Optimizing Healthcare center Discovery using clustering and association mining-based Recommendation System: A Library Approach

Abstract

The advent of personalized recommendation systems has revolutionized numerous sectors, including healthcare. This paper introduces a novel recommendation system aimed at helping users discover suitable healthcare centers by utilizing clustering and association mining techniques. Our approach integrates these methodologies to improve the accuracy and relevance of recommendations, particularly in addressing the challenges associated with healthcare center selection. The system employs clustering to group healthcare centers with similar attributes, while association mining is used to uncover hidden patterns and relationships that enhance recommendation quality. We assess the effectiveness of our system through empirical testing, benchmarking its performance against traditional recommendation approaches. One of the key challenges addressed in this paper is the Cold Start issue and data Sparsity, which commonly hinder recommendation systems. To overcome these challenges, we propose a hybrid method that combines data clustering with the Eclat Algorithm to generate more effective rules. Initially, we cluster the rating matrix based on user similarity. Subsequently, we convert the clustered data into Boolean form and apply the Eclat Algorithm to the Boolean data, resulting in a more refined and efficient recommendation algorithm.