Active exploration for learning rankings from clickthrough data
Active exploration for learning rankings from clickthrough data
复制标题
DOI：
10.1145/1281192.1281254
复制
发表时间：
2007-08
期刊：
影响因子：
--
通讯作者：
Filip Radlinski;T. Joachims
中科院分区：
文献类型：
--
作者：
Filip Radlinski;T. Joachims
我们解决了从用户行为的搜索引擎中学习文档排名的任务。之前对这个问题的研究依赖于被动收集的点击量数据。相反，我们表明主动探索策略可以提供数据，从而更快地学习。具体来说，我们开发了一种贝叶斯方法来选择排名以呈现用户，以便交互产生更多信息的训练数据。我们使用TREC-10 Web语料库以及合成数据的结果表明，在在线学习设置中，定向探索策略可以快速地为用户提供改进的排名。我们发现主动勘探实质上优于被动观察和随机勘探。
We address the task of learning rankings of documents from search enginelogs of user behavior. Previous work on this problem has relied onpassively collected clickthrough data. In contrast, we show that anactive exploration strategy can provide data that leads to much fasterlearning. Specifically, we develop a Bayesian approach for selectingrankings to present users so that interactions result in more informativetraining data. Our results using the TREC-10 Web corpus, as well assynthetic data, demonstrate that a directed exploration strategy quicklyleads to users being presented improved rankings in an online learningsetting. We find that active exploration substantially outperformspassive observation and random exploration.