Forbidden Nodes Aware Community Search

Authors: Chaokun Wang, Junchao Zhu758-765

AAAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental In this section, extensive experiments are conducted on real data sets to evaluate the three proposed algorithms for the problem of forbidden nodes aware community search, i.e., k-core based FORTE, k-truss based FORTE and CW based FORTE.
Researcher Affiliation Academia Chaokun Wang, Junchao Zhu School of Software, Tsinghua University, Beijing 100084, China chaokun@tsinghua.edu.cn, zhu-jc17@mails.tsinghua.edu.cn
Pseudocode Yes Algorithm 1 k-core based FORTE, Algorithm 2 k-truss based FORTE, Algorithm 3 CW based FORTE
Open Source Code No The paper does not provide any concrete access to source code for the methodology described, nor does it explicitly state that the code is being released.
Open Datasets Yes The data sets are downloaded from the Stanford Large Network Data set Collection (http://snap.stanford.edu/data/).
Dataset Splits No The paper does not provide specific training/validation/test dataset splits. It mentions using 'ground truth communities' and 'test cases' but not in the context of defined splits like percentages or counts.
Hardware Specification Yes The experiments are conducted on a Server with Intel Xeon E5-2650 2.0 GHZ and 256 GB main memory. The Operation System is Windows Server 2008.
Software Dependencies No The paper states 'All the codes are implemented using Python 3.6.1.' but does not list specific version numbers for any libraries, frameworks, or self-contained solvers used in the implementation.
Experiment Setup Yes As for parameters, we set the threshold λ in CW based FORTE to 0.54 according to Figure 3, which presents the f-measure changes from 0.51 to 0.69 on the DBLP data set. The parameter k in the k-core based FORTE is tested from 2 to 10, and then the largest k with valid results is remained, which varies among different tests.