TheoriaApofaseonProject2022

In this project we developed a heart disease prediction system using the UCI heart disease dataset for training and a multitude of machine learning predictors like
Naive-Bayes,K-Nearest-Neighbors,Support Vector Machine and a custom made neutral network optimized for this spesific application.We conducted a statistical analysis 
of the input features of the dataset and removed detected feature class outliers after detecting them with methods like z-score.We also created a special function 
with which we perfomed the classification of the training set with the various machine learning predictors and allowed us to standardize or normalize the data before 
the prediction so we can access the different scenarios and scoring that would result from the training.To access the training that was conducted we used measurment 
score likes F1-score,precision,recall,accuracy as well as ROC curves.After training our machine learning model with the various predictors and also with the 2 
preprocessing methods being used in different scenarios (standardization or normalization) we then visualized our results with graphical represantantions of the 
measurment scores we mentioned above as well as comparison graphs of these metrics for each different preprocessing method used.