Supervised User Ranking in Signed Social Networks
Authors: Xiaoming Li, Hui Fang, Jie Zhang184-191
AAAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experimental results demonstrate the superiority of our approach over the state-of-the-art approaches. We conduct experiments on four real-world datasets and the results show that SSRW’s performance has an improvement of 6.05% compared to the state-of-the-art approaches. |
| Researcher Affiliation | Academia | Xiaoming Li, Hui Fang, Jie Zhang School of Computer Science and Engineering, Nanyang Technological University, Singapore Research Institute for Interdisciplinary Sciences and School of Information Management and Engineering Shanghai University of Finance and Economics, China |
| Pseudocode | Yes | Algorithm 1 Computation of r+ im, r im, r+ im wi , and r im wi |
| Open Source Code | No | The paper does not provide an explicit statement about open-sourcing the code for the described methodology or a link to a code repository. |
| Open Datasets | Yes | We employ the four datasets (i.e. Epinions, Slashdot, Wikipedia RFA and Bitcoin), which are the only public available datasets with signed structure. 1http://www.trustlet.org/epinions.html 2snap.stanford.edu/data/soc-Slashdot0902.html 3snap.stanford.edu/data/wiki-Rf A.html 4cs.umd.edu/ srijan/wsn |
| Dataset Splits | No | We use 2-fold cross-validation for training and testing, and utilize GAUC (Generalized AUC) (Song and Meyer 2015) to measure the ranking performance, which is formulated as:. The paper specifies 2-fold cross-validation but does not explicitly mention a separate validation set with specific split percentages or counts. |
| Hardware Specification | Yes | We then further empirically check the actual runtime of our approaches conducted on a four CPU 3.7GHz machine with 16GB memory. |
| Software Dependencies | No | The paper does not list specific software dependencies with version numbers (e.g., Python version, specific library versions like PyTorch 1.9 or scikit-learn 0.24). |
| Experiment Setup | Yes | In this experiment, we consider 5 features for the vector x to describe a user pair... we utilize the linear model to represent the link strength, i.e., fw(x) = w T x... In SSRW, there are five hyperparameters: δ, α, β, γ and the restart probability c. The first four are application-dependent, and in view of simplicity and fair comparisons, we make them equal to 1 respectively. Besides, we set c = 0.2 for SSRW in the comparative experiments considering favourable performance of our model in this setting. |