Prediction of Individual Drug Addiction Vulnerability using Machine Learning Techniques

Objective

The main objective of this project is to develop a machine learning–based prediction model that can detect individual risk factors leading to substance addiction at an early stage and contribute to intervention strategies aimed at prevention. By identifying individuals with a tendency toward addiction before dependency develops, the project seeks to strengthen preventive and protective public health policies.

Scope

Within the scope of the project, the role of demographic, socioeconomic, and psychosocial factors in substance use disorders outside of medical indications will be examined. Data collected from healthy individuals and individuals diagnosed with substance addiction will be analyzed using various machine learning algorithms to estimate individual vulnerability. This research, to be conducted in Türkiye, will align with national strategies for combating addiction and adopt a multidisciplinary approach.

Expected Outcomes

• Development of predictive models enabling early identification of individuals prone to substance addiction
• Strengthening the capacity of healthcare professionals for early intervention by systematically determining risk factors
• Supporting data-driven, evidence-based prevention policies against addiction
• Providing a methodological framework that can later be adapted to other types of addictions (e.g., alcohol, tobacco)
• Raising public awareness of substance addiction and contributing to community health

This project directly aligns with the objectives of the 2024–2028 National Strategy and Action Plan on Combating Drugs and is being carried out under the coordination of TÜSEB in collaboration with relevant institutions. Project activities have been officially initiated and have reached a stage worthy of being shared in the public interest.