Embedding-based Query Language Models

By Hamed Zamani, W. Bruce Croft. 2016

150 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Information Systems

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Information Systems

Citations in the corpus

Relevance-based Word Embedding (2017, methodology): “For instance, Zamani and Croft [40] presented a set of embedding-based query language models using the query expansion and pseudo-relevance feedback techniques that benefit from the word embedding vectors.”
A Survey of Cross-lingual Word Embedding Models (2017, background): “They also offer support to Information Retrieval tasks (IR) (Zamani & Croft, 2016; Mitra & Craswell, 2017, inter alia) serving as useful features which can link semantics of the query to semantics of the target document collection, even when query terms are not explicitly mentioned in the relevant…”
Neural Ranking Models with Weak Supervision (2017, methodology): “Word embedding vectors have been applied to term re-weighting in IR models [32, 47], query expansion [10, 33, 43], query classification [25, 44], etc.”
BERT-QE: Contextualized Query Expansion for Document Re-ranking (2020, methodology): “Some query expansion methods use word embeddings to find the relevant terms to the query (Diaz et al., 2016; Zamani and Croft, 2016).”
EXS: Explainable Search Using Local Model Agnostic Interpretability (2018, unknown): “…networks have been widely successful in an array of learning tasks including vision, speech and NLP. Recently, the IR community has also seen successful application of deep learning models in document representation [3, 9], query modeling [4, 19, 20] and design of ranking models [6, 10, 11, 16].”
Estimating Embedding Vectors for Queries (2016, methodology): “As suggested by Xu and Croft [36], query expansion techniques can use global or local document analysis.”
Guided Transformer: Leveraging Multiple External Sources for Representation Learning in Conversational Search (2020, methodology): “, we do not use external knowledge bases in our experiments. For example, we use the top retrieved documents as one information source, which is similar to the pseudo-relevance feedback (PRF) methods [2, 14, 27, 60]. We showed that our method significantly outperforms state-of-the-art PRF methods. 3”
Pseudo-Relevance Feedback Based on Matrix Factorization (2016, background): “The weight function can potentially be enhanced using these kinds of information that have been shown to be useful in the PRF task [3, 21, 23, 40].”
Query Expansion Using Word Embeddings (2016, methodology): “The merits of integrating embedding-based query expansion with a relevance model were demonstrated as in our work; and, the merits of using embeddings for relevance model construction were shown [17].”
Critically Examining the "Neural Hype": Weak Baselines and the Additivity of Effectiveness Gains from Neural Ranking Models (2019, background): “While there are much more sophisticated approaches to integrating multiple relevance signals, this approach is commonly used [6, 8, 16, 23, 24].”
Word Embedding Causes Topic Shifting; Exploit Global Context! (2017, methodology): “More recently, the approaches to exploit the advancement in word embedding for IR has shown not only to be competitive to the local approaches but also that combining the approaches brings further improvements in comparison to each of them alone [12, 16, 17].”
Neural information retrieval: at the end of the early years (2017, methodology): “Zamani and Croft (2016b) develop a method for query embedding based on the embedding vectors of its individual words.”
A Deep Look into Neural Ranking Models for Information Retrieval (2019, background): “In this area, embedding-based relevance models [151] extend the original relevance models [121] by considering word embedding vectors.”
Situational Context for Ranking in Personal Search (2017, methodology): “Note that the embedding layers diﬀer from pre-trained word embedding vectors, which have been previously used in [13, 29, 48], in that our embeddings are directly learned from the click training data.”