Skip to content

Classification Elo Arena

Across 38 TabArena benchmark datasets (30 binary, 8 multiclass).
Click any column header to sort. Frozen key columns (ELO, Fit Time, Model) remain sticky.

Navigation:
- 1. All Datasets (Test Accuracy)
- 2. Binary Datasets (30)
- 3. Multiclass Datasets (8)
- 4. All Datasets (Test LogLoss)
- ← Overall Combined Arena | Regression Arena →


Executive Commentary & Analysis

1. Benchmark Methodology

  • Scope: 38 TabArena-v0.1 classification datasets spanning industrial failure diagnosis (APSFailure, seismic-bumps), customer retention (Bank_Customer_Churn, churn, online_shoppers_intention), clinical screening (Bioresponse, Diabetes130US, maternal_health_risk), cybersecurity (website_phishing, NATICUSdroid), and corporate bankruptcy (polish_companies_bankruptcy, taiwanese_bankruptcy_prediction).
  • Elo Rating Engine: Pairwise round-robin head-to-head matches across all pairs on holdout test accuracy (\(K=32\), initialized at \(R_0 = 1500\)).
  • Calibration Sub-Arena: Section 4 evaluates multiclass cross-entropy (LogLoss) to penalize overconfident and miscalibrated models.

2. Tier Analysis & Core Findings

  • Tier 1 — Symmetric & Categorical Boosters (Elo 1300–1390):
    CatBoost (1388.0) and perpetual (1302.3) dominate tabular classification. CatBoost's target-based categorical encoders and oblivious symmetric decision trees provide strong regularization against overfitting on sparse categories.
  • Tier 2 — Random Forests & Additive Ensembles (Elo 1240–1280):
    sklearn RF (1279.0), ChimeraBoost (1265.3), EBM (1264.0), and ModelTree(all) (1244.3). ModelTree(all) shows remarkable competitiveness, outscoring standard LightGBM (1148.2) and HistGradientBoosting (1170.8) on complex mixed-type tasks by utilizing local regularized logistic regressions at leaves.
  • Tier 3 — Classical & Single-Tree Baselines (Elo < 1000):
    ctree (948.1), sklearn tree (800.1), KNN (791.0), and rpart (760.4). Single decision trees experience severe boundary discretization error, while model trees with local logistic models bridge the gap to full ensembles.

1. All Datasets (Test Accuracy Elo)

Evaluates raw classification accuracy across all 38 datasets.

