Role Of Machine Learning and Random Forest in Accuracy Enhancement During Asthma Prediction

• Published in: 2022 10th International Conference on Reliability, Infocom Technologies and Optimization (Trends and Future Directions) (ICRITO) pp. 1 - 10
• Main Authors: Rani, Anchal, Sehrawat, Hakesh
• Format: Conference Proceeding
• Language: English
• Published: IEEE 13.10.2022
• Subjects: Accuracy; Asthma Prediction; F1-Score; Fault diagnosis; Machine learning; Machine learning algorithms; nd Sensitivity; Predictive models; Random Forest; Respiratory system; Sensitivity; Sociology; Specificity; Statistics
• DOI: 10.1109/ICRITO56286.2022.9965149

It is noticed that relatively little research have employed the machine learning prediction approach to diagnose asthma in among individuals. If ML methods are applied for identification of asthma then the identification will be very much effective and also flawless. A very essential aspect of the study will be examining the function of ML in the prediction of asthma. ML may be utilized for the evaluation of the association between the incidence of the illness and the quality of the environment. This suggests that accurate ML approaches were utilized to diagnose asthma among diverse people. The discovery of several factors that would be employed by machine learning algorithms to diagnose asthma was therefore another significant objective of the research. These traits were also revealed as a consequence of the research that involved diagnosing asthma using ML methods. One of the key research topics of the study was to establish the sort of ML algorithm that is a good proportion for the diagnosis of asthma. This information was also identified in the literature study. It is obvious that ML methods are vital for detecting asthma; nevertheless additional processes and learning techniques must be integrated for a correct diagnosis. The CAPE & CAPP models were able to outperform identical models constructed using standard logistic regression-based methodologies by incorporating ML. In contrast to conventional models that were based on CAPE and CAPP models., the suggested work employs a random forest classifier and delivers improved accuracy., specificity., f1-score, and sensitivity.