---
id: "diabetes"
slug: "diabetes"
name: "Diabetes"
domain: "Healthcare & Biomedicine"
taskType: "binary_classification"
numRows: 768
numColumns: 9
targetColumn: "diabetes_diagnosis"
primaryMetric: "1 - AUC Error"
primaryScore: "0.1632"
parquetUrl: "https://huggingface.co/datasets/carlahq/demo-tabular-benchmarks/resolve/main/data/diabetes/diabetes.parquet"
spreadsheetUrl: "https://docs.google.com/spreadsheets/d/1CzV5tH7CyqC4Qzi7zQQtB7BvWbXHR2cXQhF3cOKr1Xo/edit?usp=sharing"
canonicalUrl: "https://carlahq.eu/datasets/diabetes"
---

# Dataset: Diabetes (diabetes)

> Clinical diagnostic benchmark comprising 768 patient records designed to predict the onset of diabetes mellitus in high-risk adult females based on diagnostic and metabolic biomarkers.

## Dataset Overview

| Attribute | Value |
| :--- | :--- |
| **Task Type** | `binary_classification` |
| **Business Domain** | Healthcare & Biomedicine |
| **Target Column** | `diabetes_diagnosis` |
| **Rows (Instances)** | 768 |
| **Features (Columns)** | 9 |
| **Primary Evaluation Metric** | 1 - AUC Error |
| **Best Benchmark Score** | **0.1632** |

## Business Decision Context
**Context:** Primary care physicians, metabolic health clinics, and health insurance providers seek early-stage risk stratifications to identify individuals at elevated risk of Type 2 diabetes before irreversible vascular or organ damage occurs.
**Objective:** Accurately predict binary diabetes onset (tested_positive vs. tested_negative) utilizing non-invasive demographic factors and routine clinical lab measurements.
**Decision Impact:** Enables proactive preventative interventions (lifestyle programs, Metformin therapy) reducing long-term chronic care expenditure by up to 60%. Mitigates severe false negatives (delayed diagnosis leading to neuropathy, nephropathy, and cardiovascular events) while controlling false positives to avoid unnecessary clinical anxiety and second-line diagnostic costs.

## Feature Schema
| Column Name | Data Type | Role | Distinct Values | Missing Values (%) | Description |
| :--- | :--- | :--- | :--- | :--- | :--- |
| `diabetes_diagnosis` | `categorical` | target | 2 | 0 | Binary diagnostic indicator reflecting whether the patient was tested positive ('tested_positive') or negative ('tested_negative') for diabetes according to WHO criteria. |
| `pregnancy_count` | `numeric` | feature | 17 | 0 | Total number of times the patient has been pregnant; associated with gestational metabolic strain and insulin resistance. |
| `glucose_concentration_2h` | `numeric` | feature | 136 | 0 | Plasma glucose concentration measured 2 hours after an oral glucose tolerance test (OGTT). Key diagnostic indicator where values >= 200 mg/dL indicate diabetes. |
| `diastolic_blood_pressure` | `numeric` | feature | 47 | 0 | Diastolic blood pressure reading; elevated values are correlated with metabolic syndrome and systemic vascular stiffness. |
| `triceps_skinfold_thickness` | `numeric` | feature | 51 | 0 | Measurement of the fold of skin and subcutaneous fat over the triceps muscle, serving as an anthropometric proxy for peripheral body fat percentage. |
| `serum_insulin_2h` | `numeric` | feature | 186 | 0 | Serum insulin level measured 2 hours after glucose loading; assesses pancreatic beta-cell response and peripheral insulin sensitivity. |
| `body_mass_index` | `numeric` | feature | 248 | 0 | Body mass index calculated as weight in kilograms divided by height in meters squared (kg/m²), indicating obesity levels. |
| `diabetes_pedigree_function` | `numeric` | feature | 517 | 0 | Continuous genetic scoring function synthesizing family history of diabetes to estimate genetic risk predisposition. |
| `patient_age_years` | `numeric` | feature | 52 | 0 | Chronological age of the patient in years (all subjects are females >= 21 years old). |