ELO FIT TIME (±sd) MODELAPSFailureAmazon_employee_accessBank_Customer_ChurnBioresponseDiabetes130USE-CommereShippingDataFitness_ClubGiveMeSomeCreditHR_Analytics_Job_Change_of_Data_ScientistsIs-this-a-good-customerMICMarketing_CampaignNATICUSdroidSDSS17annealbank-marketingblood-transfusion-service-centerchurncoil2000_insurance_policiescredit-gcredit_card_clients_defaultcustomer_satisfaction_in_airlinediabeteshazelnut-spread-contaminant-detectionhelochiva_agnosticin_vehicle_coupon_recommendationjm1kddcup09_appetencymaternal_health_riskonline_shoppers_intentionpolish_companies_bankruptcyqsar-biodegseismic-bumpssplicestudents_dropout_and_academic_successtaiwanese_bankruptcy_predictionwebsite_phishing
1388.07.75±23.64sCatBoost(200) 38/380.99370.95440.86750.80400.91240.66230.79670.93550.79590.88660.89120.90180.94130.97710.99440.89280.76670.96200.93840.71000.81480.95900.70130.95000.72180.96360.77850.81900.98210.79700.90470.96700.87680.93040.94830.78310.97360.8856
1302.320.59±33.81sperpetual (Rust) 24/380.99410.94520.8660·0.91210.67450.78000.93650.79540.8895·0.8772···0.89450.76000.9530·0.71500.81130.95620.68830.94370.7242·0.77770.82090.9822·0.9055··0.9342····
1279.05.28±13.05ssklearn RF 38/380.99360.94700.85900.79730.91200.65360.78330.93630.78440.88370.88820.87500.92660.97940.99440.88750.73330.95300.93020.74500.81570.95570.70130.91460.72850.95970.75960.81810.98220.80690.90920.95690.85780.92260.93890.77630.97140.8856
1265.35.25±7.00sChimeraBoost 38/380.99400.95160.86700.81470.91220.67140.79000.93730.79670.88660.88530.89060.92790.97770.97780.89390.74000.96000.93990.73500.81600.95390.72730.95630.72560.96490.73910.82040.98210.73270.90590.96530.87200.93420.95920.76610.97070.8856
1264.034.09±32.57sEBM 29/38··0.8700··0.66820.8000·0.79510.88370.89410.88620.9346·0.98890.89460.76000.95800.93890.75500.8185·0.66880.95210.7290·0.76670.8190·0.75250.90630.96620.87200.93230.94980.78190.97650.8745
1244.318.55±41.36sModelTree(all) 38/380.98630.88590.86500.76130.90820.67450.76670.93730.78940.88080.86180.86830.93120.96850.96670.89370.76670.94000.92720.70000.81450.91470.66880.78330.71030.96490.72880.80620.98220.64850.90920.93490.89100.92650.93100.76270.97360.8819
1228.811.16±13.28sbart 19/38·0.94410.87050.7867·0.6809··0.79740.8837··0.9332··0.89470.76670.94600.94090.74000.8178·0.70130.93330.7357·0.7718·····0.85310.9342····
1204.018.03±33.88sranger 38/380.99350.94900.86000.81200.91200.65140.78670.93660.79140.88660.88530.88170.93520.97920.99440.89490.74670.95200.93990.74500.81630.95540.72730.91040.72900.96490.75680.82180.98220.80690.90670.95940.87680.93040.94670.78310.97140.8782
1192.29.63±21.42ssklearn GBM 37/38·0.94120.86850.80000.91180.66820.80000.93760.79560.87790.87940.85940.93320.97711.00000.89450.74670.95300.93890.73500.81900.91850.70780.91460.72990.95580.74060.81400.93530.73270.90510.96360.86730.92260.95300.76720.97070.8819
1187.83.99±5.63sc50_rules 33/38·0.05780.86400.7707·0.67640.8100·0.79720.88370.86180.85710.91990.9735·0.89430.73330.94000.94040.71500.8197·0.70130.86460.71320.01040.73000.81260.98220.74750.89940.93570.85780.93420.93420.74350.96700.8967
1170.81.54±3.56sHistGradientBoosting 33/380.9921·0.86250.8040·0.66500.76670.93770.80110.86630.8765·0.9339·0.99440.89530.71330.95600.93740.74500.81900.95260.69480.93540.72320.95580.77340.8227·0.81680.90190.96450.87680.91490.94830.79440.97430.8708
1159.45.00±7.76sc50_tree 36/380.99070.94220.86400.7667·0.67550.81330.93610.79720.88370.86760.85710.92120.9730·0.89400.73330.94400.94040.71000.81850.94770.70130.84380.70840.96490.72330.81260.98220.74750.90020.93820.83410.93420.92790.74460.96480.8967
1148.20.72±0.94sLightGBM(200) 38/380.99300.95220.85950.79070.91120.64680.77330.93690.79350.86050.88530.88840.93590.97470.98330.89360.71330.96100.92970.72500.81380.95570.70780.95830.71750.95840.78280.82040.98180.82670.89900.96360.87200.92070.95770.78640.97360.8708
1110.90.57±1.41sCTBoost 29/38··0.86600.7947·0.66820.7933·0.80240.88660.8882·0.9352·1.00000.89560.76000.94500.94040.73500.8185·0.71430.88330.72800.96490.75720.8176·0.7277·0.96450.87680.93420.96390.76950.97210.8708
