Dataset for: Systematic Literature Review of Health Recommender Systems (136 Studies)

Published: 11 February 2026| Version 1 | DOI: 10.17632/krfvhrkwpn.1
Contributors:
Oumaima EL MIAYAR , Abdelaziz berrado

Description

This table summarizes the systematic extraction and coding of 136 peer-reviewed Health Recommender Systems (HRS) published between 2014 and 2025. Each study was analyzed using a lifecycle-oriented analytical framework composed of six complementary pillars, operationalized through a total of 20 explicit analytical attributes. The six pillars capture key dimensions of HRS design and translation, including: (i) Clinical need and intended use (Clinical Domain, Clinical Task, Target User,Intended Use), (ii) Data acquisition and governance (Data Source Type, Data Sensitivity, Governance Reporting, Reproducibility Signal), (iii) Recommendation logic and output semantics (Recommendation Paradigm, Output Type, Explainability Level), (iv) User experience and interaction (Delivery Platform, User Interaction, Human-in-the-loop), (v) Evaluation and validation strategy (Evaluation Level, Outcome Focus), (vi) Translation, clinical integration, and ethical operationalization (Clinical Integration, Deployment Stage, Integration Barriers, Ethical Operationalization).

Files

Categories

Systematic Review, Recommendation System, Digital Health

Licence