Tesi etd-01282019-164923 |
Link copiato negli appunti
Tipo di tesi
Tesi di laurea magistrale
Autore
ALIPERTI, ANDREA
URN
etd-01282019-164923
Titolo
A novel fuzzy density-based clustering algorithm for streaming data
Dipartimento
INGEGNERIA DELL'INFORMAZIONE
Corso di studi
COMPUTER ENGINEERING
Relatori
relatore Prof. Marcelloni, Francesco
relatore Ing. Bechini, Alessio
correlatore Dott. Renda, Alessandro
relatore Ing. Bechini, Alessio
correlatore Dott. Renda, Alessandro
Parole chiave
- concept drift
- fuzzy DbScan
- streaming
- temporal decay
Data inizio appello
22/02/2019
Consultabilità
Non consultabile
Data di rilascio
22/02/2089
Riassunto
The aim of this thesis is the deepening of the principal clustering techniques and frameworks to process data in streaming and the implementation of a novel fuzzy density based algorithm .
Thesis evolves in three parts.The first part is dedicated to the study and description of principal Apache open source frameworks used for streaming. The second part is devoted to development and comparison between a crisp and a fuzzy version of a streaming algorithm in absence of temporal fading.
In the last part is discussed an online-offline version of the algorithm with temporal fading . In this part online phase collects some aggregations in form of vector of neighbours and vector of borders and offline phase computes clusters at some time intervals.
Thesis evolves in three parts.The first part is dedicated to the study and description of principal Apache open source frameworks used for streaming. The second part is devoted to development and comparison between a crisp and a fuzzy version of a streaming algorithm in absence of temporal fading.
In the last part is discussed an online-offline version of the algorithm with temporal fading . In this part online phase collects some aggregations in form of vector of neighbours and vector of borders and offline phase computes clusters at some time intervals.
File
| Nome file | Dimensione |
|---|---|
Tesi non consultabile. |
|