Recognition of Emotions in Czech Newspaper Headlines
The result's identifiers
Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26220%2F11%3APU91508" target="_blank" >RIV/00216305:26220/11:PU91508 - isvavai.cz</a>
Result on the web
—
DOI - Digital Object Identifier
—
Alternative languages
Result language
angličtina
Original language name
Recognition of Emotions in Czech Newspaper Headlines
Original language description
With the growth of internet community, many different text-based documents are produced. Emotion detection and classification in text becomes very important in human-machine interaction or in human-to-human internet communication with this growth. This article refers to this issue in Czech texts. Headlines were extracted from Czech newspapers and Fear, Joy, Anger, Disgust, Sadness, and Surprise emotions are detected. In this work, several algorithms for learning were assessed and compared according to their accuracy of emotion detection and classification of news headlines. The best results were achieved using the SVM (Support Vector Machine) method with a linear kernel, where the presence of the dominant emotion or emotions was analyzed. For individual emotions the following results were obtained: Anger was detected in 87.3 %, Disgust 95.01%, Fear 81.32 %, Joy 71.6 %, Sadness 75.4 %, and Surprise 71.09 %.
Czech name
—
Czech description
—
Classification
Type
J<sub>x</sub> - Unclassified - Peer-reviewed scientific article (Jimp, Jsc and Jost)
CEP classification
BD - Information theory
OECD FORD branch
—
Result continuities
Project
<a href="/en/project/FR-TI2%2F679" target="_blank" >FR-TI2/679: *Media-informatics system supporting advanced multimedia services</a><br>
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
Radioengineering
ISSN
1210-2512
e-ISSN
—
Volume of the periodical
2011
Issue of the periodical within the volume
1
Country of publishing house
CZ - CZECH REPUBLIC
Number of pages
9
Pages from-to
1-9
UT code for WoS article
—
EID of the result in the Scopus database
—