Conditions For Cardiovascular Diseases Prediction

Abstract

data  mining  technology. Today  data  mining has  lots  of  application in  every  aspects  of human  life. Applications  of  data  mining  are  wide  and  diverse.  Among  this  health  care  is  a major  application  of  data  mining.  Medical  field  has  get  benefited  more  from  data  mining.  Heart  Disease  is  the  most  dangerous  life-threatening  chronic  disease  globally.  The  objective  of  the  work  is  to  predicts  the occurrence of heart disease of a patient using random forest algorithm. The dataset was accessed from Kaggle site. The  dataset contains 303 samples and  14  attributes are taken for  features of the dataset.

Then  it  was  processed  using  python  open  access  software  in  jupyter  notebook.  The  datasets  are classified  and  processed  using  machine  learning  algorithm  Random  forest.  The  outcomes  of  the dataset  are  expressed in  terms  of accuracy,  sensitivity  and  specificity in  percentage.  Using  random forest algorithm,  we obtained accuracy of 86.9% for prediction of  heart disease with sensitivity value 90.6%  and  specificity  value  82.7%.  From  the  receiver  operating  characteristics,  we  obtained  the diagnosis  rate  for  prediction  of  heart  disease  using  random  forest  is  93.3%.  The  random  forest algorithm has proven to be the most efficient algorithm for classification of heart disease and therefore it is used in the proposed system.

Large amount of data can be extracted by a technique known as Data Mining Technology. In day to day human life, Data mining is widely used. It has created a vertical and significant role in field of health care. Data mining has been increasingly advantageous for the medical industry. The deadliest chronic condition in the world that can cause death is heart disease. The death rate can be decreased by conducting the early diagnosis of heart disease. This programme aids in the early diagnosis and prediction of heart disease. Healthcare organisations nowadays produce enormous amounts of data, yet those data are incredibly disorganised. This data can be used to forecast cardiac illnesses with simplicity if it is properly organised using data mining techniques. The goal of this research is to use the Random Forest and Decision Tree algorithms to forecast if a patient may develop cardiac disease. From GitHub, the dataset was accessed.  The data was then processed in a Jupyter notebook using open-source Python tools. Using the machine learning algorithms Random Forest and Decision Tree, the datasets are categorised and processed.