A Markov random field model for term dependencies
- Donald Metzler, W. Bruce Croft
- Computer ScienceAnnual International ACM SIGIR Conference on…
- 15 August 2005
A novel approach is developed to train the model that directly maximizes the mean average precision rather than maximizing the likelihood of the training data, and significant improvements are possible by modeling dependencies, especially on the larger web collections.
Search Engines - Information Retrieval in Practice
- W. Bruce Croft, Donald Metzler, Trevor Strohman
- Computer Science
- 16 February 2009
This text provides the background and tools needed to evaluate, compare and modify search engines and numerous programming exercises make extensive use of Galago, a Java-based open source search engine.
Indri : A language-model based search engine for complex queries ( extended version )
- Trevor Strohman, Donald Metzler, Howard R. Turtle, W. Bruce Croft
- Computer Science
- 2005
The Indri system is described and it is shown how the query language is designed to support modern language technologies and results demonstrating that Indri is both effective and efficient are presented.
Linear feature-based models for information retrieval
- Donald Metzler, W. Bruce Croft
- Computer ScienceInformation retrieval (Boston)
- 1 June 2007
This paper details supervised training algorithms that directly maximize the evaluation metric under consideration, such as mean average precision, and shows that linear feature-based models can consistently and significantly outperform current state of the art retrieval models with the correct choice of features.
Efficient Transformers: A Survey
- Yi Tay, M. Dehghani, Dara Bahri, Donald Metzler
- Computer ScienceACM Computing Surveys
- 14 September 2020
This article characterizes a large and thoughtful selection of recent efficiency-flavored “X-former” models, providing an organized and comprehensive overview of existing work and models across multiple domains.
Position Bias Estimation for Unbiased Learning to Rank in Personal Search
- Xuanhui Wang, Nadav Golbandi, Michael Bendersky, Donald Metzler, Marc Najork
- Computer ScienceWeb Search and Data Mining
- 2 February 2018
This paper proposes a regression-based Expectation-Maximization (EM) algorithm that is based on a position bias click model and that can handle highly sparse clicks in personal search and compares the pointwise and pairwise learning-to-rank models.
Synthesizer: Rethinking Self-Attention in Transformer Models
- Yi Tay, Dara Bahri, Donald Metzler, Da-Cheng Juan, Zhe Zhao, Che Zheng
- Computer ScienceInternational Conference on Machine Learning
- 2 May 2020
The true importance and contribution of the dot product-based self-attention mechanism on the performance of Transformer models is investigated and a model that learns synthetic attention weights without token-token interactions is proposed, called Synthesizer.
Latent concept expansion using markov random fields
- Donald Metzler, W. Bruce Croft
- Computer ScienceAnnual International ACM SIGIR Conference on…
- 23 July 2007
A robust query expansion technique based on the Markov random field model for information retrieval, called latent concept expansion, provides a mechanism for modeling term dependencies during expansion and the use of arbitrary features within the model provides a powerful framework for going beyond simple term occurrence features.
Improving the estimation of relevance models using large external corpora
- Fernando Diaz, Donald Metzler
- Computer ScienceAnnual International ACM SIGIR Conference on…
- 6 August 2006
The results show that using a high quality corpus that is comparable to the evaluation corpus can be as, if not more, effective than using the web.
Combining the language model and inference network approaches to retrieval
- Donald Metzler, W. Bruce Croft
- Computer ScienceInformation Processing & Management
- 1 September 2004
...
...