Streaming Data Clustering in MOA using the Leader Algorithm

This master thesis presents a novel stream clustering algorithm, called StreamLeader. It presents a way to deliver clustering without the need of resorting to conventional clustering algorithms, like most other algorithms do. We test it, outperforming its state of the art rivals in most of the cases

Bibliographic Details
Author: Andrés Merino, Jaime
Format: master thesis
Publication Date:2015
Country:España
Institution:Universitat Politècnica de Catalunya (UPC)
Repository:UPCommons. Portal del coneixement obert de la UPC
Language:English
OAI Identifier:oai:upcommons.upc.edu:2117/79235
Online Access:https://hdl.handle.net/2117/79235
Access Level:Open access
Keyword:Computer algorithms
StreamLeader
LeaderKernel
stream
clustering
MOA
leader
Hartigan
Clustream
Denstream
Clustree
Network Intrusion
Forest Cover Type
dimensionality
noise
Algorismes computacionals
Àrees temàtiques de la UPC::Informàtica
Description
Summary:This master thesis presents a novel stream clustering algorithm, called StreamLeader. It presents a way to deliver clustering without the need of resorting to conventional clustering algorithms, like most other algorithms do. We test it, outperforming its state of the art rivals in most of the cases