Improving PageRank using sports results modeling
作者:
Highlights:
• PageRank is often applied on sports data to rank teams or individuals.
• We set up and calibrate a model to create synthetic sports results.
• Synthetic data are used to assess various ranking algorithms.
• We find that PageRank performs well only when there is little randomness in data.
• We propose a PageRank variant that outperforms PageRank in all evaluated settings.
摘要
•PageRank is often applied on sports data to rank teams or individuals.•We set up and calibrate a model to create synthetic sports results.•Synthetic data are used to assess various ranking algorithms.•We find that PageRank performs well only when there is little randomness in data.•We propose a PageRank variant that outperforms PageRank in all evaluated settings.
论文关键词:Networks,Modeling sports results,Ranking in sport,Stochastic modeling,PageRank
论文评审过程:Received 16 September 2021, Revised 1 December 2021, Accepted 4 January 2022, Available online 17 January 2022, Version of Record 5 February 2022.
论文官网地址:https://doi.org/10.1016/j.knosys.2022.108168