Scalable Attributed-Graph Subspace Clustering
Authors: Chakib Fettal, Lazhar Labiod, Mohamed Nadif
AAAI 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experiments on several real-world attributed datasets demonstrate the cost-effective nature of our method with respect to the state of the art. In this section, we conduct experimentation to showcase the effectiveness and efficiency of our SAGSC model. |
| Researcher Affiliation | Collaboration | 1 Centre Borelli UMR 9010, Universit e Paris Cit e, 75006 Paris, France 2 Informatique Caisse des D epˆots et Consignations |
| Pseudocode | Yes | Algorithm 1: Scalable Attributed-Graph Subspace Clustering (SAGSC). |
| Open Source Code | Yes | Code for our paper can be found in 1. 1https://github.com/chakib401/sagsc |
| Open Datasets | Yes | In our experiments, We use six commonly used benchmark datasets to compare the different models including three citation network datasets (ACM, DBLP (Wang et al. 2019); Pub Med (Sen et al. 2008); and Wiki (Yang et al. 2015)), an Amazon sales dataset (Computers) (Shchur et al. 2018) and one large scale dataset (OGBN-ar Xiv) (Hu et al. 2020). |
| Dataset Splits | No | The paper does not provide specific details on the training, validation, and test dataset splits used for its own experiments, such as percentages or sample counts. While it mentions tuning hyperparameters, the exact splitting methodology for reproduction is not described. |
| Hardware Specification | No | The paper states: 'All experiments were implemented in Tensor Flow and conducted on a standard computer with a 12GB memory GPU an a RAM of 12GB.' This lacks specific hardware details like GPU or CPU models. |
| Software Dependencies | No | The paper mentions 'implemented in Tensor Flow' but does not provide specific version numbers for TensorFlow or any other software dependencies. |
| Experiment Setup | Yes | For our model, we use a quadratic kernel feature map with a bias term equal to 1/2. We start by choosing an interval for the powers we wish to consider e.g. the multiples of five plus one between one and a hundred i.e. {1, 6, . . . , 96}. |