1076.22.33±3.89sExtraTrees 38/380.99320.94250.85800.80270.91180.65500.75000.93460.77420.88660.88820.87280.92590.97371.00000.87600.70670.94300.92620.72500.81170.95640.69480.91460.72370.95190.73790.81900.98220.80690.89620.95350.87680.92070.93730.76500.97070.8745
1036.91.17±0.79sforust (Rust) 30/380.99270.94550.86100.78800.91140.66320.76670.93600.78960.8576·0.84150.9372··0.89150.70670.95800.93330.70000.81530.95390.71430.95420.7161·0.77300.81530.9823·0.90190.96110.86730.9110··0.9721·
1018.835.68±50.56sMARS (mars-earth) 8/38·····0.6491···0.8924······0.7667··0.7000··0.7143···0.6926··0.5792···0.9323····
948.12.50±3.05sctree 27/38··0.83900.7653·0.67910.7933··0.88660.8824·0.9085·0.95560.89470.73330.94100.94040.73500.8187·0.71430.82290.69740.96490.71380.8107·0.6188·0.93060.83410.93420.9404·0.96920.8672
947.13.45±10.75sModelTree(opt) 38/380.98900.94320.85100.75730.91010.66860.80330.93530.79480.88660.87940.85040.90590.96390.98330.89450.71330.93100.94040.70000.82020.90760.70130.69790.70700.96490.66020.80660.98220.70790.89250.95520.85310.93420.92320.75030.96770.8708
925.67.01±34.82sModelTree(path) 38/380.98730.86300.85900.76000.91160.67450.80670.93670.79010.88660.88820.87280.89450.96770.83330.89470.76670.94100.93990.73000.81380.87940.68830.80000.70840.96490.70710.80980.98070.59410.90840.95520.85310.92840.92630.74350.97360.8413
924.46.48±34.58sModelTree(mean) 38/380.98630.94050.85250.76130.91200.68050.79330.93620.79330.88660.88820.87050.89450.96150.76670.89450.73330.93400.94040.73000.81900.87880.72080.77710.70120.96490.71150.80980.98220.53960.90390.95260.83410.93420.92950.74460.97290.8413
909.510.23±8.57sgrf_rf 9/38···0.7800············0.7667·····0.70130.88750.7165··0.8163·0.6782·0.9535····0.9677·
836.16.79±10.78sModelTree(winnow) 38/380.98630.81750.86450.76000.91190.67320.79330.93720.78390.88950.85880.85710.93190.95910.96670.89490.75330.93600.93230.74000.81570.89710.68830.78120.70360.96490.72010.80620.98220.59900.89980.93570.87680.93040.94200.76050.96920.8303
800.11.09±4.66ssklearn tree 38/380.98930.90870.78600.74400.83180.65270.70670.89680.71290.81690.81470.80130.89920.96521.00000.81860.66000.92700.89710.65500.73320.93310.68830.86460.62140.92590.68190.75290.90790.80200.86130.94160.82940.89560.91220.67570.95450.8745
791.00.34±0.62sKNN 38/380.98620.93870.76400.74800.90630.64550.76330.93170.74890.87210.85290.83040.91660.67680.85560.87540.77330.88900.93430.63000.76020.70230.70130.83960.66250.96490.62040.78500.98200.67330.85810.92890.82940.92070.65050.62490.96850.8561
760.42.34±6.81srpart 36/380.98900.93300.85750.75330.91200.68050.79330.93530.79640.88080.88240.77680.8932·0.98890.89460.72670.94000.94040.72500.82020.86800.72080.82920.69980.96490.68820.80660.97380.61390.90310.92890.83890.93420.9357·0.96700.8266
737.723.99±39.03smodeltrees glmtree 16/38··0.8690··0.6723···0.8808··0.9259···0.7600·0.93840.73500.8192·0.70780.92500.7266·0.7237···0.8885·0.84830.9226··0.9648·
727.60.03±0.01sstatsmodels MNLogit 5/38················0.7533·····0.71430.91870.7141····0.5693········
701.312.56±23.52smboost 19/38··0.86450.7907·0.67180.7933··0.8866··0.9166··0.89450.7600··0.74000.8172·0.69480.90000.7266····0.5495·0.93230.86260.9304·0.7198·0.8155
701.111.53±25.66smodeltrees multinomtree 15/38·····0.6741···0.8808····0.0111····0.73500.8117·0.70780.92500.72800.0104···0.6584··0.84830.92650.23980.3209·0.8524
687.927.82±41.22sglmtree 16/38··0.86900.6653·0.6723···0.8808··0.9259···0.7600·0.93840.73500.8173·0.70780.92500.7266·0.7237·····0.84830.9246··0.9560·
666.82.10±3.93searth 19/38··0.86050.7787·0.6650··0.77860.8837··0.9199··0.89380.78000.87400.93840.74500.8167·0.68830.90210.7309·0.6996·····0.86730.9207··0.9604·
658.74.27±6.32ssklearn LogReg 38/380.98910.94220.80750.76130.91180.63640.80000.93430.77080.88660.88820.83480.92860.77980.89440.88270.75330.86500.93890.69000.79800.81710.70780.66670.69360.94800.62670.80520.98220.59410.89010.92550.85780.92260.82600.77060.96410.8155

