Text Mining-based Formation of Dictionaries Expressing Opinions in Natural Languages
The result's identifiers
Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F62156489%3A43110%2F11%3A00173333" target="_blank" >RIV/62156489:43110/11:00173333 - isvavai.cz</a>
Result on the web
—
DOI - Digital Object Identifier
—
Alternative languages
Result language
angličtina
Original language name
Text Mining-based Formation of Dictionaries Expressing Opinions in Natural Languages
Original language description
Automatic formation of dictionaries containing words significant for expressing different customers' opinions written in natural languages is demonstrated. The research used very large real-world data concerning the hotel accommodation booking via the Internet. The hotel companies could be interested in characteristic words expressing positive and negative opinions because it could help improve the offered service. The suggested method uses unstructured plain text reviews of many customers from different countries. The data is transformed into vectors using the bag-of-words procedure with the word representation by their frequencies in the reviews. Significant words are selected as relevant attributes for the classification to given categories using trained decision trees. Each tree branch leading to a leaf represents a subset of significant words for a category. The individual word importance is weighted by the word frequency in all the branches combined with their occurrence in branc
Czech name
—
Czech description
—
Classification
Type
J<sub>x</sub> - Unclassified - Peer-reviewed scientific article (Jimp, Jsc and Jost)
CEP classification
IN - Informatics
OECD FORD branch
—
Result continuities
Project
—
Continuities
Z - Vyzkumny zamer (s odkazem do CEZ)
Others
Publication year
2011
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Data specific for result type
Name of the periodical
Mendel
ISSN
1803-3814
e-ISSN
—
Volume of the periodical
—
Issue of the periodical within the volume
1
Country of publishing house
CZ - CZECH REPUBLIC
Number of pages
8
Pages from-to
374-381
UT code for WoS article
—
EID of the result in the Scopus database
—