Performance Comparison of Recent Imputation Methods for Classification Tasks over Binary Data
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
This paper evaluates the effect on the predictive accuracy of different models of two recently proposed imputation methods, namely missForest (MF) and Multiple Imputation based on Expectation-Maximization (MIEM), along with two other imputation methods: Sequential Hot-deck and Multiple Imputation based on Logistic Regression (MILR). Their effect is assessed over the classification accuracy of four different models, namely Tree Augmented Naive Bayes (TAN) which has received little attention, Naive Bayes (NB), Logistic Regression (LR), and Support Vector Machine (SVM) with Radial Basis Function (RBF) kernel. Experiments are conducted over fourteen binary datasets with large feature sets, and across a wide range of missing data rates (between 5 and 50%). The results from 10 fold cross-validations show that the performance of the imputation methods varies substantially between different classifiers and at different rates of missing values. The MIEM method is shown to generally give the best results for all the classifiers across all rates of missing data. While NB model does not benefit much from imputation compared to a no imputation baseline, LR and TAN are highly susceptible to gain from the imputation methods at higher rates of missing values. The results also show that MF works best with TAN, and Hot-deck degrades the predictive performance of SVM and NB models at high rates of missing values (over 30%). Detailed analysis of the imputation methods over the different datasets is reported. Implications of these findings on the choice of an imputation method are discussed.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.003 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it