Exploratory analysis on heart attack

Data Description

age : Age of the patient

sex : Sex of the patient
	1 = male
	0 = female

exng: exercise induced angina (1 = yes; 0 = no)

cp : Chest Pain type chest pain type
	Value 1: typical angina
	Value 2: atypical angina
	Value 3: non-anginal pain
	Value 4: asymptomatic

trtbps : resting blood pressure (in mm Hg)

chol : cholestoral in mg/dl fetched via BMI sensor

fbs : (fasting blood sugar > 120 mg/dl) (1 = true; 0 = false)

rest_ecg : resting electrocardiographic results

	Value 0: normal 
	Value 1: having ST-T wave abnormality (T wave inversions and/or ST elevation or depression
		of > 0.05 mV) 
	Value 2: showing probable or definite left ventricular hypertrophy by Estes' criteria

thalach : maximum heart rate achieved

exang: exercise induced angina
	1 = yes
	0 = no
	
oldpeak = ST depression induced by exercise relative to rest
slope: the slope of the peak exercise ST segment
	Value 0: upsloping
	Value 1: flat
	Value 2: downsloping

ca: number of major vessels (0-3)

thal:
	0 = error (in the original dataset 0 maps to NaN's)
	1 = fixed defect
	2 = normal
	3 = reversable defect

target :
	0= less chance of heart attack
	1= more chance of heart attack

Statistical description of data is in report.html

For Report : click

general statistical description

pearson correlation

exang induced on gender

male = 1
exercise induced angina have a high possibility in Male then female.

Frequency of different type on chest pain

Typical-Angina have the highest possibility of occurrence
,also NON-Anginal pain have high possibility.

How cholesterol affects chest pain type?

individuals having chol around 200-275 have the most possibility of having chest pain.

chol and age

since age and chol have a +ve co-relation as the age inc. chol tends to inc.

caa relation ship with chance of heart attack

(0,4) have a significant high chance of heart attack.
(1,2,3) have a significant low chance of heart attack.

How max heart rate affects the probability of having heart attack?

There's positive correlation co-efficient of 0.421741 ("pearson") between them,
so we can assume that having a higher heart rate is a factor contributing to heart attack.

Model to predict heart attack

train_data = data.drop(['output'], axis=1)

X_train, X_test, y_train, y_test = train_test_split(
        train_data, data['output'], random_state=0)

knn = KNeighborsClassifier(n_neighbors=4)
knn.fit(X_train, y_train)
prediction = knn.predict(X_test)

print(np.mean(prediction == y_test))

accuracy of model => 0.6842105263157895 or 68.42%

increase accuracy to 86% using auto-sklearn