Enhancing Ensemble Clustering with Adaptive High-Order Topological Weights
Authors: Jiaxuan Xu, Taiyong Li, Lei Duan
AAAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive experiments on multiple datasets demonstrate the effectiveness of the proposed method. |
| Researcher Affiliation | Academia | 1 School of Computer Science, Sichuan University, Chengdu, China 2 School of Computing and Artificial Intelligence, Southwestern University of Finance and Economics, Chengdu, China |
| Pseudocode | Yes | Algorithm 1: ALM update Z and Algorithm 2: AWEC are provided. |
| Open Source Code | Yes | The source code of the proposed approach is available at https://github.com/ltyong/awec. |
| Open Datasets | Yes | We conduct extensive experiments on 14 real datasets from different domains. Characteristics of these datasets are provided in Table 1. We randomly run the k-means algorithm 100 times on each dataset (http://archive.ics.uci.edu/datasets) (Huang, Wang, and Lai 2017; Zhou, Zheng, and Pan 2019; Yu et al. 2022) to generate the base clustering result set. |
| Dataset Splits | No | The paper mentions running k-means on datasets to generate base clustering results and conducting repeated experiments, but it does not specify explicit train/validation/test dataset splits or their percentages/counts for reproduction. |
| Hardware Specification | No | The paper does not provide specific hardware details such as GPU or CPU models, memory, or cloud instance types used for running the experiments. |
| Software Dependencies | No | The paper mentions algorithms and methods used (e.g., k-means, ADMM, spectral clustering) but does not list specific software dependencies with version numbers (e.g., Python, PyTorch, scikit-learn versions). |
| Experiment Setup | Yes | For the number of neighbors parameter in Eq. (4), we set it to 0.5s in all datasets, where s = n/c represents the average sample number in each category. AWEC has two main parameters: the noise regularization parameter λ and the parameter γ. We perform a 6*6 grid search for λ in the set {0.01, 0.02, 0.04, 0.08, 0.1, 0.2} and γ in the set {0.1, 0.5, 1, 5, 10, 50}. We set the ensemble size M = 20, conduct 10 repeated experiments with different base clustering combinations. |