A Cluster-Weighted Kernel K-Means Method for Multi-View Clustering
Authors: Jing Liu, Fuyuan Cao, Xiao-Zhi Gao, Liqin Yu, Jiye Liang4860-4867
AAAI 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experimental results on both synthetic and real data sets demonstrate the effectiveness of the proposed method. |
| Researcher Affiliation | Academia | 1School of Computer and Information Technology, Shanxi University, Taiyuan 030006, P.R. China 2School of Software, Shanxi Agricultural University, Taigu 030801, P.R. China 3School of Computing, University of Eastern Finland, Kuopio 70211, Finland |
| Pseudocode | Yes | We summarize the above optimization process in Algorithm 1. |
| Open Source Code | No | The paper does not provide concrete access to source code. |
| Open Datasets | Yes | MSRC-v1 1: This is an image data set... 1https://www.microsoft.com/en-us/research/project/image-understanding/ Caltech101-7 2: This is a subset of Caltech101 image data set... 2http://www.vision.caltech.edu/Image Data sets/Caltech101/ Handwritten numerals (HW) 3: This data set consists of 2000 instances... 3https://archive.ics.uci.edu/ml/datasets/Multiple+Features Reuters 4: This is a multilingual data set... 4https://archive.ics.uci.edu/ml/datasets 5http://membres-lig.imag.fr/grimal/data.html |
| Dataset Splits | No | The paper does not provide specific dataset split information (exact percentages, sample counts, citations to predefined splits, or detailed splitting methodology) for training, validation, or testing, beyond mentioning some standard datasets. |
| Hardware Specification | No | The paper does not provide specific hardware details used for running its experiments. |
| Software Dependencies | No | The paper does not provide specific ancillary software details with version numbers. |
| Experiment Setup | Yes | For MVKKM and CWK2M, the parameter p for each data set is searched in logarithm form (log10 p from 0.1 to 2 with step size 0.2), and the initial centers are selected on a single view with the best performance by the global kernel k-means initialization algorithm. ... We use Guassian kernel for the MSRC-v1, Caltech101-7 and HW data sets by setting the standard deviation to be the median of the pairwise Euclidean distances between instances of each view, and use linear kernel for the Reuters data set. |