## Empirical Benchmark Leaderboard
| Rank | Algorithm / Model | Model Family | Runtime | Accuracy | ROC AUC | F1-Score |
| :--- | :--- | :--- | :--- | :--- | :--- | :--- |
| 1 | **AttributeSelectedClassifier HoeffdingTree** | Decision Tree | Python | **77.34%** | 0.8190 | 0.7665 |
| 2 | **ClassificationViaRegression M5P** | Decision Tree | Python | **77.34%** | 0.8304 | 0.7656 |
| 3 | **AttributeSelectedClassifier SMO PolyKernel** | Support Vector Machine (SVM) | Python | **77.08%** | 0.7132 | 0.7592 |
| 4 | **AttributeSelectedClassifier Bagging JRip** | Bagging Ensemble | Python | **77.08%** | 0.7984 | 0.7615 |
| 5 | **AttributeSelectedClassifier Bagging LMT** | Bagging Ensemble | Python | **77.08%** | 0.8248 | 0.7666 |
| 6 | **AttributeSelectedClassifier LMT** | Machine Learning Model | Python | **77.08%** | 0.8212 | 0.7622 |
| 7 | **RandomForest** | Random Forest | Python | **76.95%** | 0.8312 | 0.7679 |
| 8 | **Bagging REPTree** | Decision Tree | Python | **76.95%** | 0.8270 | 0.7648 |
| 9 | **KernelLogisticRegression RBFKernel** | Logistic / Linear Model | Python | **76.82%** | 0.8381 | 0.7618 |
| 10 | **SMO PolyKernel** | Support Vector Machine (SVM) | Python | **76.82%** | 0.7129 | 0.7577 |
| 11 | **TabPFN v3** | Tabular Foundation Model | PyTorch/Python | **76.73%** | 0.8393 | 0.6354 |
| 12 | **EXAONE Tabular** | Tabular Foundation Model | PyTorch/Python | **76.50%** | 0.8362 | 0.6332 |
| 13 | **Google TabFM v1.0** | Tabular Foundation Model | PyTorch/Python | **76.42%** | 0.8377 | 0.6335 |
| 14 | **nanotabicl Vanilla** | Tabular Foundation Model | PyTorch/Python | **76.20%** | 0.8368 | 0.6252 |
| 15 | **Streaming nanotabicl** | Tabular Foundation Model | PyTorch/Python | **76.16%** | 0.8368 | 0.6248 |

### Repeated Cross-Validation Fold Scores (TabICLv2 Reference)
| Fold | Train Samples | Test Samples | Metric Score | Latency |
| :--- | :--- | :--- | :--- | :--- |
| Fold 0 | 512 | 256 | 0.1341 | 696ms |
| Fold 1 | 512 | 256 | 0.1663 | 643ms |
| Fold 2 | 512 | 256 | 0.1802 | 649ms |
| Fold 0 | 512 | 256 | 0.1720 | 630ms |
| Fold 1 | 512 | 256 | 0.1240 | 646ms |
| Fold 2 | 512 | 256 | 0.1850 | 640ms |
| Fold 0 | 512 | 256 | 0.1772 | 654ms |
| Fold 1 | 512 | 256 | 0.1497 | 643ms |
| Fold 2 | 512 | 256 | 0.1636 | 656ms |
| Fold 0 | 512 | 256 | 0.1579 | 634ms |
| Fold 1 | 512 | 256 | 0.1882 | 656ms |
| Fold 2 | 512 | 256 | 0.1520 | 633ms |
| Fold 0 | 512 | 256 | 0.1948 | 653ms |
| Fold 1 | 512 | 256 | 0.1483 | 638ms |
| Fold 2 | 512 | 256 | 0.1453 | 657ms |
| Fold 0 | 512 | 256 | 0.1350 | 640ms |
| Fold 1 | 512 | 256 | 0.1998 | 656ms |
| Fold 2 | 512 | 256 | 0.1683 | 638ms |
| Fold 0 | 512 | 256 | 0.1606 | 680ms |
| Fold 1 | 512 | 256 | 0.1720 | 670ms |
| Fold 2 | 512 | 256 | 0.1399 | 644ms |
| Fold 0 | 512 | 256 | 0.1788 | 640ms |
| Fold 1 | 512 | 256 | 0.1539 | 657ms |
| Fold 2 | 512 | 256 | 0.1551 | 643ms |
| Fold 0 | 512 | 256 | 0.1714 | 679ms |
| Fold 1 | 512 | 256 | 0.1508 | 635ms |
| Fold 2 | 512 | 256 | 0.1580 | 674ms |
| Fold 0 | 512 | 256 | 0.1585 | 643ms |
| Fold 1 | 512 | 256 | 0.1985 | 645ms |
| Fold 2 | 512 | 256 | 0.1576 | 661ms |

## Provenance & Downloadable Artifacts
- **Dataset Parquet File:** [Direct Parquet Download](https://huggingface.co/datasets/carlahq/demo-tabular-benchmarks/resolve/main/data/diabetes/diabetes.parquet)
- **Google Sheets Interactive Demo:** [Open in Google Sheets](https://docs.google.com/spreadsheets/d/1CzV5tH7CyqC4Qzi7zQQtB7BvWbXHR2cXQhF3cOKr1Xo/edit?usp=sharing)
- **OpenML Source:** [OpenML Dataset](https://www.openml.org/d/37)
- **Academic Paper / Citation:** [Read Research Paper](https://www.jair.org/index.php/jair/article/view/10129)
- **Canonical HTML Page:** [https://carlahq.eu/datasets/diabetes](https://carlahq.eu/datasets/diabetes)