StructInf: Mining Structural Influence from Social Streams

Authors: Jing Zhang, Jie Tang, Yuanyi Zhong, Yuchen Mo, Juanzi Li, Guojie Song, Wendy Hall, Jimeng Sun

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

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments on a big microblogging dataset show that the proposed sampling algorithms can achieve a 10 speedup compared to the exact influence pattern mining algorithm, with an average error rate of only 1.0%.
Researcher Affiliation Academia Information School, Renmin University of China Department of Computer Science and Technology, Tsinghua University Key Laboratory of Machine Perception, Peking University Electronics and Computer Science, University of Southampton Computational Science and Engineering at College of Computing, Georgia Institute of Technology
Pseudocode Yes Algorithm 1: Struct Inf-Basic" and "Algorithm 2: Enum Inf Pattern
Open Source Code Yes The dataset and code are available online now.3 3http://aminer.org/structinf
Open Datasets Yes The dataset and code are available online now.3 3http://aminer.org/structinf
Dataset Splits No The paper mentions training a classifier and using a balanced dataset but does not provide specific train/validation/test dataset splits (percentages, counts, or references to predefined splits).
Hardware Specification No The paper does not provide specific hardware details such as CPU/GPU models, memory specifications, or cloud instance types used for running experiments.
Software Dependencies No The paper does not specify any software dependencies with version numbers (e.g., Python version, library versions like TensorFlow, PyTorch, scikit-learn).
Experiment Setup Yes The results are obtained by executing Struct Inf-S3 with τ = 25 hours, q = 0.9, px = 0.6 and py = 0.1, where px and py are the node sampling probabilities for estimating xk and yk respectively, and q is for sampling edges, and is the same for estimating xk and yk.