论文信息 - Web-Scale Semantic Ranking

Web-Scale Semantic Ranking

Semantic ranking models go beyond keyword matching to score documents based on closeness in meaning to the query. The use of semantic ranking in Web search has been limited due to the high cost of these models. To address this issue, we have designed and implemented a new Web-scale ranking system that enables us to integrate semantic ranking techniques into a commercial search engine. We have explored several types of models and will describe our implementation of translation models (TM) in this paper. The experiments demonstrate that these models significantly improve relevance over our existing baseline system. Our new ranking system is deployed online and is currently serving many millions of users.

[1] Jian-Yun Nie,et al. Using query contexts in information retrieval , 2007, SIGIR.

[2] Thomas Hofmann,et al. Probabilistic latent semantic indexing , 1999, SIGIR '99.

[3] W. Bruce Croft,et al. A language modeling approach to information retrieval , 1998, SIGIR '98.

[4] Robert L. Mercer,et al. The Mathematics of Statistical Machine Translation: Parameter Estimation , 1993, CL.

[5] Jian-Yun Nie,et al. Query expansion using term relationships in language models for information retrieval , 2005, CIKM '05.

[6] Jian-Yun Nie,et al. Integrating word relationships into language models , 2005, SIGIR '05.

[7] John D. Lafferty,et al. Information Retrieval as Statistical Translation , 2017 .

[8] Jianfeng Gao,et al. Clickthrough-based translation models for web search: from word models to phrase models , 2010, CIKM.

[9] Jaana Kekäläinen,et al. Cumulated gain-based evaluation of IR techniques , 2002, TOIS.

[10] Gregory N. Hullender,et al. Learning to rank using gradient descent , 2005, ICML.

[11] Michael I. Jordan,et al. Latent Dirichlet Allocation , 2001, J. Mach. Learn. Res..