Ranking Users in Social Networks With Higher-Order Structures

Authors: Huan Zhao, Xiaogang Xu, Yangqiu Song, Dik Lun Lee, Zhao Chen, Han Gao

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

Reproducibility Variable Result LLM Response
Research Type Experimental We conduct extensive experiments in three real-world networks, i.e., DBLP, Epinions, and Ciao, to show that MPR can significantly improve the effectiveness of Page Rank for ranking users in social networks.
Researcher Affiliation Collaboration Huan Zhao, 1 Xiaogang Xu, 1 Yangqiu Song, Dik Lun Lee, Zhao Chen, Han Gao Department of Computer Science & Engineering, Hong Kong University of Science and Technology, Hong Kong College of Information Science & Electronic Engineering, Zhejiang University, Hangzhou, China Tencent Technology (SZ) Co., Ltd., China
Pseudocode No The paper describes its method using textual descriptions and mathematical formulas but does not provide structured pseudocode or algorithm blocks.
Open Source Code Yes The code of this work is available at https://github.com/HKUST-Know Comp/Motifbased-Page Rank.
Open Datasets Yes Our experiments are conducted on three real-world networks. The first is a scholar network, DBLP, which is provided by Arnet Miner (Tang et al. 2008). The other two are trust networks, Epinions and Ciao, which are provided by (Tang, Gao, and Liu 2012; Tang et al. 2012).
Dataset Splits No The paper does not provide specific dataset split information (exact percentages, sample counts, citations to predefined splits, or detailed splitting methodology) for training, validation, or testing.
Hardware Specification No The paper does not provide specific hardware details (exact GPU/CPU models, processor types with speeds, memory amounts, or detailed computer specifications) used for running its experiments.
Software Dependencies No The paper mentions using FANMOD for Z-score calculation but does not provide specific version numbers for software dependencies or libraries used in their main implementation.
Experiment Setup Yes In Eq. (3), we have a parameter α to control the balance between edge-based and motif-based higher-order relations. When α = 0, it means we only use higher-order relations for authority computation. When α = 1, it means we only use the original edge-based relations for authority computation. In this section, we show how this parameter affects NDCG performance. ... The best performance of top500 ranking results on Epinion and Ciao and top50 ranking results on Ciao is achieved at α = 0, i.e., using only the higher-order relations.