Fast Network Embedding Enhancement via High Order Proximity Approximation

Authors: Cheng Yang, Maosong Sun, Zhiyuan Liu, Cunchao Tu

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

Reproducibility Variable Result LLM Response
Research Type Experimental We conduct experiments on multi-label classification and link prediction tasks. Experimental results show that NEU can make a consistent and significant improvement over a number of NRL methods with almost negligible running time on all three publicly available datasets.
Researcher Affiliation Academia 1Department of Computer Science and Technology, Tsinghua University, Beijing, China 2Jiangsu Collaborative Innovation Center for Language Ability, Jiangsu Normal University, Xuzhou, China
Pseudocode No The paper describes the NEU algorithm using mathematical equations (Eq. 3 and 5) but does not present a formal pseudocode block or algorithm listing.
Open Source Code Yes The source code of this paper can be obtained from https://github.com/thunlp/NEU.
Open Datasets Yes We conduct experiments on three publicly available datasets: Cora1 [Sen et al., 2008], Blog Catalog and Flickr2 [Tang and Liu, 2011]. 1http://linqs.cs.umd.edu/projects/ /projects/lbc/index.html. 2http://socialcomputing.asu.edu/pages/ datasets.
Dataset Splits Yes For multi-label classification task, we randomly select a portion of vertices as training set and leave the rest as test set. We set the hyperparameters of NEU as follows: λ1 = 0.5, λ2 = 0.25 for all three datasets, T = 3 for Cora and Blog Catalog and T = 1 for Flickr. Here λ1, λ2 are set empirically following the intuition that lower proximity matrix should have a higher weight and T is set as the maximum iteration before the performance on 10% random validation set begins to drop.
Hardware Specification Yes The experiments are executed on a single CPU for the ease of running time comparison and the CPU type is Intel Xeon E5-2620 @ 2.0GHz.
Software Dependencies No The paper mentions using "Lib Linear [Fan et al., 2008]" but does not provide specific version numbers for any software dependencies.
Experiment Setup Yes We set the hyperparameters of NEU as follows: λ1 = 0.5, λ2 = 0.25 for all three datasets, T = 3 for Cora and Blog Catalog and T = 1 for Flickr.