DocumentCode :
3101612
Title :
Discovery of SIP/DRIP approach in distributed inter process communication
Author :
Hamid, Hakim ; Jais, Jamilin
Author_Institution :
Coll. of Inf. Technol., UNITEN, Kajang
fYear :
2008
fDate :
27-28 Aug. 2008
Firstpage :
476
Lastpage :
482
Abstract :
Classification modeling in data mining has evolved since 1990psilas. Many methods have been introduced and experimented. Among them were Multi Layer Perceptron and Radial Basis Function in Neural Network and Multiple Regressions in Statistical Analysis. Not many researches have been proposed in the field of rough classification modeling. When SIP/DRIP algorithm was ported on rough classification model, its accuracy has shown competitive results. The performance of the proposed rough model is compared with neural classifiers on different datasets. This paper made experiments on the combination of SIP/DRIP algorithm with DIPC distributed system to increase the computation speed of the method. Comparison made with another distributed computing system will be presented.
Keywords :
Unix; data mining; distributed object management; multilayer perceptrons; neural nets; radial basis function networks; regression analysis; rough set theory; MOSIX; SIP-DRIP algorithm; classification modeling; data mining; multilayer perceptron; neural network; radial basis function; regression analysis; rough classification model; rough sets; Data mining; Distributed computing; Educational institutions; Fuzzy set theory; Information technology; Neural networks; Probability distribution; Space missions; Space technology; Statistical distributions; DIPC; MOSIX; Rough set; classification; distributed computing;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Telecommunications, 2008. IST 2008. International Symposium on
Conference_Location :
Tehran
Print_ISBN :
978-1-4244-2750-5
Electronic_ISBN :
978-1-4244-2751-2
Type :
conf
DOI :
10.1109/ISTEL.2008.4651349
Filename :
4651349
Link To Document :
بازگشت