Missing Value Imputation Based on Data Clustering

Abstract

Missing values widely exist in many real-world datasets, which hinders the performing of advanced data analytics. Properly filling these missing values is crucial but challenging, especially when the missing rate is high. Many approaches have been proposed for missing value imputation (MVI), but they are mostly heuristics-based, lacking a principled foundation and do not perform satisfactorily in practice. In this paper, we propose a probabilistic framework based on deep generative models for MVI. Under this framework, imputing the missing entries amounts to seeking a fixed-point solution between two conditional distributions defined on the missing entries and latent variables respectively. These distributions are parameterized by deep neural networks (DNNs) which possess high approximation power and can capture the nonlinear relationships between missing entries and the observed values. The learning of weight parameters of DNNs is performed by maximizing an approximation of the log-likelihood of observed values. We conducted extensive evaluation on 13 datasets and compared with 11 baselines methods, where our methods largely outperforms the baselines.

Shell-neighbor method and its application in missing data imputation

“…The most popular method in statistics is the regression imputation method. Common regression methods include the parametric methods (such as linear regression and the nonlinear imputation method) and the non-parametric methods (such as kernel imputation in [28]). The parametric regression imputations are superior if a dataset can be adequately modeled parametrically, or if users can correctly specify the parametric forms for the dataset.…”

“…Another commonly used and efficient imputation is the k nearest neighbor imputation (called kNN imputation, or kNNI), which is one of the hot deck techniques used to compensate for missing data [3,28]. It uses only the k most relevant complete instances in the dataset for imputing a missing datum.…”

Application in the Teaching of Principal Component Analysis

“…A grade table is constructed where courses are columns and students are rows, courses are labeled from T1, students are labeled from S1.The missing value in the table should be processed. The techniques of missing value imputation are: list wise deletion, mean imputation and some types of hot-deck imputation [8][9][10]. The listwise deletion is used to deal with missing value in the paper.…”

Imputation And Classification Of Missing Data Using Least Square Support Vector Machines – A New Approach In Dementia Diagnosis

“…Parametric methods like Nearest Neighbour [4][10] [25] have been used for the prediction of missing attribute(s). Non-parametric technique such as empirical likelihood [32], clustering [26], Semi-parametric techniques [21] [33] have also been applied for missing data imputation. Techniques like mixture model clustering [9], machine learning [12] have been used for imputing missing data.…”