Synthetic Turkish Clinical Event Dataset for Temporal Risk Prediction, Sequence Modeling, and NLP

This dataset contains a synthetic Turkish clinical event dataset designed for temporal risk prediction, sequence modeling, and healthcare-related machine learning research. The dataset simulates longitudinal patient records, including visits, diagnoses, lab tests, reports, and prescriptions within a structured temporal framework. Key characteristics:- 5,000 synthetic patients- 108,000+ clinical events- Temporal event structure with irregular visit intervals- Category-based clinical signals (labs, diagnoses, reports, prescriptions)- Latent risk modeling using a causal generative process The data generation process incorporates:- Beta-distributed baseline risk- Ornstein-Uhlenbeck temporal dynamics- Poisson-based event intensity- Category-specific signal amplification Potential applications:- Temporal risk prediction- Sequence modeling (RNN, Transformer)- Anomaly detection in healthcare data- Clinical trajectory analysis For valid modeling:- Temporal splits must be used- Latent variables (e.g., base_risk) must not be used as features License: Open for research and educational use.DATA DISCLAIMER:This dataset is fully synthetic and does not contain any real patient data. It is generated using a probabilistic simulation framework and is intended strictly for research, experimentation, and educational use.