2. Binary Classification Datasets (30)

Evaluates models strictly on binary classification problems.

ELO FIT TIME (±sd) MODELAPSFailureAmazon_employee_accessBank_Customer_ChurnBioresponseDiabetes130USE-CommereShippingDataFitness_ClubGiveMeSomeCreditHR_Analytics_Job_Change_of_Data_ScientistsIs-this-a-good-customerMarketing_CampaignNATICUSdroidbank-marketingblood-transfusion-service-centerchurncoil2000_insurance_policiescredit-gcredit_card_clients_defaultcustomer_satisfaction_in_airlinediabeteshazelnut-spread-contaminant-detectionhelocin_vehicle_coupon_recommendationjm1kddcup09_appetencyonline_shoppers_intentionpolish_companies_bankruptcyqsar-biodegseismic-bumpstaiwanese_bankruptcy_prediction
1368.034.09±32.57sEBM 23/30··0.8700··0.66820.8000·0.79510.88370.88620.93460.89460.76000.95800.93890.75500.8185·0.66880.95210.72900.76670.8190·0.90630.96620.87200.93230.9765
1297.720.59±33.81sperpetual (Rust) 24/300.99410.94520.8660·0.91210.67450.78000.93650.79540.88950.8772·0.89450.76000.9530·0.71500.81130.95620.68830.94370.72420.77770.82090.98220.9055··0.9342·
1291.40.57±1.41sCTBoost 22/30··0.86600.7947·0.66820.7933·0.80240.8866·0.93520.89560.76000.94500.94040.73500.8185·0.71430.88330.72800.75720.8176··0.96450.87680.93420.9721
1281.27.75±23.64sCatBoost(200) 30/300.99370.95440.86750.80400.91240.66230.79670.93550.79590.88660.90180.94130.89280.76670.96200.93840.71000.81480.95900.70130.95000.72180.77850.81900.98210.90470.96700.87680.93040.9736
1226.711.16±13.28sbart 19/30·0.94410.87050.7867·0.6809··0.79740.8837·0.93320.89470.76670.94600.94090.74000.8178·0.70130.93330.73570.7718····0.85310.9342·
1205.418.55±41.36sModelTree(all) 30/300.98630.88590.86500.76130.90820.67450.76670.93730.78940.88080.86830.93120.89370.76670.94000.92720.70000.81450.91470.66880.78330.71030.72880.80620.98220.90920.93490.89100.92650.9736
1182.26.48±34.58sModelTree(mean) 30/300.98630.94050.85250.76130.91200.68050.79330.93620.79330.88660.87050.89450.89450.73330.93400.94040.73000.81900.87880.72080.77710.70120.71150.80980.98220.90390.95260.83410.93420.9729
1181.81.54±3.56sHistGradientBoosting 26/300.9921·0.86250.8040·0.66500.76670.93770.80110.8663·0.93390.89530.71330.95600.93740.74500.81900.95260.69480.93540.72320.77340.8227·0.90190.96450.87680.91490.9743
1170.77.01±34.82sModelTree(path) 30/300.98730.86300.85900.76000.91160.67450.80670.93670.79010.88660.87280.89450.89470.76670.94100.93990.73000.81380.87940.68830.80000.70840.70710.80980.98070.90840.95520.85310.92840.9736
1153.15.25±7.00sChimeraBoost 30/300.99400.95160.86700.81470.91220.67140.79000.93730.79670.88660.89060.92790.89390.74000.96000.93990.73500.81600.95390.72730.95630.72560.73910.82040.98210.90590.96530.87200.93420.9707
1131.318.03±33.88sranger 30/300.99350.94900.86000.81200.91200.65140.78670.93660.79140.88660.88170.93520.89490.74670.95200.93990.74500.81630.95540.72730.91040.72900.75680.82180.98220.90670.95940.87680.93040.9714
1127.30.72±0.94sLightGBM(200) 30/300.99300.95220.85950.79070.91120.64680.77330.93690.79350.86050.88840.93590.89360.71330.96100.92970.72500.81380.95570.70780.95830.71750.78280.82040.98180.89900.96360.87200.92070.9736
1077.335.68±50.56sMARS (mars-earth) 7/30·····0.6491···0.8924···0.7667··0.7000··0.7143··0.6926·····0.9323·
1034.612.56±23.52smboost 16/30··0.86450.7907·0.67180.7933··0.8866·0.91660.89450.7600··0.74000.8172·0.69480.90000.7266····0.93230.86260.9304·
