Automated Mining of Relevant N-grams in Relation to Predominant Topics of Text Documents

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F62156489%3A43110%2F15%3A43906663" target="_blank" >RIV/62156489:43110/15:43906663 - isvavai.cz</a>

Výsledek na webu

<a href="http://link.springer.com/chapter/10.1007%2F978-3-319-24033-6_52" target="_blank" >http://link.springer.com/chapter/10.1007%2F978-3-319-24033-6_52</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1007/978-3-319-24033-6_52" target="_blank" >10.1007/978-3-319-24033-6_52</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Automated Mining of Relevant N-grams in Relation to Predominant Topics of Text Documents

Popis výsledku v původním jazyce

The article describes a method focused on the automatic analysis of large collections of short Internet textual documents, freely written in various natural languages and represented as sparse vectors, to reveal whatmulti-word phrases are relevant in relation to a given basic categorization. In addition, the revealed phrases serve for discovering additional different predominant topics, which are not explicitly expressed by the basic categories. Themain idea is to look for n-grams where an n-gram is a collocation of n consecutive words. This leads to the problem of relevant feature selection where a feature is an n-gram that provides more information than an individual word. The feature selection is carried out by entropy minimization which returns a set of combined relevant n-grams and can be used for creating rules, decision trees, or information retrieval. The results are demonstrated for English, German, Spanish, andRussian customer reviews of hotel services publicly available on t

Název v anglickém jazyce

Automated Mining of Relevant N-grams in Relation to Predominant Topics of Text Documents

Popis výsledku anglicky

The article describes a method focused on the automatic analysis of large collections of short Internet textual documents, freely written in various natural languages and represented as sparse vectors, to reveal whatmulti-word phrases are relevant in relation to a given basic categorization. In addition, the revealed phrases serve for discovering additional different predominant topics, which are not explicitly expressed by the basic categories. Themain idea is to look for n-grams where an n-gram is a collocation of n consecutive words. This leads to the problem of relevant feature selection where a feature is an n-gram that provides more information than an individual word. The feature selection is carried out by entropy minimization which returns a set of combined relevant n-grams and can be used for creating rules, decision trees, or information retrieval. The results are demonstrated for English, German, Spanish, andRussian customer reviews of hotel services publicly available on t

Druh

D - Stať ve sborníku

CEP obor

IN - Informatika

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

Název statě ve sborníku

Text, Speech and Dialogue (TSD 2015)

ISBN

978-3-319-24032-9

ISSN

0302-9743

e-ISSN

—

Počet stran výsledku

9

Strana od-do

461-469

Název nakladatele

Springer Switzerland

Místo vydání

Cham

Místo konání akce

Plzeň

Datum konání akce

14. 9. 2015

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

000365947800052