Cross-validation based K nearest neighbor imputation for software quality datasets: An empirical study.
Abstract
Being able to predict software quality is essential, but also it pose significant challenges in software engineering. Historical software project datasets are often being utilized together with various machine learning algorithms for fault-proneness classification. Unfortunately, the missing values in datasets have negative impacts on the estimation accuracy and therefore, could lead to inconsistent results. As a method handling missing data, K nearest neighbor (KNN) imputation gradually gains acceptance in empirical studies by its exemplary performance and simplicity. To date, researchers still call for optimized parameter setting for KNN imputation to further improve its performance. In the work, we develop a novel incomplete-instance based KNN imputation technique, which utilizes a cross-validation scheme to optimize the parameters for each missing value. An experimental assessment is conducted on eight quality datasets under various missingness scenarios. The study also compared the proposed imputation approach with mean imputation and other three KNN imputation approaches. The results show that our proposed approach is superior to others in general. The relatively optimal fixed parameter settings for KNN imputation for software quality data is also 1 determined. It is observed that the classification accuracy is improved or at least maintained by using our approach for missing data imputation. Keywords: empirical software engineering estimation, KNN, imputation, cross-validation, missing data Abbreviations BMI CA CCkNNI CK CVBkNNI DkNNI FP FWGkNNI G3D GRA GRC GRG ICkNNI IDWM IRWM KNN LOC MAR MCAR MDT MEI MI MM MP MR NI PROMISE RMSE RQ SEE SVM Bayes Multiple Imputation Classification Accuracy Complete-case based KNN Imputation Chidamber and Kemerer object-oriented metric Cross-validation based KNN Imputation Default KNN Imputation Fault-proneness Feature Weighted Grey based KNN Imputation GRA-based distance, K = 3 with Dudani adaptation based imputation Grey Relational Analysis Grey Relational Coefficient Grey Relational Grade Incomplete-case based KNN Imputation Inverse Distance Weighted Mean Inverse Rank Weighted Mean K Nearest Neighbor Lines of Code M