1034.45.28±13.05ssklearn RF 30/300.99360.94700.85900.79730.91200.65360.78330.93630.78440.88370.87500.92660.88750.73330.95300.93020.74500.81570.95570.70130.91460.72850.75960.81810.98220.90920.95690.85780.92260.9714
1028.41.17±0.79sforust (Rust) 30/300.99270.94550.86100.78800.91140.66320.76670.93600.78960.85760.84150.93720.89150.70670.95800.93330.70000.81530.95390.71430.95420.71610.77300.81530.98230.90190.96110.86730.91100.9721
990.60.03±0.01sstatsmodels MNLogit 4/30·············0.7533·····0.71430.91870.7141········
984.52.50±3.05sctree 21/30··0.83900.7653·0.67910.7933··0.8866·0.90850.89470.73330.94100.94040.73500.8187·0.71430.82290.69740.71380.8107··0.93060.83410.93420.9692
983.99.63±21.42ssklearn GBM 29/30·0.94120.86850.80000.91180.66820.80000.93760.79560.87790.85940.93320.89450.74670.95300.93890.73500.81900.91850.70780.91460.72990.74060.81400.93530.90510.96360.86730.92260.9707
981.26.79±10.78sModelTree(winnow) 30/300.98630.81750.86450.76000.91190.67320.79330.93720.78390.88950.85710.93190.89490.75330.93600.93230.74000.81570.89710.68830.78120.70360.72010.80620.98220.89980.93570.87680.93040.9692
964.02.33±3.89sExtraTrees 30/300.99320.94250.85800.80270.91180.65500.75000.93460.77420.88660.87280.92590.87600.70670.94300.92620.72500.81170.95640.69480.91460.72370.73790.81900.98220.89620.95350.87680.92070.9707
954.63.45±10.75sModelTree(opt) 30/300.98900.94320.85100.75730.91010.66860.80330.93530.79480.88660.85040.90590.89450.71330.93100.94040.70000.82020.90760.70130.69790.70700.66020.80660.98220.89250.95520.85310.93420.9677
946.011.53±25.66smodeltrees multinomtree 9/30·····0.6741···0.8808······0.73500.8117·0.70780.92500.7280·····0.84830.9265·
912.13.99±5.63sc50_rules 26/30·0.05780.86400.7707·0.67640.8100·0.79720.88370.85710.91990.89430.73330.94000.94040.71500.8197·0.70130.86460.71320.73000.81260.98220.89940.93570.85780.93420.9670
907.410.23±8.57sgrf_rf 8/30···0.7800·········0.7667·····0.70130.88750.7165·0.8163··0.9535··0.9677
895.52.34±6.81srpart 30/300.98900.93300.85750.75330.91200.68050.79330.93530.79640.88080.77680.89320.89460.72670.94000.94040.72500.82020.86800.72080.82920.69980.68820.80660.97380.90310.92890.83890.93420.9670
845.35.00±7.76sc50_tree 29/300.99070.94220.86400.7667·0.67550.81330.93610.79720.88370.85710.92120.89400.73330.94400.94040.71000.81850.94770.70130.84380.70840.72330.81260.98220.90020.93820.83410.93420.9648
730.023.99±39.03smodeltrees glmtree 16/30··0.8690··0.6723···0.8808·0.9259·0.7600·0.93840.73500.8192·0.70780.92500.72660.7237··0.8885·0.84830.92260.9648
694.00.34±0.62sKNN 30/300.98620.93870.76400.74800.90630.64550.76330.93170.74890.87210.83040.91660.87540.77330.88900.93430.63000.76020.70230.70130.83960.66250.62040.78500.98200.85810.92890.82940.92070.9685
689.24.27±6.32ssklearn LogReg 30/300.98910.94220.80750.76130.91180.63640.80000.93430.77080.88660.83480.92860.88270.75330.86500.93890.69000.79800.81710.70780.66670.69360.62670.80520.98220.89010.92550.85780.92260.9641
683.527.82±41.22sglmtree 16/30··0.86900.6653·0.6723···0.8808·0.9259·0.7600·0.93840.73500.8173·0.70780.92500.72660.7237····0.84830.92460.9560
661.72.10±3.93searth 19/30··0.86050.7787·0.6650··0.77860.8837·0.91990.89380.78000.87400.93840.74500.8167·0.68830.90210.73090.6996····0.86730.92070.9604
384.91.09±4.66ssklearn tree 30/300.98930.90870.78600.74400.83180.65270.70670.89680.71290.81690.80130.89920.81860.66000.92700.89710.65500.73320.93310.68830.86460.62140.68190.75290.90790.86130.94160.82940.89560.9545

