Word classes in computational linguistics and artificial intelligence

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F61989592%3A15210%2F24%3A73628826" target="_blank" >RIV/61989592:15210/24:73628826 - isvavai.cz</a>

Výsledek na webu

<a href="http://dx.doi.org/10.1093/oxfordhb/9780198852889.013.41" target="_blank" >http://dx.doi.org/10.1093/oxfordhb/9780198852889.013.41</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1093/oxfordhb/9780198852889.013.41" target="_blank" >10.1093/oxfordhb/9780198852889.013.41</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Word classes in computational linguistics and artificial intelligence

Popis výsledku v původním jazyce

Word classes, or parts of speech, are groupings of words that share common characteristics, both syntactic and semantic. In the fields of Natural Language Processing and Artificial Intelligence, word classes have been seen as the first level of annotation for text tokens within a text processing pipeline: (POS) tagging. These POS tags were originally conceived as the first step for syntactic processing, but are often used to prepare a stream of unstructured text in for the purpose of natural language understanding. The level of understanding achieved by such a system can be as deep as achieving a semantically motivated representation of meaning or as superficial as recognising the usage of certain types of words that are important for the task at hand. The importance of word classes in these fields is that they allow a level of abstraction that can alleviate data sparsity and are generally quite accurate, especially when considering left and right contexts and other contextual information. Further, they can help in suggesting the function of a token within a sentence, as well as suggesting its semantic type (such as event, attribute or referent). Scientists and practitioners in NLP and AI are consumers of word classes. However, the field of Computational Linguistics has slightly different goals that overlap more with the interest of linguists. Computational linguists ask questions that a linguist might about word classes but the answers to these questions are arrived at either by methods employed in Grammar Engineering or big data methods employed in Natural Language Processing.

Název v anglickém jazyce

Word classes in computational linguistics and artificial intelligence

Popis výsledku anglicky

Word classes, or parts of speech, are groupings of words that share common characteristics, both syntactic and semantic. In the fields of Natural Language Processing and Artificial Intelligence, word classes have been seen as the first level of annotation for text tokens within a text processing pipeline: (POS) tagging. These POS tags were originally conceived as the first step for syntactic processing, but are often used to prepare a stream of unstructured text in for the purpose of natural language understanding. The level of understanding achieved by such a system can be as deep as achieving a semantically motivated representation of meaning or as superficial as recognising the usage of certain types of words that are important for the task at hand. The importance of word classes in these fields is that they allow a level of abstraction that can alleviate data sparsity and are generally quite accurate, especially when considering left and right contexts and other contextual information. Further, they can help in suggesting the function of a token within a sentence, as well as suggesting its semantic type (such as event, attribute or referent). Scientists and practitioners in NLP and AI are consumers of word classes. However, the field of Computational Linguistics has slightly different goals that overlap more with the interest of linguists. Computational linguists ask questions that a linguist might about word classes but the answers to these questions are arrived at either by methods employed in Grammar Engineering or big data methods employed in Natural Language Processing.

Druh

C - Kapitola v odborné knize

CEP obor

—

OECD FORD obor

60203 - Linguistics

Projekt

—

Návaznosti

I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Rok uplatnění

2024

Kód důvěrnosti údajů

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

Název knihy nebo sborníku

The Oxford Handbook of Word Classes

ISBN

978-0-19-885288-9

Počet stran výsledku

17

Strana od-do

930-946

Počet stran knihy

1136

Název nakladatele

Oxford University Press

Místo vydání

Oxford

Kód UT WoS kapitoly

—