Filtering outliers to improve classification. First results
The result's identifiers
Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216224%3A14330%2F19%3A00111021" target="_blank" >RIV/00216224:14330/19:00111021 - isvavai.cz</a>
Result on the web
<a href="https://hi.kkui.fei.tuke.sk/daz2019/DaZ_WIKT_2019_Zbornik.pdf" target="_blank" >https://hi.kkui.fei.tuke.sk/daz2019/DaZ_WIKT_2019_Zbornik.pdf</a>
DOI - Digital Object Identifier
—
Alternative languages
Result language
angličtina
Original language name
Filtering outliers to improve classification. First results
Original language description
In this work we describe preliminary experiments for exploring influence of outlier elimination to classifier accuracy. A workflow is a couple consisting of some pre-processing methods - outlier detection and filtering in our case - and a classifier. We describe a method for testing various workflows and bring first results.
Czech name
—
Czech description
—
Classification
Type
D - Article in proceedings
CEP classification
—
OECD FORD branch
10200 - Computer and information sciences
Result continuities
Project
—
Continuities
S - Specificky vyzkum na vysokych skolach
Others
Publication year
2019
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Data specific for result type
Article name in the collection
Data a Znalosti & WIKT 2019
ISBN
9788055333540
ISSN
—
e-ISSN
—
Number of pages
5
Pages from-to
124-128
Publisher name
Fakulta elektrotechniky a informatiky - Technická univerzita v Košiciach
Place of publication
Košice
Event location
Košice, Slovenská republika
Event date
Oct 10, 2019
Type of event by nationality
CST - Celostátní akce
UT code for WoS article
—