A Hybrid Model- and Memory-Based Collaborative Filtering Algorithm for Baseline Data Prediction of Friedreich's Ataxia Patients

Friedreich's ataxia (FRDA) is the most common inherited that causes progressive damage of nervous systems and performance deterioration physical movements. FRDA baseline data analysis plays a crucial role in advancing disease research, where main obstacle comes from collection primarily due to degenerative symptoms patients. Inspired by nowadays popular collaborative filtering (CF) method, new algorithm proposed this article, with which patients (or their families) are only required provide certain reliable acquired home uncertain/missing parts can then be predicted acceptable accuracy utilizing existing patient information. The framework constructed based on novel hybrid model combining merits model- memory-based CF methods, thereby facilitating improved prediction accuracy. exhibits following two features: when does not have neighbors sharing similar data, model-based component activated employ clustering method find attributes; case neighbors, similarity measure, accounts for more statistical characteristics integrating rating habits degree co-rated items, developed order adjust initial similarities between To evaluate advantages algorithm, Scale Assessment Rating Ataxia selected European Consortium Translational Studies database. Experimental results demonstrate our approach superior other conventional approaches.