Class center-based firefly algorithm for handling missing data

Heru Nugroho

doi:10.1186/s40537-021-00424-y

2021-02-23

Heru Nugroho *

Nugraha Priya Utama

Kridanto Surendro

DOI：10.1186/s40537-021-00424-y

原文链接

查看原文

摘要

En 中文

A significant advancement that occurs during the data cleaning stage is estimating missing data. Studies have shown that improper data handling leads to inaccurate analysis. Furthermore, most studies indicate the occurrence of missing data irrespective of the correlation between attributes. However, an adaptive search procedure helps to determine the estimates of the missing data when correlations between attributes are considered in the process. Firefly Algorithm (FA) implements an adaptive search procedure in the imputation of the missing data by determining the estimated value closest to others' value. Therefore, this study proposes a class center-based adaptive approach model for retrieving missing data by considering the attribute correlation in the imputation process (C3-FA). The result showed that the class center-based firefly algorithm (FA) is an efficient technique for obtaining the actual value in handling missing data with the Pearson correlation coefficient (r) and root mean squared error (RMSE) close to 1 and 0, respectively. In addition, the proposed method has the ability to maintain the true distribution of data values. This is indicated by the Kolmogorov-Smirnov test, which stated that the value of D-KS for most attributes in the dataset is generally closer to 0. Furthermore, the accuracy evaluation results using three classifiers showed that the proposed method produces good accuracy.

Keyword:

Missing data

Correlation

Imputation

Firefly algorithm

Class center