Analysis of Crowdsourced Sampling Strategies for HodgeRank with Sparse Random Graphs

@article{Osting2015AnalysisOC,
  title={Analysis of Crowdsourced Sampling Strategies for HodgeRank with Sparse Random Graphs},
  author={Braxton Osting and Jiechao Xiong and Q. Xu and Y. Yao},
  journal={ArXiv},
  year={2015},
  volume={abs/1503.00164}
}
  • Braxton Osting, Jiechao Xiong, +1 author Y. Yao
  • Published 2015
  • Mathematics, Computer Science
  • ArXiv
  • Abstract Crowdsourcing platforms are now extensively used for conducting subjective pairwise comparison studies. In this setting, a pairwise comparison dataset is typically gathered via random sampling, either with or without replacement. In this paper, we use tools from random graph theory to analyze these two random sampling methods for the HodgeRank estimator. Using the Fiedler value of the graph as a measurement for estimator stability (informativeness), we provide a new estimate of the… CONTINUE READING
    Parsimonious Mixed-Effects HodgeRank for Crowdsourced Preference Aggregation
    5
    Spectral Method and Regularized MLE Are Both Optimal for Top-$K$ Ranking
    43
    Towards Spectral Sparsification of Simplicial Complexes Based on Generalized Effective Resistance
    5
    Spectral MLE: Top-K Rank Aggregation from Pairwise Comparisons
    91
    Active Learning for Crowdsourced QoE Modeling
    7

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 63 REFERENCES
    HodgeRank on Random Graphs for Subjective Video Quality Assessment
    52
    Optimal data collection for informative rankings expose well-connected graphs
    12
    Pairwise ranking aggregation in a crowdsourced setting
    239
    Statistical ranking and combinatorial Hodge theory
    220
    Enhanced statistical rankings via targeted data collection
    24
    Iterative ranking from pair-wise comparisons
    194
    Random partial paired comparison for subjective video quality assessment via hodgerank
    31
    Active Ranking using Pairwise Comparisons
    166
    Active Ranking in Practice: General Ranking Functions with Sample Complexity Bounds
    5
    A Statistical Convergence Perspective of Algorithms for Rank Aggregation from Pairwise Data
    113