Using word embeddings for analysing texts from the educational domain
The result's identifiers
Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F62156489%3A43110%2F18%3A43913697" target="_blank" >RIV/62156489:43110/18:43913697 - isvavai.cz</a>
Result on the web
<a href="http://pefnet.mendelu.cz/wcd/w-rek-pefnet/pefnet17_fin.pdf" target="_blank" >http://pefnet.mendelu.cz/wcd/w-rek-pefnet/pefnet17_fin.pdf</a>
DOI - Digital Object Identifier
—
Alternative languages
Result language
angličtina
Original language name
Using word embeddings for analysing texts from the educational domain
Original language description
The application of machine learning to natural language data is very attractive for a data scientist. Modern algorithms and text representations allow the discovery of information about the semantic content of texts with more possibilities than before. The article deals with algorithms for word embeddings - namely word2vec and fastText. The goal was to analyse Facebook posts from the pages of universities in the Czech Republic. After creating and querying the natural language models, we were able to discover what is the most interesting information and the relations for the names of universities, job opportunities, specialisations, events, freshmen, or out of school activities.
Czech name
—
Czech description
—
Classification
Type
D - Article in proceedings
CEP classification
—
OECD FORD branch
10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)
Result continuities
Project
—
Continuities
S - Specificky vyzkum na vysokych skolach
Others
Publication year
2018
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Data specific for result type
Article name in the collection
PEFnet 2017: Proceedings
ISBN
978-80-7509-555-8
ISSN
—
e-ISSN
neuvedeno
Number of pages
8
Pages from-to
129-136
Publisher name
Mendelova univerzita v Brně
Place of publication
Brno
Event location
Brno
Event date
Nov 30, 2017
Type of event by nationality
EUR - Evropská akce
UT code for WoS article
—