Classification of Complex Ecological Objects with the Use of Information Granules
Fragment książki (Materiały konferencyjne)
MNiSW
140
konferencja
| Status: | |
| Autorzy: | Kiersztyn Adam, Kiersztyn Krystyna, Karczmarek Paweł, Kamiński Marek, Kitowski Ignacy, Zbyryt Adam, Łopucki Rafał, Pitucha Grzegorz, Pedrycz Witold |
| Dyscypliny: | |
| Aby zobaczyć szczegóły należy się zalogować. | |
| Wersja dokumentu: | Drukowana | Elektroniczna |
| Język: | angielski |
| Strony: | 1 - 6 |
| Web of Science® Times Cited: | 0 |
| Scopus® Cytowania: | 2 |
| Bazy: | Web of Science | Scopus |
| Efekt badań statutowych | NIE |
| Finansowanie: | The work was co-financed by the Lublin University of Technology Scientific Fund: FD-ITIT-KIER. |
| Materiał konferencyjny: | TAK |
| Nazwa konferencji: | IEEE International Conference on Fuzzy Systems (FUZZ-IEEE) 2021 |
| Skrócona nazwa konferencji: | FUZZ-IEEE 2021 |
| URL serii konferencji: | LINK |
| Termin konferencji: | 11 lipca 2021 do 14 lipca 2021 |
| Miasto konferencji: | Virtual Conference |
| Państwo konferencji: | LUKSEMBURG |
| Publikacja OA: | NIE |
| Abstrakty: | angielski |
| The selection of an appropriate method of data analysis is a key problem for researchers from various fields of applications. They consider different methods of data classification,often based on the thematic scope of the data at their disposal.However, various data characteristics, such as data set size, datatype and quality, gaps, outliers and other anomalies, can makeproper selection significantly difficult. Therefore, in this study wepropose a method based on a very universal classifier designedon the basis of calculations using information granules. The mainobjective of the work is to present and comprehensively verifythe effectiveness of the classifier. As an example of application,we propose complicated yet currently important data comingfrom widely understood ecological research. Detailed numerical experiments indicate the high efficiency of the proposed methodand the possibility of easy application to data appearing in otherfields. In addition, various types of aggregation functions of theclassification results are considered in order to obtain the mostreliable results for the discussed problems. |
