Random-Radius Ball Method for Estimating Closeness Centrality

Authors: Wataru Inariba, Takuya Akiba, Yuichi Yoshida

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

Reproducibility Variable Result LLM Response
Research Type Experimental The effectiveness of the RRB method over existing algorithms is demonstrated through experiments on real-world networks.
Researcher Affiliation Collaboration Wataru Inariba The University of Tokyo and JST, ERATO, Kawarabayashi Large Graph Project oinari17@gmail.com Takuya Akiba Preferred Networks, Inc. akiba@preferred.jp Yuichi Yoshida National Institute of Informatics and Preferred Infrastructure, Inc. yyoshida@nii.ac.jp
Pseudocode Yes Algorithm 1: Random-radius ball (RRB) method. Algorithm 2: Random-radius ball (RRB) method with bootstrapping.
Open Source Code No The paper does not provide an explicit statement or link for the open-sourcing of the RRB method code. It mentions using a third-party framework: "For HB, we used the Web Graph framework, which is the Java program provided by the authors (Boldi and Vigna 2004)."
Open Datasets Yes Networks. All of the networks were collected from the Stanford Large Network Dataset Collection (Leskovec and Krevl 2014) and Laboratory for Web Algorithms (Boldi and Vigna 2004; Boldi et al. 2011).
Dataset Splits No The paper does not explicitly state the training, validation, or test dataset splits. It mentions using various networks for experiments, but no details on how these were partitioned for evaluation.
Hardware Specification Yes All of the experiments were conducted on a machine with two Intel Xeon E5540 processors and 48 Gi B of main memory.
Software Dependencies Yes We implemented RRB, RRB-BS, and ADS in C++11 and compiled them with gcc 4.8.2. For HB, we used the Web Graph framework, which is the Java program provided by the authors (Boldi and Vigna 2004).
Experiment Setup Yes For RRB-BS, we set s = 3. For all of these methods, we can control the trade-off between the time complexity and accuracy by varying select parameters. Throughout all of the experiments, we estimated the harmonic centrality of the vertices, i.e., the distance-decay function α(x) was set to 1/x. To evaluate the (normalized) RMSE of an estimate, we ran an algorithm 100 times with different random seeds and then took the average.