Medical Papers Cluster Supported Medication-Symptom Names Using Multi Read Nonnegative Matrix Factorization

Abstract

Clinical documents are rich free-text knowledge sources containing valuable medication and symptom data, that have a great potential to enhance health care. Existing system, a brand new convolutional neural network primarily based multimodal disease risk prediction algorithmic rule mistreatment structured and unstructured knowledge from hospital. To the simplest of our information, none of the existing work focused on each knowledge types within the space of medical big knowledge analytics. Compared to many typical prediction algorithms, the prediction accuracy of our proposed algorithm reaches with a convergence speed. Proposed system, we tend to build an integrating system for extracting medication names and symptom names from clinical notes. Then we apply nonnegative Matrix factorization (NMF) and multi-view NMF to cluster clinical notes into purposeful clusters supported sample-feature matrices. Our experimental results show that multi-view NMF could be a preferred methodology for clinical document cluster. Moreover, we discover that using extracted medication symptom names to cluster clinical documents outperforms simply using words.