A neural network-based framework for the reconstruction of incomplete data sets

摘要

The treatment of incomplete data is an important step in the pre-processing of data. We propose a novel nonparametric algorithm Generalized regression neural network Ensemble for Multiple Imputation (GEMI). We also developed a single imputation (SI) version of this approach-GESI. We compare our algorithms with 25 popular missing data imputation algorithms on 98 real-world and synthetic datasets for various percentage of missing values. The effectiveness of the algorithms is evaluated in terms of (i) the accuracy of output classification: three classifiers (a generalized regression neural network, a multilayer perceptron and a logistic regression technique) are separately trained and tested on the dataset imputed with each imputation algorithm, (ii) interval analysis with missing observations and (iii) point estimation accuracy of the missing value imputation. GEMI outperformed GESI and all the conventional imputation algorithms in terms of all three criteria considered. Abbreviations: EM, expectation maximization; GA, genetic algorithm: GRNN, generalized regression neural networks; HD, hot-deck imputation; HUX, half uniform crossover; KNN, K-nearest neighbours; MAR, missing at random; MCAR, missing complete at random; MCMC, Markov chain Monte Carlo; Ml, multiple imputation; MLP, multilayer perceptrons; MNAR, missing not at random; MS, mean substitution; PCA. principal component analysis; PSO, particle swarm optimization; RBFN, radial basis function networks; SA, simulated annealing algorithm; SI, single Imputation; WKNN, weighted K-nearest neighbours; Zl, zero imputation