Big data growth in the healthcare community, accurate analysis of medical data supports early disease detection, patient care and community services. However, the accuracy of the analysis decreases when the quality of the medical data is incomplete. Feature selection is a process that selects a subset of features that are relevant for a predictive modeling problem. This method can identify and remove unnecessary, irrelevant, and redundant attributes from the dataset, which do not contribute to the accuracy of the model or reduce the accuracy of the model. The challenge in research in the field of medical records is in structured and unstructured data which results in a method being needed to assist the algorithm in selecting good features. This paper provides an overview of the proposed hybrid method that can be used for feature selection algorithms in the field of medical records.

Link Publikasi : https://greenpub.org/IJPHS/article/view/339