Distributional Semantics in Language Modeling

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F49777513%3A23520%2F15%3A43925553" target="_blank" >RIV/49777513:23520/15:43925553 - isvavai.cz</a>

Výsledek na webu

—

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Distributional Semantics in Language Modeling

Popis výsledku v původním jazyce

Language models are crucial for many tasks in natural language processing and n-grams are probably the best way to build them. Huge effort is being invested in improving the n-gram language models. By introducing external knowledge (morphology, syntax, etc.) into the models, a significant improvement can be achieved. The models can, however, be improved without external knowledge and the better smoothing is an excellent example of such improvement. By discovering hidden patterns in unlabeled training corpora, we can enhance the language modeling with the information that is already present in the corpora. This thesis studies three different ways of latent information discovery. Global semantics is modeled by latent Dirichlet allocation and brings long-range dependencies into language models. Word clusters given by semantic spaces enrich these language models with short-range semantics. Finally, our own unsupervised stemming algorithm is used to further enhance the performance of langua

Název v anglickém jazyce

Distributional Semantics in Language Modeling

Popis výsledku anglicky

Language models are crucial for many tasks in natural language processing and n-grams are probably the best way to build them. Huge effort is being invested in improving the n-gram language models. By introducing external knowledge (morphology, syntax, etc.) into the models, a significant improvement can be achieved. The models can, however, be improved without external knowledge and the better smoothing is an excellent example of such improvement. By discovering hidden patterns in unlabeled training corpora, we can enhance the language modeling with the information that is already present in the corpora. This thesis studies three different ways of latent information discovery. Global semantics is modeled by latent Dirichlet allocation and brings long-range dependencies into language models. Word clusters given by semantic spaces enrich these language models with short-range semantics. Finally, our own unsupervised stemming algorithm is used to further enhance the performance of langua

Druh

O - Ostatní výsledky

CEP obor

JD - Využití počítačů, robotika a její aplikace

OECD FORD obor

—

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2015

Kód důvěrnosti údajů

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