Subspace Enhancement for High-Dimensional Data Classification.

Clustering Algorithm Based Selective Ensemble

Clustering ensemble algorithm for categorical data

A Novel Clustering-Ensemble Approach

A novel ensemble classifier based on multiple diverse classification methods

Adaptive Random Subspace Ensemble Classification Aided by X-means Clustering

Dynamic Classifier Ensemble Using Classification Confidence

A Clustering Ensemble Algorithm for Handling Deep Embeddings Using Cluster Confidence

A Cluster Ensemble Method for Clustering Categorical Data

Improved Selective Deep-Learning-Based Clustering Ensemble

A Novel SVM Ensemble Approach Using Clustering Analysis

A Study of Classification Algorithm for Data Mining Based on Hybrid Intelligent Systems