3. Multiclass Classification Datasets (8)

Evaluates models supporting multi-label / multi-class classification.

ELO FIT TIME (±sd) MODELMICSDSS17annealhiva_agnosticmaternal_health_risksplicestudents_dropout_and_academic_successwebsite_phishing
1356.07.75±23.64sCatBoost(200) 8/80.89120.97710.99440.96360.79700.94830.78310.8856
1305.85.28±13.05ssklearn RF 8/80.88820.97940.99440.95970.80690.93890.77630.8856
1283.25.25±7.00sChimeraBoost 8/80.88530.97770.97780.96490.73270.95920.76610.8856
1259.318.03±33.88sranger 8/80.88530.97920.99440.96490.80690.94670.78310.8782
1228.89.63±21.42ssklearn GBM 8/80.87940.97711.00000.95580.73270.95300.76720.8819
1209.55.00±7.76sc50_tree 7/80.86760.9730·0.96490.74750.92790.74460.8967
1195.53.99±5.63sc50_rules 7/80.86180.9735·0.01040.74750.93420.74350.8967
1186.334.09±32.57sEBM 6/80.8941·0.9889·0.75250.94980.78190.8745
1177.00.72±0.94sLightGBM(200) 8/80.88530.97470.98330.95840.82670.95770.78640.8708
1160.31.54±3.56sHistGradientBoosting 7/80.8765·0.99440.95580.81680.94830.79440.8708
1134.018.55±41.36sModelTree(all) 8/80.86180.96850.96670.96490.64850.93100.76270.8819
1090.50.57±1.41sCTBoost 7/80.8882·1.00000.96490.72770.96390.76950.8708
1086.42.33±3.89sExtraTrees 8/80.88820.97371.00000.95190.80690.93730.76500.8745
990.210.23±8.57sgrf_rf 1/8····0.6782···
953.12.50±3.05sctree 6/80.8824·0.95560.96490.61880.9404·0.8672
945.63.45±10.75sModelTree(opt) 8/80.87940.96390.98330.96490.70790.92320.75030.8708
941.91.09±4.66ssklearn tree 8/80.81470.96521.00000.92590.80200.91220.67570.8745
834.06.79±10.78sModelTree(winnow) 8/80.85880.95910.96670.96490.59900.94200.76050.8303
788.035.68±50.56sMARS (mars-earth) 1/8····0.5792···
785.82.34±6.81srpart 6/80.8824·0.98890.96490.61390.9357·0.8266
784.66.48±34.58sModelTree(mean) 8/80.88820.96150.76670.96490.53960.92950.74460.8413
765.00.03±0.01sstatsmodels MNLogit 1/8····0.5693···
760.37.01±34.82sModelTree(path) 8/80.88820.96770.83330.96490.59410.92630.74350.8413
757.30.34±0.62sKNN 8/80.85290.67680.85560.96490.67330.65050.62490.8561
735.34.27±6.32ssklearn LogReg 8/80.88820.77980.89440.94800.59410.82600.77060.8155
677.611.53±25.66smodeltrees multinomtree 6/8··0.01110.01040.65840.23980.32090.8524
608.612.56±23.52smboost 3/8····0.5495·0.71980.8155

4. All Datasets (Test LogLoss Elo)

Evaluates cross-entropy / probability calibration across all 38 datasets.

