Matches in SemOpenAlex for { <https://semopenalex.org/work/W2955448166> ?p ?o ?g. }
Showing items 1 to 95 of
95
with 100 items per page.
- W2955448166 abstract "Statistical machine learning models should be evaluated and validated before putting to work. Conventional k-fold Monte Carlo Cross-Validation (MCCV) procedure uses a pseudo-random sequence to partition instances into k subsets, which usually causes subsampling bias, inflates generalization errors and jeopardizes the reliability and effectiveness of cross-validation. Based on ordered systematic sampling theory in statistics and low-discrepancy sequence theory in number theory, we propose a new k-fold cross-validation procedure by replacing a pseudo-random sequence with a best-discrepancy sequence, which ensures low subsampling bias and leads to more precise Expected-Prediction-Error estimates. Experiments with 156 benchmark datasets and three classifiers (logistic regression, decision tree and naive bayes) show that in general, our cross-validation procedure can extrude subsampling bias in the MCCV by lowering the EPE around 7.18% and the variances around 26.73%. In comparison, the stratified MCCV can reduce the EPE and variances of the MCCV around 1.58% and 11.85% respectively. The Leave-One-Out (LOO) can lower the EPE around 2.50% but its variances are much higher than the any other CV procedure. The computational time of our cross-validation procedure is just 8.64% of the MCCV, 8.67% of the stratified MCCV and 16.72% of the LOO. Experiments also show that our approach is more beneficial for datasets characterized by relatively small size and large aspect ratio. This makes our approach particularly pertinent when solving bioscience classification problems. Our proposed systematic subsampling technique could be generalized to other machine learning algorithms that involve random subsampling mechanism." @default.
- W2955448166 created "2019-07-12" @default.
- W2955448166 creator A5007241832 @default.
- W2955448166 creator A5022890638 @default.
- W2955448166 creator A5033916746 @default.
- W2955448166 date "2019-07-04" @default.
- W2955448166 modified "2023-09-27" @default.
- W2955448166 title "Subsampling Bias and The Best-Discrepancy Systematic Cross Validation" @default.
- W2955448166 cites W1520812622 @default.
- W2955448166 cites W1607458528 @default.
- W2955448166 cites W1680392829 @default.
- W2955448166 cites W174737001 @default.
- W2955448166 cites W1964916079 @default.
- W2955448166 cites W1975869018 @default.
- W2955448166 cites W1978501336 @default.
- W2955448166 cites W1999490823 @default.
- W2955448166 cites W2002667752 @default.
- W2955448166 cites W2018066814 @default.
- W2955448166 cites W2043419861 @default.
- W2955448166 cites W2045422729 @default.
- W2955448166 cites W2057580922 @default.
- W2955448166 cites W2105981176 @default.
- W2955448166 cites W2112081648 @default.
- W2955448166 cites W2117427558 @default.
- W2955448166 cites W2135699773 @default.
- W2955448166 cites W2144836152 @default.
- W2955448166 cites W2149736324 @default.
- W2955448166 cites W2150073804 @default.
- W2955448166 cites W2187483593 @default.
- W2955448166 cites W2319003951 @default.
- W2955448166 cites W2962833123 @default.
- W2955448166 doi "https://doi.org/10.48550/arxiv.1907.02437" @default.
- W2955448166 hasPublicationYear "2019" @default.
- W2955448166 type Work @default.
- W2955448166 sameAs 2955448166 @default.
- W2955448166 citedByCount "0" @default.
- W2955448166 crossrefType "posted-content" @default.
- W2955448166 hasAuthorship W2955448166A5007241832 @default.
- W2955448166 hasAuthorship W2955448166A5022890638 @default.
- W2955448166 hasAuthorship W2955448166A5033916746 @default.
- W2955448166 hasBestOaLocation W29554481661 @default.
- W2955448166 hasConcept C105795698 @default.
- W2955448166 hasConcept C107673813 @default.
- W2955448166 hasConcept C11413529 @default.
- W2955448166 hasConcept C12267149 @default.
- W2955448166 hasConcept C13280743 @default.
- W2955448166 hasConcept C154945302 @default.
- W2955448166 hasConcept C169258074 @default.
- W2955448166 hasConcept C185798385 @default.
- W2955448166 hasConcept C19499675 @default.
- W2955448166 hasConcept C205649164 @default.
- W2955448166 hasConcept C27181475 @default.
- W2955448166 hasConcept C2778112365 @default.
- W2955448166 hasConcept C33923547 @default.
- W2955448166 hasConcept C41008148 @default.
- W2955448166 hasConcept C49898467 @default.
- W2955448166 hasConcept C52001869 @default.
- W2955448166 hasConcept C54355233 @default.
- W2955448166 hasConcept C86803240 @default.
- W2955448166 hasConceptScore W2955448166C105795698 @default.
- W2955448166 hasConceptScore W2955448166C107673813 @default.
- W2955448166 hasConceptScore W2955448166C11413529 @default.
- W2955448166 hasConceptScore W2955448166C12267149 @default.
- W2955448166 hasConceptScore W2955448166C13280743 @default.
- W2955448166 hasConceptScore W2955448166C154945302 @default.
- W2955448166 hasConceptScore W2955448166C169258074 @default.
- W2955448166 hasConceptScore W2955448166C185798385 @default.
- W2955448166 hasConceptScore W2955448166C19499675 @default.
- W2955448166 hasConceptScore W2955448166C205649164 @default.
- W2955448166 hasConceptScore W2955448166C27181475 @default.
- W2955448166 hasConceptScore W2955448166C2778112365 @default.
- W2955448166 hasConceptScore W2955448166C33923547 @default.
- W2955448166 hasConceptScore W2955448166C41008148 @default.
- W2955448166 hasConceptScore W2955448166C49898467 @default.
- W2955448166 hasConceptScore W2955448166C52001869 @default.
- W2955448166 hasConceptScore W2955448166C54355233 @default.
- W2955448166 hasConceptScore W2955448166C86803240 @default.
- W2955448166 hasLocation W29554481661 @default.
- W2955448166 hasLocation W29554481662 @default.
- W2955448166 hasOpenAccess W2955448166 @default.
- W2955448166 hasPrimaryLocation W29554481661 @default.
- W2955448166 hasRelatedWork W2969635709 @default.
- W2955448166 hasRelatedWork W3044272884 @default.
- W2955448166 hasRelatedWork W3089416646 @default.
- W2955448166 hasRelatedWork W3106359073 @default.
- W2955448166 hasRelatedWork W3133324635 @default.
- W2955448166 hasRelatedWork W3204641204 @default.
- W2955448166 hasRelatedWork W4220803308 @default.
- W2955448166 hasRelatedWork W4283016678 @default.
- W2955448166 hasRelatedWork W4285225238 @default.
- W2955448166 hasRelatedWork W4294892151 @default.
- W2955448166 isParatext "false" @default.
- W2955448166 isRetracted "false" @default.
- W2955448166 magId "2955448166" @default.
- W2955448166 workType "article" @default.