Notice: This website is continuously updated as new conference proceedings and journal volumes are published. Consequently, the figures presented here may differ from those reported in our publications.
A brief overview of our methodology is provided on the Methods page, along with references to publications documenting our approach and links to the code and data used in each study.
The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates.
Outlier Aware Network Embedding for Attributed Networks
Authors: Sambaran Bandyopadhyay, N. Lokesh, M. N. Murty45279
AAAI 2019 | Venue PDF | LLM Run Details | Input Tokens: 15,638 Total number of tokens sent to the LLM as input for this paper's analysis. | Output Tokens: 4,887 Total number of tokens produced by the LLM (including reasoning/thinking tokens) for this paper's analysis.
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We experimented on publicly available real networks and manually planted different types of outliers to check the performance of the proposed algorithm. Results demonstrate the superiority of our approach to detect the network outliers compared to the state-of-the-art approaches. We also consider different downstream machine learning applications on networks to show the efficiency of ONE as a generic network embedding technique. |
| Researcher Affiliation | Collaboration | Sambaran Bandyopadhyay IBM Research, India EMAIL Lokesh N. Indian Institute of Science, Bangalore EMAIL M. N. Murty Indian Institute of Science, Bangalore EMAIL |
| Pseudocode | Yes | Algorithm 1 ONE |
| Open Source Code | Yes | The source code is made available at https://github.com/sambaranban/ONE. |
| Open Datasets | Yes | The datasets are Web KB, Cora, Citeseer and Pubmed2. Datasets: https://linqs.soe.ucsc.edu/data |
| Dataset Splits | No | We split the set of nodes of the graph into training set and testing set. The training set size is varied from 10% to 50% of the entire data. The remaining (test) data is used to compare the performance of different algorithms. |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., GPU/CPU models, memory) used for running its experiments. |
| Software Dependencies | No | The paper mentions using algorithms and classifiers (e.g., random forest classifier, KMeans++, isolation forest) and notes implementation details for baseline algorithms, but it does not specify exact version numbers for software dependencies or libraries. |
| Experiment Setup | Yes | For ONE, we set the values of α and β in such a way that three components in the joint losss function in Eq. 5 contribute equally before the first iteration of the for loop in Algorithm 1. For all the experiments we keep embedding space dimension to be three times the number of ground truth communities. For each of the datasets, we run the for loop (Steps 4 to 6 in Alg. 1) of ONE only 5 times. |