ELO FIT TIME (±sd) MODELAPSFailureAmazon_employee_accessBank_Customer_ChurnBioresponseDiabetes130USE-CommereShippingDataFitness_ClubGiveMeSomeCreditHR_Analytics_Job_Change_of_Data_ScientistsIs-this-a-good-customerMICMarketing_CampaignNATICUSdroidSDSS17annealbank-marketingblood-transfusion-service-centerchurncoil2000_insurance_policiescredit-gcredit_card_clients_defaultcustomer_satisfaction_in_airlinediabeteshazelnut-spread-contaminant-detectionhelochiva_agnosticin_vehicle_coupon_recommendationjm1kddcup09_appetencymaternal_health_riskonline_shoppers_intentionpolish_companies_bankruptcyqsar-biodegseismic-bumpssplicestudents_dropout_and_academic_successtaiwanese_bankruptcy_predictionwebsite_phishing
1515.27.75±23.64sCatBoost(200) 38/380.0220.1390.3270.4460.2800.5170.4440.1820.4360.3250.4670.2510.1660.0760.0110.3030.5220.1580.2090.5230.4350.0970.6110.1930.5470.2000.4790.4250.0740.4910.2200.1060.3010.2480.1490.5750.0750.306
1498.534.09±32.57sEBM 29/38··0.326··0.5090.436·0.4420.3130.4530.2700.177·0.0220.3010.5010.1720.2080.5120.430·0.5630.1230.540·0.4930.424·0.5640.2200.1200.3050.2440.1980.5710.0710.320
1495.818.03±33.88sranger 38/380.0310.1580.3370.4500.2880.5180.4680.1820.4470.3410.4630.2910.1860.0830.0650.3040.5500.1810.2290.5060.4360.1230.5890.2730.5470.2450.5290.4230.0780.5150.2230.1350.2970.2550.3270.5870.0740.301
1433.29.63±21.42ssklearn GBM 37/38·0.1980.3290.4660.2840.5080.4430.1790.4380.3390.6430.3150.1850.0770.0000.3020.5580.1720.2080.5390.4300.1960.6110.2310.5440.2130.5430.4280.1810.5720.2240.1130.3340.2510.1550.5700.0860.311
1432.55.25±7.00sChimeraBoost 38/380.0220.1450.3280.4220.2820.5090.4380.1780.4380.3160.4980.2600.1830.0710.0470.3030.4980.1640.2030.5290.4310.1030.5760.1360.5450.1740.5390.4240.0720.5270.2220.1050.3270.2410.1460.5730.0790.321
1415.311.16±13.28sbart 19/38·0.1820.3210.469·0.504··0.4400.317··0.177··0.3020.4910.1900.2030.5010.430·0.5730.2030.542·0.497·····0.3110.242····
1382.520.59±33.81sperpetual (Rust) 24/380.0360.1740.335·0.2890.5120.4460.1830.4370.329·0.298···0.3040.5020.174·0.5380.4500.0980.6320.2800.547·0.4970.4240.122·0.225··0.237····
1362.50.57±1.41sCTBoost 29/38··0.3310.454·0.5250.451·0.4340.3250.471·0.179·0.0180.3000.5060.1950.2060.5210.430·0.6090.3400.5490.1790.5150.429·0.684·0.1150.3270.2440.1360.6110.0750.332
1307.410.23±8.57sgrf_rf 9/38···0.516············0.500·····0.5620.3320.553··0.423·0.638·0.137····0.081·
1203.735.68±50.56sMARS (mars-earth) 8/38·····0.534···0.319······0.492··0.555··0.549···0.587··0.845···0.241····
1143.85.28±13.05ssklearn RF 38/380.0380.3070.3430.4540.3070.5280.8000.2830.5350.5270.8110.3040.2380.1210.0260.4840.8270.2130.5610.5120.4620.1260.6060.3310.5510.7010.5220.6170.1230.4220.2610.2100.3060.6640.3150.6950.0970.408
1127.35.00±7.76sc50_tree 36/380.0320.2210.3790.550·0.5120.4560.2040.4450.3310.6980.3890.2350.104·0.3280.5360.1950.2260.6780.4510.1370.7720.3980.5900.1730.6030.4820.0890.6180.2570.1980.4730.2420.2960.7280.1170.345
1063.111.53±25.66smodeltrees multinomtree 11/38·····0.567···0.321·········0.5170.466·0.5760.8740.556····0.654··0.3490.345···0.421
1050.81.17±0.79sforust (Rust) 30/380.0300.1690.3460.4730.2870.5450.5720.1810.4550.479·0.3590.176··0.3070.6560.1910.2330.7080.4450.1071.0150.1390.570·0.4830.4380.085·0.2380.1370.3700.333··0.097·
1048.11.54±3.56sHistGradientBoosting 33/380.024·0.3350.434·0.5170.5480.1780.4370.4221.070·0.174·0.0510.3020.6450.1890.2270.6590.4290.1130.9310.1470.5480.4620.4880.425·0.4440.2290.1390.3650.3490.2680.6270.1170.472
