Hybrid Context-Aware Recommendation System For Hypertension Problem رسالة ماجستير

Abstract

This research aims to develop an intelligent recommendation system based on arti ficial intelligence to provide personalized and data-driven recommendations for hyper tension management. The proposed system, named Hybrid Context-Aware Recommen dation System, integrates two complementary methodologies Collaborative Filtering and Content-Based Filtering within a unified scientific and practical framework. The study was conducted using an applied analytical methodology that combined soft ware design with experimental evaluation. The study population relied on real-world data from the Kaggle Health Dataset, which includes clinical, demographic, and behavioral at tributes of adult individuals. The study sample was selected and preprocessed according to medical standards, incorporating blood pressure levels, cholesterol categories, cardio vascular risk levels, age, gender, and lifestyle factors. Multiple research tools were employed, including data analysis, feature engineering, and context extraction techniques, in addition to performance evaluation metrics such as precision, recall, and ranking quality. The results revealed that the proposed hybrid model outperformed the standalone Collaborative Filtering and Content-Based Filtering models, achieving Precision 0.72 and Recall 0.90, indicating highly accurate and comprehensive recommendations. The system also demonstrated strong interpretability through an inter active dashboard built on the Streamlit platform, enhancing transparency and applicability in digital healthcare environments. The study recommends adopting hybrid context-aware systems in clinical decision sup port due to their ability to combine personalization, reliability, and clinical interpretability. It further encourages expanding the dataset to include multi-institutional medical records and applying the same framework to other chronic diseases such as diabetes and cardio vascular disorders