Improved CURE clustering for big data using Hadoop and Mapreduce

In the Era of Information, Extracting useful information out of massive amount of data and process them in less span of time has become crucial part of Data mining. CURE is very useful hierarchical algorithm which has ability to identify cluster of arbitrary shape and able to identify outliers. In t...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:2016 International Conference on Inventive Computation Technologies (ICICT) Ročník 3; s. 1 - 5
Hlavní autoři: Lathiya, Piyush, Rani, Rinkle
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 01.08.2016
Témata:
Algorithm design and analysis
Classification algorithms
Clustering
Clustering algorithms
CURE
Data mining
Distributed databases
Hadoop
Mapreduce
Outliers
Partitioning algorithms
Shape
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:In the Era of Information, Extracting useful information out of massive amount of data and process them in less span of time has become crucial part of Data mining. CURE is very useful hierarchical algorithm which has ability to identify cluster of arbitrary shape and able to identify outliers. In this paper we have implemented CURE clustering algorithm over distributed environment using Apache Hadoop. Now a days, to handle large store and handle huge data, Mapreduce has become very popular paradigm. Mapper and Reducer routines are designed for CURE algorithm. We have also discussed how different parameters affect quality of clusters by removing outliers.
DOI:10.1109/INVENTIVE.2016.7830238