Berkson Probit Regressive Adaptive Boost Whisper Data Clustering For Healthcare Data Analytics

• Published in: 2024 International Conference on Cognitive Robotics and Intelligent Systems (ICC - ROBINS) pp. 564 - 571
• Main Authors: Sathishkumar, S., Parameswari, P.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 17.04.2024
• Subjects: Adaptive Boost Whisper; Berkson Probit Regression; Classification algorithms; Clustering algorithms; Data analysis; Data Clustering; Decision making; Ensemble Learning; Feature extraction; Healthcare Data Analytics; Machine Learning; Performance analysis; Prediction algorithms
• DOI: 10.1109/ICC-ROBINS60238.2024.10533990

Data analytics involves analysing raw data to ascertain inclinations and ensure finer decision-making. Data analytics applies to numerous applications, organisations, and health care. Healthcare data analytics refers to exploring prevailing historical data to predict trends enhancing outreach and control in disease advancements. When integrated with business intelligence and data visualisation, healthcare data analytics assists managers in performing better by providing information that supports precise decision-making and delivers actionable perceptions. Many researchers and academicians have recently proposed Machine Learning (ML) algorithms for healthcare data analytics to improve assessment accuracy. In recent studies, Ensemble Learning (EL) techniques have played a crucial role in healthcare data analysis of chronic diseases like cardiovascular and have achieved good performance metrics. Although most prevailing methods worked independently and generated an admissible result, they were still short of optimal accuracy. This article presents an ensemble learning method called Berkson Probit Regressive Adaptive Boost Whisper Data Clustering (BPR-ABWDC) for healthcare data analytics. In the proposed BPR-ABWDC method, the outcomes of two algorithms, Berkson Probit Regressive Feature Selection and Adaptive Boost Whisper Data Clustering, are integrated. The weighted voting model is used for aggregating precision. Experimental investigations of the proposed BPR-ABWDC method employing the cardiovascular disease dataset confirm that the performance of the proposed method is comparatively better than the traditional methods in terms of higher precision by 10%, recall by 16%, and reduced time by 32%, respectively.