Evaluation and Explanation of Next Best Action Recommendations in Healthcare - overview
Evaluation and Explanation of Next Best Action Recommendations in Healthcare
Description: "Recommender systems are one of the most used classes of machine learning techniques. While they are widely used across the most diverse business applications, recommender systems have not yet been extensively adopted and successfully leveraged in healthcare scenarios. Challenges remain in bringing diverse data sources and producing explainable and interpretable results. Strict and evolving regulations regarding patient privacy healthcare systems face another challenge in gaining both patients’ and doctors’ trust in these systems.
The objective of the project is to evaluate the effectiveness and performance of two prebuilt recommender systems in healthcare, suggest improvements and investigate explainability of the results.
- Strong proficiency in Java
- Good understanding of recommendation systems, knowledge graphs and neural networks.
- (Preferred) Knowledge of system evaluation and evaluation statistics