There is an enormous increase in digital information, such as biomedical literature, clinical notes, etc. Clinical notes contain valuable information about a patient’s symptoms, diagnosis, treatment, etc., which could greatly help health care. Retrieving relevant information efficiently and effectively from huge amount of data is a challenge. The paper proposed a hybrid model, named NNKM using clustering as a tool for clustering search results. Clustering search results in context of health care data improves relevant information retrieval, thereby fostering improved clinical decision making, and patient care outcome. The NNKM model integrates non-negative matrix factorization and the K-means clustering tehnique. We compare the results of NNKM model with the results of state-of-the-art techniques. The experimental results demonstrate that the NNKM model achieves a promising result, achieving a highest silhouette score, i.e., 0.054, and Computation time 0.096 in seconds.