Abstract

Recently, machine learning has become a hot research topic. Therefore, this study investigates the interaction between software engineering and machine learning within the context of health systems. We proposed a novel framework for health informatics: the framework and methodology of software engineering for machine learning in health informatics (SEMLHI). The SEMLHI framework includes four modules (software, machine learning, machine learning algorithms, and health informatics data) that organize the tasks in the framework using a SEMLHI methodology, thereby enabling researchers and developers to analyze health informatics software from an engineering perspective and providing developers with a new road map for designing health applications with system functions and software implementations. The SEMLHI approach utilizes the principal component analysis (PCA) for feature extraction and feature reduction. Further, SEMLHI model also utilizes the extreme learning machine (ELM) for prediction problems. The SEMLHI approach considers the Indian Diabetes dataset to perform the simulations, and proposed ELM is outperformed as compared to state of art approaches

Description