Semantic search

{{Short description|Contextual queries}} '''Semantic search''' denotes search with meaning, as distinguished from lexical search where the search engine looks for literal matches of the query words or variants of them, without understanding the overall meaning of the query.{{cite journal |last1=Bast |first1=Hannah |last2=Buchhold |first2=Björn | last3=Haussmann | first3=Elmar | title=Semantic search on text and knowledge bases |journal=Foundations and Trends in Information Retrieval |date=2016 |volume=10 |issue=2–3 |pages=119–271 |doi=10.1561/1500000032 |url=https://www.nowpublishers.com/article/Details/INR-032 |access-date=1 December 2018|url-access=subscription }} Semantic search is an approach to [[information retrieval]] that seeks to improve [[search engine technology|search]] accuracy by understanding [[User_intent|the searcher's intent]] and the [[context (language use)|context]]ual meaning of terms as they appear in the searchable dataspace, whether on the [[World Wide Web|Web]] or within a closed system, to generate more relevant results. Modern semantic search systems often use vector embeddings to represent words, phrases, or documents as numerical vectors, allowing the retrieval engine to measure similarity based on meaning rather than exact keyword matches.{{Cite journal |last=Klampanos |first=Iraklis A. |date=2009-06-02 |title=Manning Christopher, Prabhakar Raghavan, Hinrich Schütze: Introduction to information retrieval |url=https://doi.org/10.1007/s10791-009-9096-x |journal=Information Retrieval |volume=12 |issue=5 |pages=609–612 |doi=10.1007/s10791-009-9096-x |issn=1386-4564|url-access=subscription }}{{Cite journal |last=Kim |first=Bosung |last2=Hong |first2=Taesuk |last3=Ko |first3=Youngjoong |last4=Seo |first4=Jungyun |date=2020 |title=Multi-Task Learning for Knowledge Graph Completion with Pre-trained Language Models |url=https://doi.org/10.18653/v1/2020.coling-main.153 |journal=Proceedings of the 28th International Conference on Computational Linguistics |location=Stroudsburg, PA, USA |publisher=International Committee on Computational Linguistics |doi=10.18653/v1/2020.coling-main.153|doi-access=free }}

Some authors regard semantic search as a set of techniques for retrieving knowledge from richly structured data sources like [[Ontology (information science)|ontologies]] and [[XML]] as found on the [[Semantic Web]].{{cite book|last1=Dong|first1=Hai|title=A survey in semantic search technologies|date=2008|publisher=IEEE|pages=403–408|url=https://www.researchgate.net/publication/224331268|access-date=1 May 2009}} Such technologies enable the formal articulation of [[domain knowledge]] at a high level of expressiveness and could enable the user to specify their intent in more detail at query time.{{cite conference|chapter=Domain Specific Data Retrieval on the Semantic Web|last1=Ruotsalo |first1=T. |title=The Semantic Web: Research and Applications |date=May 2012 |conference=Eswc2012 |series=Lecture Notes in Computer Science |volume=7295 |doi=10.1007/978-3-642-30284-8_35|pages=422–436|isbn=978-3-642-30283-1 |doi-access=free}} The articulation enhances content relevance and depth by including specific places, people, or concepts relevant to the query.

== Models and tools == Tools like Google's [[Knowledge Graph (Google)|Knowledge Graph]] provide structured relationships between entities to enrich query interpretation.Singhal, A. (2012). Introducing the Knowledge Graph: things, not strings. Google Blog. https://blog.google/products/search/introducing-knowledge-graph-things-not/

Models like [[BERT (language model)|BERT]] and Sentence-BERT convert words or sentences into dense vectors for similarity comparison.Reimers, N., & Gurevych, I. (2019). Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks. ''EMNLP 2019''. https://arxiv.org/abs/1908.10084

Semantic ontologies like [[Web Ontology Language]], [[Resource Description Framework]], and [[Schema.org]] organize concepts and relationships, allowing systems to infer related terms and deeper meanings.Bodenreider, O. (2004). The Unified Medical Language System (UMLS): integrating biomedical terminology. ''Nucleic Acids Research'', 32(suppl_1), D267–D270.

Hybrid search models combine [[lexical retrieval]] (e.g., BM25) with [[semantic ranking]] using pretrained transformer models for optimal performance.Lin, J., et al. (2021). Pretrained Transformers for Text Ranking: BERT and Beyond. https://arxiv.org/abs/2010.06467

==See also== *[[List of search engines]] *[[Semantic web]] *[[Semantic unification]] *[[Resource Description Framework]] *[[Natural language search engine]] *[[Semantic query]] *[[Vector database]] *[[Word embeddings]]

==References== {{reflist}}

==External links==

• [https://km.aifb.kit.edu/ws/semsearch08/ Semantic Search 2008 Workshop at ESWC'08]
• [https://km.aifb.kit.edu/ws/semsearch10/ Semantic Search 2010 Workshop at WWW2010]
• [https://web.archive.org/web/20080325224931/http://www.yr-bcn.es/esair08/ Workshop on Exploiting Semantic Annotations in Information Retrieval at ECIR'08].

{{Semantic Web}} {{Internet search}}

[[Category:Internet search engines]] [[Category:Semantic Web]] [[Category:Information retrieval genres]]

{{internet-stub}}

[[de:Semantische Suche]]