Please use this identifier to cite or link to this item:
Title: Computational intelligence techniques for data analysis
Keywords: Computational intelligence, data analysis, neural networks, evolutionary algorithms, classification, time series forecasting.
Issue Date: 16-Jan-2009
Citation: ANG JI HUA BRIAN (2009-01-16). Computational intelligence techniques for data analysis. ScholarBank@NUS Repository.
Abstract: The use of computational intelligence techniques for data analysis is illustrated in this thesis, focusing particularly on identifying the existing issues and proposing new and effective algorithms. The data analysis techniques studied can be largely classified into two main approaches, namely non-rule-based approach and rule-based approach.For non-rule-based approach, the architectural design issues of Neural Networks (NN) for classification are discussed. An improved NN architecture with reduced interference in the input space is investigated. In addition, a novel evolutionary approach, which uses a growth probability, is proposed to optimize the weights and architecture of NN.For rule-based approach, an Evolutionary Memetic Algorithm (EMA) is used for rule extraction to discover knowledge from data sets. In EMA, two different local search schemes are used to complement the global search capability of evolutionary algorithm. In addition, a multi-objective rule-based technique for time series forecasting is also proposed.
Appears in Collections:Ph.D Theses (Open)

Show full item record
Files in This Item:
File Description SizeFormatAccess SettingsVersion 
Ang Ji Hua Brian 2009 NUS ECE.pdf1.59 MBAdobe PDF



Page view(s)

checked on Apr 13, 2019


checked on Apr 13, 2019

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.