Impact of Surrogate Assessments on High-Recall Retrieval

Adam Roegiest, G. Cormack, C. Clarke, Maura R. Grossman
{"title":"Impact of Surrogate Assessments on High-Recall Retrieval","authors":"Adam Roegiest, G. Cormack, C. Clarke, Maura R. Grossman","doi":"10.1145/2766462.2767754","DOIUrl":null,"url":null,"abstract":"We are concerned with the effect of using a surrogate assessor to train a passive (i.e., batch) supervised-learning method to rank documents for subsequent review, where the effectiveness of the ranking will be evaluated using a different assessor deemed to be authoritative. Previous studies suggest that surrogate assessments may be a reasonable proxy for authoritative assessments for this task. Nonetheless, concern persists in some application domains---such as electronic discovery---that errors in surrogate training assessments will be amplified by the learning method, materially degrading performance. We demonstrate, through a re-analysis of data used in previous studies, that, with passive supervised-learning methods, using surrogate assessments for training can substantially impair classifier performance, relative to using the same deemed-authoritative assessor for both training and assessment. In particular, using a single surrogate to replace the authoritative assessor for training often yields a ranking that must be traversed much lower to achieve the same level of recall as the ranking that would have resulted had the authoritative assessor been used for training. We also show that steps can be taken to mitigate, and sometimes overcome, the impact of surrogate assessments for training: relevance assessments may be diversified through the use of multiple surrogates; and, a more liberal view of relevance can be adopted by having the surrogate label borderline documents as relevant. By taking these steps, rankings derived from surrogate assessments can match, and sometimes exceed, the performance of the ranking that would have been achieved, had the authority been used for training. Finally, we show that our results still hold when the role of surrogate and authority are interchanged, indicating that the results may simply reflect differing conceptions of relevance between surrogate and authority, as opposed to the authority having special skill or knowledge lacked by the surrogate.","PeriodicalId":297035,"journal":{"name":"Proceedings of the 38th International ACM SIGIR Conference on Research and Development in Information Retrieval","volume":"35 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2015-08-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"8","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the 38th International ACM SIGIR Conference on Research and Development in Information Retrieval","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/2766462.2767754","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 8

Abstract

We are concerned with the effect of using a surrogate assessor to train a passive (i.e., batch) supervised-learning method to rank documents for subsequent review, where the effectiveness of the ranking will be evaluated using a different assessor deemed to be authoritative. Previous studies suggest that surrogate assessments may be a reasonable proxy for authoritative assessments for this task. Nonetheless, concern persists in some application domains---such as electronic discovery---that errors in surrogate training assessments will be amplified by the learning method, materially degrading performance. We demonstrate, through a re-analysis of data used in previous studies, that, with passive supervised-learning methods, using surrogate assessments for training can substantially impair classifier performance, relative to using the same deemed-authoritative assessor for both training and assessment. In particular, using a single surrogate to replace the authoritative assessor for training often yields a ranking that must be traversed much lower to achieve the same level of recall as the ranking that would have resulted had the authoritative assessor been used for training. We also show that steps can be taken to mitigate, and sometimes overcome, the impact of surrogate assessments for training: relevance assessments may be diversified through the use of multiple surrogates; and, a more liberal view of relevance can be adopted by having the surrogate label borderline documents as relevant. By taking these steps, rankings derived from surrogate assessments can match, and sometimes exceed, the performance of the ranking that would have been achieved, had the authority been used for training. Finally, we show that our results still hold when the role of surrogate and authority are interchanged, indicating that the results may simply reflect differing conceptions of relevance between surrogate and authority, as opposed to the authority having special skill or knowledge lacked by the surrogate.
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
替代评价对高查全率检索的影响
我们关注的是使用代理评估器来训练被动(即批处理)监督学习方法对文档进行排名以供后续审查的效果,其中排名的有效性将使用被认为是权威的不同评估器进行评估。先前的研究表明,替代评估可能是这项任务的权威评估的合理代理。尽管如此,在某些应用领域(如电子发现),人们仍然担心替代训练评估中的错误会被学习方法放大,从而严重降低性能。通过对先前研究中使用的数据的重新分析,我们证明,在被动监督学习方法中,相对于在训练和评估中使用相同的被认为是权威的评估器,使用替代评估进行训练会严重损害分类器的性能。特别是,使用一个代理来代替权威评估器进行培训,通常会产生一个必须遍历更低的排名,才能达到与使用权威评估器进行培训所产生的排名相同的召回水平。我们还表明,可以采取措施减轻、有时甚至克服替代评估对培训的影响:相关性评估可以通过使用多个替代评估来实现多样化;而且,可以采用一种更自由的相关性观点,即让代理将边缘文档标记为相关。通过采取这些步骤,从替代评估得出的排名可以达到,有时甚至超过,如果将该权威用于培训,将会达到的排名表现。最后,我们表明,当代理和权威的角色互换时,我们的结果仍然成立,这表明结果可能只是反映了代理和权威之间相关性的不同概念,而不是代理缺乏特殊技能或知识的权威。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
自引率
0.00%
发文量
0
期刊最新文献
Regularised Cross-Modal Hashing Adapted B-CUBED Metrics to Unbalanced Datasets Incorporating Non-sequential Behavior into Click Models Time Pressure in Information Search Modeling Multi-query Retrieval Tasks Using Density Matrix Transformation
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1