A Comparison of Clustering Methods and Prediction Models for Spatial Data
| Autor: | Karamatić, Ivo |
|---|---|
| Přispěvatelé: | Pripužić, Krešimir |
| Jazyk: | chorvatština |
| Rok vydání: | 2022 |
| Předmět: |
usporedba
grupiranje predviđanje predikcija regresija prostorni podaci Python algoritam K-sredina model Gaussove mješavine aglomerativno hijerarhijsko grupiranje algoritam BIRCH linearna regresija robusna regresija bayesovska hrbatna regresija stroj potpornih vektora algoritam slučajnih šuma TEHNIČKE ZNANOSTI. Računarstvo TECHNICAL SCIENCES. Computing BIRCH spatial data support vector machine Bayesian ridge regression Agglomerative hierarchical clustering prediction Kmeans robust regression comparison linear regression regression Gaussian mixture random forest clustering |
| Popis: | U ovom je radu dana usporedba kombinacija različitih metoda grupiranja i modela predviđanja provedena nad prostornim podacima. Teorijski su opisane sve korištene metode grupiranja, kao i korišteni modeli predviđanja. Osim njih, objašnjene su metrike korištene za evaluaciju metoda grupiranja i metrike korištene za evaluaciju modela predviđanja. Zatim su opisani korišteni skup podataka i korišteni programski alati. Konačno, na samom kraju rada dan je pregled rezultata provedene usporedbe. The main goal of this thesis was to compare the performance of different combinations of clustering methods and prediction models for spatial data. Every clustering method that ended up in the practical implementation, as well as every prediction model, was described in the theoretical part of the thesis. Metrics used to evaluate clustering methods and prediction models were described as well. Furthermore, the thesis contains details about the dataset, tools and software libraries necessary for the practical implementation. Finally, the last part of this thesis gives an overview of the obtained results. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|