Imputation of Missing Data Using PCA, Neuro-Fuzzy and Genetic Algorithms

Authors:
Nthabiseng Hlalele;Fulufhelo Nelwamondo;Tshilidzi Marwala
Affiliations:
School of Electrical and Information Engineering, University of the Witwatersrand, Wits, 2050;School of Arts Sciences, Harvard University, Cambridge, 023138;School of Electrical and Information Engineering, University of the Witwatersrand, Wits, 2050
Venue:
Advances in Neuro-Information Processing
Year:
2009

Citing 0
Cited 1

Optimum estimation of missing values in randomized complete block design by genetic algorithm

Knowledge-Based Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents a method of imputing missing data that combines principal component analysis and neuro-fuzzy (PCA-NF) modeling in conjunction with genetic algorithms (GA). The ability of the model to impute missing data is tested using the South African HIV sero-prevalence dataset. The results indicate an average increase in accuracy from 60 % when using the neuro-fuzzy model independently to 99 % when the proposed model is used.