Mining Heavy Temporal Subgraphs: Fast Algorithms and Applications

Authors: Jose Cadena, Anil Vullikanti

AAAI 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We evaluate our algorithms in a diverse set of real and synthetic networks, and we find solutions with higher score and better detection power for anomalous events compared to earlier heuristics.
Researcher Affiliation Academia Jose Cadena, Anil Vullikanti Department of Computer Science and Biocomplexity Institute, Virginia Tech, Blacksburg, VA 24061 {jcadena,vsakumar}@vt.edu
Pseudocode Yes Algorithm 1 SELECTINTERVALS(t): Produce set of O(t log t) intervals.; Algorithm 2 FASTGW(H, w, π); Algorithm 3 FASTGW-HDS: Find a solution to HDS
Open Source Code No The paper links to a full version PDF, not source code. There is no explicit statement about releasing source code for the methodology.
Open Datasets Yes We evaluate our algorithms on five real networks (see Table 2): (i) a Twitter follower graph with communication evolving over 6 months, (ii) the highway network of Los Angeles County, California and its activity on May, 20141, (iii) a contact network of health-care workers and patients in a hospital ward during 4 days (Vanhems et al. 2013), (iv) a sample of Wikipedia page view statistics, and (v) an autonomous systems network. ... 1http://pems.dot.ca.gov/ ... Ramakrishnan et al. 2014; Vanhems et al. 2013
Dataset Splits No The paper mentions using real and synthetic networks but does not provide specific details on training, validation, or test dataset splits, percentages, or cross-validation setup.
Hardware Specification No The paper does not specify any hardware details (e.g., GPU/CPU models, memory) used for running the experiments.
Software Dependencies No The paper describes algorithms but does not mention any specific software components with version numbers (e.g., Python 3.x, PyTorch x.x, CPLEX x.x).
Experiment Setup No The paper discusses the experimental setup in terms of datasets and comparisons, but does not provide specific details on hyperparameters, training configurations, or system-level settings.