Data Management

Imputation Method for Missing Value Estimation of Mixed-Attribute Data Sets

Free registration required

Executive Summary

Missing data imputation is an important issue in learning from incomplete data. Various techniques have been developed with great successes on dealing with missing values in data sets with homogeneous attributes (their independent attributes are all either continuous or discrete). The authors propose a new setting of missing data imputation that is by imputing missing data in data sets with heterogeneous attributes thus by contributing both continuous and discrete data. They propose two consistent estimators for discrete and continuous missing target values. Then mixture kernel based iterative estimator and spherical kernel based iterative estimator is advocated to impute mixed-attribute data sets

  • Format: PDF
  • Size: 499.9 KB