Gender stereotype reinforcement: Measuring the gender bias conveyed by ranking algorithms
作者：
Highlights：
• Gender Stereotype Reinforcement (GSR) measure tailored for Search Engines.
• Evaluation of GSR within the construct validity framework.
• Audit, in terms of GSR, of several widely-known and used ranking algorithm.
• Estimation of the impact of different Word Embedding debiasing approaches both on ranking effectiveness and countering gender bias.
• Quantitative and qualitative analysis showing suitability of shared IR test collection to analyze gender stereotype reinforcement.
摘要
•Gender Stereotype Reinforcement (GSR) measure tailored for Search Engines.•Evaluation of GSR within the construct validity framework.•Audit, in terms of GSR, of several widely-known and used ranking algorithm.•Estimation of the impact of different Word Embedding debiasing approaches both on ranking effectiveness and countering gender bias.•Quantitative and qualitative analysis showing suitability of shared IR test collection to analyze gender stereotype reinforcement.
论文关键词：Fairness,Gender stereotypes,Information retrieval,Search engines,Word embeddings
论文评审过程：Received 15 May 2020, Revised 20 August 2020, Accepted 23 August 2020, Available online 3 September 2020, Version of Record 20 October 2020.
论文官网地址：https://doi.org/10.1016/j.ipm.2020.102377
