Semantic Matching in Search

By Hang Li, Jun Xu. 2014

215 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Information Systems, Information Retrieval & Search, Natural Language Processing

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

In Information Retrieval & Search

In Natural Language Processing

Citations in the corpus

Word-Entity Duet Representations for Document Ranking (2017, background): “Synonyms and related concepts stored in these resources can be added to queries and documents to reduce language varieties and may improve the recall of search results [12].”
A Deep Architecture for Semantic Matching with Multiple Positional Sentence Representations (2015, background): “Semantic matching is a critical task for many applications in natural language processing (NLP), such as information retrieval (Li and Xu 2013), question answering (Berger et al. 2000) and paraphrase identiﬁcation (Dolan, Quirk, and Brockett 2004).”
Embedding-based Retrieval in Facebook Search (2020, background): “search techniques are mostly based on various term matching methods [1], which performs well for the cases that keyword match can address. It still remains a challenging problem for semantic matching [12], which is to address desired results that are not exact match of the query text but can satisfy”
Beyond 512 Tokens: Siamese Multi-depth Transformer-based Hierarchical Encoder for Document Matching (2020, background): “Research on learning models for matching between objects from two different spaces or learning to match can potentially benefit a large family of applications and tasks [16].”
Deep Learning for Matching in Search and Recommendation (2018, methodology): “Previously, machine learning methods have been exploited to address the problem, which learns a matching function from labeled data, also referred to as “learning to match” [17].”
EsdRank: Connecting Query and Documents through External Semi-Structured Data (2015, background): “There is a long history of using external semi-structured data to improve full-text search [2, 10, 18, 20, 22, 29].”
DiPair: Fast and Accurate Distillation for Trillion-ScaleText Matching and Pair Modeling (2020, background): “For example, modeling the relationship between queries and documents / ad keywords is central to search engines / digital advertisement systems (Li and Xu, 2014; Guo et al., 2019).”
Short Text Similarity with Word Embeddings (2015, background): “Being able to do so successfully is beneficial in many settings in information retrieval like search [26], query suggestion [30], automatic summarization [3] and image finding [12].”
Neural information retrieval: at the end of the early years (2017, background): “Learn to matchLearning to match (Li and Xu 2013) is the problem of learning a matching function f(x, y) that computes a degree of similarity between two objects x and y from two different spaces X and Y.”
Document Retrieval Using Entity-Based Language Models (2016, background): “To address the potential vocabulary mismatch between a short query and documents relevant to the query, various semantic document-query similarity measures have been proposed [28].”
Generalizing Translation Models in the Probabilistic Relevance Framework (2016, background): “Recently, Li and Xu [20] published a survey on the topic, grouping the various approaches into 5 categories: 1 Both Topic Modelling and Latent Space Models are still to be conclusively proven competitive in terms of both efﬁciency and effectiveness with probabilistic and language models.”
Optimizing Dense Retrieval Model Training with Hard Negatives (2021, background): “To better deduce the users’ search intent and retrieve relevant items, the ranking algorithms are expected to conduct semantic matching between queries and documents [17], which is a challenging problem.”
Match-SRNN: Modeling the Recursive Matching Structure with Spatial RNN (2016, methodology): “However, it is usually difficult for these methods to model the complicated interaction relationship between two texts [Lu and Li, 2013] because the representations are calculated independently.”
Text Matching as Image Recognition (2016, background): “Matching two texts is central to many natural language applications, such as machine translation (Brown et al. 1993), question and answering (Xue, Jeon, and Croft 2008), paraphrase identiﬁcation (Socher et al. 2011) and document retrieval (Li and Xu 2014).”
Learning to Respond with Deep Neural Networks for Retrieval-Based Human-Computer Conversation System (2016, methodology): “A series of information retrieval-based methods are applied to short text conversation using microblog data [9, 14, 17, 16].”