Role-based Multiplex Network Embedding
Authors: Hegui Zhang, Gang Kou
ICML 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | The proposed methods are empirically evaluated on the network reconstruction, node classification, link prediction, and multi-class edge classification tasks using several real-world multiplex networks. The experimental results on eight public, real-world multiplex networks demonstrate that the proposed methods outperform state-of-the-art baseline methods. |
| Researcher Affiliation | Academia | Hegui Zhang 1 Gang Kou 1 School of Business Administration, Faculty of Business Administration, Southwestern University of Finance and Economics, Chengdu 611130, China. Correspondence to: Gang Kou <kougang@swufe.edu.cn;kougang@yahoo.com>. |
| Pseudocode | Yes | Algorithm 1 The RMNE method and Algorithm 2 The role-modified random walk method |
| Open Source Code | No | The paper does not contain any explicit statement or link indicating that the source code for the described methodology (RMNE, RMNE+) is publicly available. |
| Open Datasets | Yes | Eight real-world multiplex networks were used to validate the performance of the proposed methods. The basic statistics for the multiplex networks used are given in Table 1. The CKM (Coleman et al., 1957) physician-innovative multiplex network contains three different layers. |
| Dataset Splits | No | The paper mentions that 'the ROC-AUC was used as an evaluation metric of model performance and it is calculated by averaging the results of 10 runs,' implying a form of cross-validation, but it does not specify explicit percentages or absolute counts for training, validation, and test splits for the datasets. |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., GPU/CPU models, memory, or cloud computing instance types) used for running the experiments. |
| Software Dependencies | No | The paper mentions general software components like 'Skip Gram model' and 'Adam optimizer', but does not provide specific version numbers for any software dependencies used in their implementation. |
| Experiment Setup | Yes | For the RMNE, we set K = 20, Le = 15, and W = 10. The optimal values of r and t were chosen from a set of {0.25, 0.5, 1.0, 2.0, 4.0}. For RMNE+, we set α = β = 1, γ = 0.5, the other parameters were set the same as the MANE (Ata et al., 2021). |