DocumentCode :
2145033
Title :
Optimize Prototype Classification Based on GA ANT Hybrid Technique
Author :
Khan, Mahrukh ; Dubey, Deepika ; Gupta, H. ; Saxena, Ankur
Author_Institution :
Dept. of Comput. Eng., TIEIT Bhopal, Bhopal, India
fYear :
2013
fDate :
27-29 Sept. 2013
Firstpage :
522
Lastpage :
526
Abstract :
Learning is the process of generating useful information from a huge volume of data. Learning can be classified as supervised learning and unsupervised learning. Classification is a kind of supervised learning and Clustering is a kind of unsupervised learning. Prototype clustering and classification is new technique in data mining. The prototype clustering and classification also called as ensemble clustering and classification. The Proposed novel method for prototype clustering and classification based on ant colony optimization and genetic algorithm. We conduct experiments with the prototype clustering and classification based on GA ANT hybrid technique using UCI repository with three data sets. The experimental results show that the proposed approach can achieve superior classification performance than other commonly used data mining approaches.
Keywords :
ant colony optimisation; data mining; genetic algorithms; learning (artificial intelligence); pattern classification; pattern clustering; GA ANT hybrid technique; UCI repository; ant colony optimization; data mining; ensemble classification; ensemble clustering; genetic algorithm; prototype classification optimization; prototype clustering; supervised learning; unsupervised learning; Ant colony optimization; Classification algorithms; Clustering algorithms; Data mining; Genetic algorithms; Partitioning algorithms; Prototypes; ant colony optimization; genetic algorithm; k-prototype clustering classification;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computational Intelligence and Communication Networks (CICN), 2013 5th International Conference on
Conference_Location :
Mathura
Type :
conf
DOI :
10.1109/CICN.2013.113
Filename :
6658049
Link To Document :
بازگشت