Fusion Multiple Kernel K-means

Authors: Yi Zhang, Xinwang Liu, Jiyuan Liu, Sisi Dai, Changwang Zhang, Kai Xu, En Zhu9109-9117

AAAI 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Comprehensive experimental results demonstrate that our proposed algorithm achieves stateof-the-art performance on multiple public datasets, validating its effectiveness.
Researcher Affiliation Academia 1 School of Computer, National University of Defense Technology, Changsha, China, 410073 2 CCF Theoretical Computer Science Technical Committee, Shenzhen, China, 518064
Pseudocode Yes Algorithm 1: Solving H with orthogonality constraint via curvilinear search algorithm and Algorithm 2: Fusion Multiple Kernel K-means
Open Source Code Yes The code of this work is publicly available at https://github.com/ethan-yizhang/Fusion-Multiple-Kernel K-means.
Open Datasets Yes Multiple public datasets are adopted to evaluate the performance of our proposed FMKKM, including Texas1, Wisconsin1, Football2, BBCSport3, Willow4, Flower175, Flower1025, ALOI-1006, Reuters7. The detail information of datasets is summarized in Table 1.
Dataset Splits No The paper mentions 'For all algorithms, we repeat each experiment 50 times with random initialization to reduce the randomness effect caused by k-means' but does not provide specific train/validation/test split percentages, sample counts, or references to predefined splits for the datasets used.
Hardware Specification Yes All experiments are performed on a PC with Intel Core i9-10900X CPU and 64G RAM.
Software Dependencies No The paper mentions using 'k-means' and solving equations 'by SVD', but it does not specify version numbers for any software libraries, frameworks, or programming languages used for implementation.
Experiment Setup Yes For all datasets, the true number of clusters k is prespecified and set as the input of algorithms. We repeat each experiment 50 times with random initialization to reduce the randomness effect caused by k-means. Figure 4 presents the ACC of FMKKM on Wisconsin and ALOI-100 datasets by varying λ1 in 2[1:9] and λ2 in 2[3:10], respectively.