Spectral Clustering of Signed Graphs via Matrix Power Means
Authors: Pedro Mercado, Francesco Tudisco, Matthias Hein
ICML 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive experiments on random graphs and real world datasets confirm the theoretically predicted behaviour of the signed power mean Laplacian and show that it compares favourably with state-of-the-art methods. and 4. Experiments on Wikipedia-Elections |
| Researcher Affiliation | Academia | 1Saarland University 2University of T ubingen 3University of Strathclyde. |
| Pseudocode | Yes | Algorithm 1: Spectral clustering of signed graphs with Lp |
| Open Source Code | No | The paper does not contain an explicit statement about releasing the source code for the methodology, nor does it provide a link to a code repository. |
| Open Datasets | Yes | We now evaluate the Signed Power Mean Laplacian Lp with p { 10, 5, 2, 1, 0, 1} on Wikipedia-Elections dataset (Leskovec & Krevl, 2014). |
| Dataset Splits | No | No explicit training, validation, or test dataset splits (e.g., percentages, counts, or specific predefined split references) are provided in the main text for either the synthetic or real-world datasets. |
| Hardware Specification | No | No specific hardware details (e.g., CPU or GPU models, memory, or cluster specifications) used for running the experiments are mentioned in the paper. |
| Software Dependencies | No | No specific software dependencies, libraries, or frameworks with version numbers (e.g., 'Python 3.8, PyTorch 1.9') are explicitly mentioned in the paper. |
| Experiment Setup | Yes | We set the number of clusters to identify to k = 30 and in Fig. 4 we portray the portion of adjacency matrices of positive and negative edges W + and W corresponding to k 1 clusters sorted according to the corresponding identified clusters. and fixing the sparsity of G+ and G by setting p+ in+ p+ out = 0.1 and p in + p out = 0.1 with two clusters each of size 100. |