FBGN: A FUSION BASED GENETIC PSO NETWORK FOR HEART ATTACK PREDICTION

Abstract

The most prevalent disease is heart disease. The most challenging task is predicting cardiac disease. Although there exist tools to predict cardiac disease, they are either more pricey or unsuccessful at doing so. We may be able to reduce the death rate somewhat if we can immediately predict cardiac problems. We have access to many data in the modern world, allowing us to anticipate employing machine learning methods to study heart illness. The existing model used to forecast cardiac problems is based on a Customised ensemble classification technique that uses a decision tree, SVM and logistic regression technique. The following are some of the system's drawbacks: more classifiers should first be evaluated second because the suggested model is based on the wrapper feature selection technique, and its high data processing price and temporal complexity as the main limitations. One option is to use a genetic algorithm. Minor data restrictions are the second. So, to rectify drawbacks in this project, use a genetic algorithm with particle swarm optimisation. This paper chooses the optimal attributes for predicting heart illness with numerous equations. Using numerous equations to forecast cardiac disease is innovative in this work.