Data Imputation in Related Time Series Using Fuzzy Set-Based Techniques
Fragment książki (Materiały konferencyjne)
MNiSW
140
konferencja
| Status: | |
| Autorzy: | Kiersztyn Adam, Karczmarek Paweł, Łopucki Rafał, Pedrycz Witold, Al Ebru, Kitowski Ignacy, Zbyryt Adam |
| Dyscypliny: | |
| Aby zobaczyć szczegóły należy się zalogować. | |
| Wersja dokumentu: | Drukowana | Elektroniczna |
| Język: | angielski |
| Strony: | 1 - 8 |
| Web of Science® Times Cited: | 4 |
| Scopus® Cytowania: | 9 |
| Bazy: | Web of Science | Scopus | IEEE Xplore |
| Efekt badań statutowych | NIE |
| Finansowanie: | Funded by the National Science Centre, Poland under CHIST-ERA programme (Grant no. 2018/28/Z/ST6/00563). |
| Materiał konferencyjny: | TAK |
| Nazwa konferencji: | IEEE International Conference onFuzzy Systems (FUZZ-IEEE) 2020 |
| Skrócona nazwa konferencji: | FUZZ-IEEE 2020 |
| URL serii konferencji: | LINK |
| Termin konferencji: | 19 lipca 2020 do 24 lipca 2020 |
| Miasto konferencji: | Glasgow |
| Państwo konferencji: | WIELKA BRYTANIA |
| Publikacja OA: | NIE |
| Abstrakty: | angielski |
| One of the main challenges faced by people who use data from empirical research in their work is missing data. In many scientific disciplines and industries there are references to time series. The suitability of several methods to imputation of the missing data in the study of mutual links between the analysed time series have been presented and tested in this work. In this paper, known methods of supplementing data in time series were enriched by the use of fuzzy sets and their processing was tested on unique data from experimental research and a transport company database. Fuzzy linguistic descriptors-based methods of missing data imputation in databases containing time series are discussed. The proposed method has a high efficiency, which have been proven in a series of experiments with both artificial and real datasets. The proposed methodologies have been tested on theoretical example and empirical data sets from various fields: (1) ecological data on changes in bird arrival dates in the context of climate change and (2) data describing the transport of containers between ports on the Mediterranean. Moreover, an important novelty of this work is, in particular, an application of fuzzy techniques to the correction of the datasets containing bird migration descriptions. |
