Artificial Neural Network-Founded Big Data Solution for Accurate Heart Disease Analysis and Cataloguing

In this section, an Optimal Artificial Neural Network (OANN) model is used to present the concept of diagnosing heart disease (HD) by the utilisation of large amounts of data. The two cornerstones of the proposed OANN are a model that is referred to as TLBO-ANN, which is an acronym that signifies Teaching-Learning-Based Optimisation, and Data-Based Misclassification Reduction (DBMIR). As a result of its construction on top the popular Apache Spark structure, the algorithm is capable of making predictions both offline and online. In order to create accurate predictions for HD using the trained model, the OANN technique is first taught and evaluated offline on an example set for coronary artery disease (CAD). After that, it is utilised online to deal with current information that is given using Apache Spark. Following rigorous testing, we are able to confirm that the model is capable of managing increasing amounts of knowledge on individuals and even suggest some improvements to do analysis on a variety of conditions. According to the results of the section, selecting attributes and reduction techniques is expected to render the OANN framework more efficient for large-scale HD scheduling. This might result in categorised that is both precise and more expedient. With an accuracy of $96.42 \%$, OANN outperforms competing approaches like Naive Bayes Tree (NBtree), J48, Random Forest (RF), and TLBO-ANN.