SnapNETS: Automatic Segmentation of Network Sequences with Node Labels

Authors: Sorour Amiri, Liangzhe Chen, B. Prakash

AAAI 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments on several diverse real datasets show that it finds cut points matching ground-truth or meaningful external signals outperforming non-trivial baselines.
Researcher Affiliation Academia Sorour E. Amiri, Liangzhe Chen, B. Aditya Prakash Department of Computer Science, Virginia Tech Email: {esorour, liangzhe, badityap}@cs.vt.edu
Pseudocode Yes Alg. 1 shows the final pseudo code of SNAPNETS. Algorithm 2: LAYERED-ALP
Open Source Code Yes Our experiments were conducted on a 4 Xeon E7-4850 CPU with 512GB of 1066Mhz main memory and our code is available for research purposes3. 3http://github.com/Sorour Amiri/Snap NETS
Open Datasets Yes Datasets. We collected a number of datasets from various domains such as social and news media, epidemiology, autonomous system, and co-authorship network to evaluate SNAPNETS. See Tab. 3 for a summary description. ... 4http://topology.eecs.umich.edu/data.html 5http://ndssl.vbi.vt.edu/synthetic-data/
Dataset Splits No No explicit information on dataset splits (e.g., training, validation, test percentages or counts) is provided. The paper mentions using 'datasets' and 'ground truth'.
Hardware Specification Yes Our experiments were conducted on a 4 Xeon E7-4850 CPU with 512GB of 1066Mhz main memory
Software Dependencies No The paper states, 'We implemented SNAPNETS in MATLAB and Python,' but does not provide specific version numbers for these or any other software dependencies.
Experiment Setup Yes We use the same amount of coarsening (ρ = 0.1) as in (Purohit et al. 2014). We extracted multiple standard features (Li et al. 2012) and eliminate correlated ones to get eight features for each Gc i (See Tab. 2 for a description). Finally, we normalize them by range normalization for a meaningful comparison between the features (Li et al. 2012).