The proposed work shows how important and useful testing for any machine learning application and need to make better models such that bugs and errors are minimized. The proposed work considers patient’s health data which can be used for decision making or prediction using various
calculation, in this work, developing a heart condition prediction system mainly concentrating on artificial neural network, which uses the multilayer perceptron algorithm for the execution. Our dataset consists of vital information about the patient such as age, gender, blood pressure, ECG
measures, and stroke history. This labeled dataset predicts the probability of a patient to have heart diseases or not. To fulfill the proposed work, created a GUI to get the information about a new patient. Once the development of the model finished, complete functionality testing was applied.
The report of the testing helped in identifying the bugs and errors which on rectification by the developer helped in increasing the accuracy of the artificial neural network.
A machine learning approach for indirect human presence detection using IOT devices