1022.90.72±0.94sLightGBM(200) 38/380.0310.2060.3490.4790.3010.5390.6280.1790.4570.5871.1800.3670.1860.1250.0490.3070.6840.2380.2600.8180.4390.1011.3210.1490.5640.5450.4680.4340.1320.4660.2400.2130.5430.4500.1450.6550.1680.471
1013.32.34±6.81srpart 36/380.0390.2810.3870.5350.2980.5160.5000.2130.4520.3360.9021.6780.291·0.0410.3300.5660.2040.2260.6300.4660.3650.8570.4240.6100.1730.6050.4910.2290.7850.2490.2080.4270.2420.234·0.1100.528
991.52.50±3.05sctree 27/38··0.3950.559·0.5050.453··0.3220.632·0.262·0.1100.3110.5350.2220.2100.5460.444·1.8830.4550.5780.1730.5940.440·0.802·0.2040.4150.2300.347·0.0950.554
969.00.03±0.01sstatsmodels MNLogit 5/38················0.486·····0.5200.2120.563····0.830········
942.84.27±6.32ssklearn LogReg 38/380.0380.2190.4420.6070.2930.5560.4350.2250.4780.3200.5510.3840.1930.6410.2600.3340.4860.3330.2120.5590.4810.4240.5200.6930.5750.3480.6390.5410.0890.8310.3080.4590.3360.2650.4420.5970.1850.515
913.412.56±23.52smboost 19/38··0.3380.504·0.5240.448··0.322··0.219··0.3080.519··0.5020.435·0.5870.2830.544····0.894·0.2080.3310.234·0.806·0.673
898.118.55±41.36sModelTree(all) 38/380.0382.4780.4300.7180.4000.5100.4930.1830.4500.4821.9350.4390.2000.2550.2530.3110.5290.4880.3051.0180.4710.1991.0430.5440.6890.2560.5590.5110.0960.7550.2300.8170.5980.4490.5800.8310.1180.481
838.93.45±10.75sModelTree(opt) 38/380.0330.2150.3760.5240.3820.5090.4400.1880.4470.3531.4720.4220.2320.3010.2400.3140.5160.2110.2080.6110.4470.2100.5520.5760.5660.1730.6150.4520.0810.7310.2370.1720.3430.2380.7381.4660.0881.023
834.66.48±34.58sModelTree(mean) 38/380.0380.2240.3670.5570.2900.5050.4400.1860.4450.4290.7520.3170.2890.1400.6630.3120.5380.2710.2240.8690.4390.2592.2640.5440.5830.1730.5700.4410.0960.8020.2380.2500.5300.4750.2350.6560.1190.602
807.62.33±3.89sExtraTrees 38/380.0300.4760.4130.4580.3080.5450.7230.2940.7590.4510.8120.7130.8990.1310.0210.8473.4490.2430.8880.5220.5130.1270.5740.2630.5530.7540.6880.6670.1120.9700.2860.1460.4390.7350.3960.5960.1001.931
787.66.79±10.78sModelTree(winnow) 38/380.0384.7670.3780.6700.2860.5050.4530.1910.4510.4441.8670.3510.1990.2090.2610.3120.5030.2350.2590.9380.4510.2330.9800.5440.6720.2580.5620.5110.0860.8870.2590.3570.5400.4060.5100.7710.1910.552
766.47.01±34.82sModelTree(path) 38/380.0453.1680.3870.5530.2950.5070.4550.1850.4440.4490.8600.3260.2890.1510.6830.3110.5240.2670.2250.8680.4390.2512.2650.5440.5810.2560.5650.4410.1460.9070.2360.2600.5110.4390.3610.7420.1180.680
745.823.99±39.03smodeltrees glmtree 16/38··0.325··0.576···0.321··0.211···0.500·0.2150.5170.442·0.5760.8700.555·0.564···0.294·0.3490.338··1.215·
735.02.10±3.93searth 19/38··0.3430.508·0.522··0.4520.345··0.201··0.3060.4930.2860.2070.5360.436·0.5850.2220.542·0.589·····0.3880.272··1.367·
735.027.82±41.22sglmtree 16/38··0.32511.559·0.576···0.321··0.211···0.500·0.2150.5170.444·0.5760.8700.555·0.564·····0.3490.265··1.519·
338.10.34±0.62sKNN 38/380.1251.2752.5191.6551.9631.8542.1991.6082.1192.1323.6172.4680.7704.2670.6491.9282.4322.0771.4402.2652.3141.5173.1201.3222.0481.0761.5852.3160.5851.8611.6891.1350.7811.5182.3344.2080.8271.974
243.73.99±5.63sc50_rules 33/38·0.6933.0595.946·8.9766.451·7.0054.0164.5854.9342.5050.857·3.6199.2101.6890.6936.7206.004·6.5163.3398.9941.0993.0696.4420.6936.4981.5402.1112.7700.6931.3367.0721.0172.731
-63.61.09±4.66ssklearn tree 38/380.3703.1527.3918.8425.81011.99410.1313.5549.9096.3256.4006.7923.2861.2010.0006.26510.2392.5213.42411.9169.2112.31210.7654.67712.3032.56010.9748.2863.1812.4184.7902.0165.8933.6073.03211.2011.5703.222