Key result
An optimized hybrid SVM-KNN model using reduced features improves classification performance and cuts training time.
Why the study?
Larger medical datasets complicate understanding feature connections for early disease diagnosis, creating a need to simplify heart disease classification by identifying key risk variables from high-dimensional data.
Does a hybrid SVM-KNN model optimized with GSO and a reduced feature subset improve heart disease classification performance compared to models using a complete feature set?
Does a hybrid SVM-KNN model optimized with GSO and a reduced feature subset improve heart disease classification performance compared to models using a complete feature set?
A hybrid SVM-KNN machine learning model using feature reduction and GSO optimization improves the efficiency and accuracy of heart disease classification.
The convergence of the IoT and the Cloud makes our lives better by allowing for frictionless communication between humans and inanimate objects. Projecting analytics in the medical arena may help transform a sensitive healthcare one, and cutting-edge artificial intelligence technologies are increasingly making their way into the healthcare industry. In order to provide early intervention and preventative therapy to people at risk, it is crucial to accurately predict the beginning of diseases. As the use of electronic clinical data spreads, it becomes more important than ever to refine existing prediction models. With larger medical datasets, clinicians have to put in more effort to understand complicated feature connections and diagnose illnesses in advance. Therefore, this research study simplifies the process of classifying heart disease by identifying the most important risk variables from a high-dimensional dataset. We have employed two datasets with a wide range of clinical characteristics for our investigation. To begin, we examined the interplay and dependency between a variety of clinical characteristics and coronary heart disease. Last but not least, we looked into a hybrid machine learning classification model (Support Vector Machine with K-Nearest Neighbour) by testing it with both full and reduced feature subsets. Accuracy, ROC curve, and F1-Score were used to assess the trained classifiers, while Galactic Swarm Optimisation (GSO) was used to choose the appropriate kernel weight of the SVM. When associated to models trained on a complete feature set, the presentation of classification models trained with a slighter feature set enhanced melodramatically with a reduced amount of training period.
No takes yet. Share an insight, caveat, or question.
Sudhakar et al. (2023) studied Coronary heart disease. Hybrid machine learning classification model (SVM with KNN) optimized with Galactic Swarm Optimisation (GSO) using a reduced feature set vs. Models trained on a complete feature set was evaluated on Classification performance (Accuracy, ROC curve, and F1-Score). A hybrid machine learning classification model (SVM with KNN) optimized with Galactic Swarm Optimisation using a reduced feature set improved classification performance and reduced training time.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: