Semantic Vector Encoding and Similarity Search Using Fulltext Search Engines

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F03892620%3A_____%2F17%3AN0000005" target="_blank" >RIV/03892620:_____/17:N0000005 - isvavai.cz</a>

Nalezeny alternativní kódy

RIV/00216224:14330/17:00094366

Výsledek na webu

<a href="http://www.aclweb.org/anthology/W17-2611" target="_blank" >http://www.aclweb.org/anthology/W17-2611</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.18653/v1/W17-2611" target="_blank" >10.18653/v1/W17-2611</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Semantic Vector Encoding and Similarity Search Using Fulltext Search Engines

Popis výsledku v původním jazyce

Vector representations and vector space modeling (VSM) play a central role in modern machine learning. We propose a novel approach to ‘vector similarity searching’ over dense semantic representations of words and documents that can be deployed on top of traditional inverted-index-based fulltext engines, taking advantage of their robustness, stability, scalability and ubiquity. We show that this approach allows the indexing and querying of dense vectors in text domains. This opens up exciting avenues for major efficiency gains, along with simpler deployment, scaling and monitoring. The end result is a fast and scalable vector database with a tunable trade-off between vector search performance and quality, backed by a standard fulltext engine such as Elasticsearch. We empirically demonstrate its querying performance and quality by applying this solution to the task of semantic searching over a dense vector representation of the entire English Wikipedia.

Název v anglickém jazyce

Semantic Vector Encoding and Similarity Search Using Fulltext Search Engines

Popis výsledku anglicky

Vector representations and vector space modeling (VSM) play a central role in modern machine learning. We propose a novel approach to ‘vector similarity searching’ over dense semantic representations of words and documents that can be deployed on top of traditional inverted-index-based fulltext engines, taking advantage of their robustness, stability, scalability and ubiquity. We show that this approach allows the indexing and querying of dense vectors in text domains. This opens up exciting avenues for major efficiency gains, along with simpler deployment, scaling and monitoring. The end result is a fast and scalable vector database with a tunable trade-off between vector search performance and quality, backed by a standard fulltext engine such as Elasticsearch. We empirically demonstrate its querying performance and quality by applying this solution to the task of semantic searching over a dense vector representation of the entire English Wikipedia.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

20205 - Automation and control systems

Projekt

<a href="/cs/project/TD03000295" target="_blank" >TD03000295: Inteligentní software pro sémantické hledání dokumentů</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2017

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Proceedings of the 2nd Workshop on Representation Learning for NLP

ISBN

978-1-945626-62-3

ISSN

—

e-ISSN

—

Počet stran výsledku

10

Strana od-do

81-90

Název nakladatele

Association for Computational Linguistics

Místo vydání

Vancouver, Canada

Místo konání akce

Vancouver, Canada

Datum konání akce

1. 1. 2017

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—