In [ ]:
df.corr()['stroke'][:-1].sort_values().plot(kind='bar')
Out[ ]:
<Axes: >
#Importing the libraries
import numpy as np
import matplotlib.pyplot as plt
import pandas as pd
import seaborn as sns
from sklearn.preprocessing import LabelEncoder
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LogisticRegression
from sklearn.tree import DecisionTreeClassifier
from sklearn.svm import SVC
from sklearn.neighbors import KNeighborsClassifier
from sklearn import metrics
from sklearn.metrics import accuracy_score
from sklearn.metrics import mean_absolute_error
from sklearn.metrics import f1_score
from sklearn.metrics import mean_squared_error
from sklearn.metrics import log_loss
#Loading the dataset
df = pd.read_csv('healthcare-dataset-stroke-data.csv')
df.head()
| id | gender | age | hypertension | heart_disease | ever_married | work_type | Residence_type | avg_glucose_level | bmi | smoking_status | stroke | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 9046 | Male | 67.0 | 0 | 1 | Yes | Private | Urban | 228.69 | 36.6 | formerly smoked | 1 |
| 1 | 51676 | Female | 61.0 | 0 | 0 | Yes | Self-employed | Rural | 202.21 | NaN | never smoked | 1 |
| 2 | 31112 | Male | 80.0 | 0 | 1 | Yes | Private | Rural | 105.92 | 32.5 | never smoked | 1 |
| 3 | 60182 | Female | 49.0 | 0 | 0 | Yes | Private | Urban | 171.23 | 34.4 | smokes | 1 |
| 4 | 1665 | Female | 79.0 | 1 | 0 | Yes | Self-employed | Rural | 174.12 | 24.0 | never smoked | 1 |
df.drop('id', axis=1, inplace=True)
df.describe()
| age | hypertension | heart_disease | avg_glucose_level | bmi | stroke | |
|---|---|---|---|---|---|---|
| count | 5110.000000 | 5110.000000 | 5110.000000 | 5110.000000 | 4909.000000 | 5110.000000 |
| mean | 43.226614 | 0.097456 | 0.054012 | 106.147677 | 28.893237 | 0.048728 |
| std | 22.612647 | 0.296607 | 0.226063 | 45.283560 | 7.854067 | 0.215320 |
| min | 0.080000 | 0.000000 | 0.000000 | 55.120000 | 10.300000 | 0.000000 |
| 25% | 25.000000 | 0.000000 | 0.000000 | 77.245000 | 23.500000 | 0.000000 |
| 50% | 45.000000 | 0.000000 | 0.000000 | 91.885000 | 28.100000 | 0.000000 |
| 75% | 61.000000 | 0.000000 | 0.000000 | 114.090000 | 33.100000 | 0.000000 |
| max | 82.000000 | 1.000000 | 1.000000 | 271.740000 | 97.600000 | 1.000000 |
df.info()
<class 'pandas.core.frame.DataFrame'> RangeIndex: 5110 entries, 0 to 5109 Data columns (total 11 columns): # Column Non-Null Count Dtype --- ------ -------------- ----- 0 gender 5110 non-null object 1 age 5110 non-null float64 2 hypertension 5110 non-null int64 3 heart_disease 5110 non-null int64 4 ever_married 5110 non-null object 5 work_type 5110 non-null object 6 Residence_type 5110 non-null object 7 avg_glucose_level 5110 non-null float64 8 bmi 4909 non-null float64 9 smoking_status 5110 non-null object 10 stroke 5110 non-null int64 dtypes: float64(3), int64(3), object(5) memory usage: 439.3+ KB
df['age'].astype(int)
0 67
1 61
2 80
3 49
4 79
..
5105 80
5106 81
5107 35
5108 51
5109 44
Name: age, Length: 5110, dtype: int32
#Checking for null values
df.isnull().sum()
gender 0 age 0 hypertension 0 heart_disease 0 ever_married 0 work_type 0 Residence_type 0 avg_glucose_level 0 bmi 201 smoking_status 0 stroke 0 dtype: int64
#replacing the missing values with the most frequent value
df['bmi'].fillna(df['bmi'].mode()[0], inplace=True)
print(df['ever_married'].value_counts())
print(df['work_type'].value_counts())
print(df['gender'].value_counts())
print(df['Residence_type'].value_counts())
print(df['smoking_status'].value_counts())
ever_married Yes 3353 No 1757 Name: count, dtype: int64 work_type Private 2925 Self-employed 819 children 687 Govt_job 657 Never_worked 22 Name: count, dtype: int64 gender Female 2994 Male 2115 Other 1 Name: count, dtype: int64 Residence_type Urban 2596 Rural 2514 Name: count, dtype: int64 smoking_status never smoked 1892 Unknown 1544 formerly smoked 885 smokes 789 Name: count, dtype: int64
df['ever_married'].replace({'Yes':1, 'No':0}, inplace=True)
df['gender'].replace({'Male':1, 'Female':0,'Other':2}, inplace=True)
df['Residence_type'].replace({'Urban':1, 'Rural':0}, inplace=True)
df['smoking_status'].replace({'formerly smoked':0, 'never smoked':1, 'smokes':2, 'Unknown':3}, inplace=True)
df['work_type'].replace({'Private':0, 'Self-employed':1, 'children':2, 'Govt_job':3, 'Never_worked':4}, inplace=True)
df.corr()['stroke'][:-1].sort_values().plot(kind='bar')
<Axes: >
plt.figure(figsize=(10,10))
sns.heatmap(df.corr(), annot=True)
<Axes: >
# replace age with number wrt to age group
# 0 = 0-12 , 1 = 13-19 , 2 = 20-30 , 3 = 31-60 , 4 = 61-100
df['age'] = pd.cut(x=df['age'], bins=[0, 12, 19, 30, 60, 100], labels=[0, 1, 2, 3,4])
df.head()
| gender | age | hypertension | heart_disease | ever_married | work_type | Residence_type | avg_glucose_level | bmi | smoking_status | stroke | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 1 | 4 | 0 | 1 | 1 | 0 | 1 | 228.69 | 36.6 | 0 | 1 |
| 1 | 0 | 4 | 0 | 0 | 1 | 1 | 0 | 202.21 | 28.7 | 1 | 1 |
| 2 | 1 | 4 | 0 | 1 | 1 | 0 | 0 | 105.92 | 32.5 | 1 | 1 |
| 3 | 0 | 3 | 0 | 0 | 1 | 0 | 1 | 171.23 | 34.4 | 2 | 1 |
| 4 | 0 | 4 | 1 | 0 | 1 | 1 | 0 | 174.12 | 24.0 | 1 | 1 |
sns.countplot(x = 'gender', data = df)
<Axes: xlabel='gender', ylabel='count'>
fig, ax = plt.subplots(4,4,figsize=(20, 20))
sns.countplot(x = 'gender', data = df,hue = 'stroke', ax=ax[0,0])
sns.countplot(x = 'age', data = df,hue = 'hypertension', ax=ax[0,1])
sns.countplot(x = 'age', data = df,hue = 'heart_disease', ax=ax[0,2])
sns.countplot(x = 'age', data = df,hue = 'stroke', ax=ax[0,3])
sns.countplot(x = 'hypertension', data = df,hue = 'stroke', ax=ax[1,0])
sns.countplot(x = 'heart_disease', data = df,hue = 'stroke', ax=ax[1,1])
sns.countplot(x = 'ever_married', data = df,hue = 'stroke', ax=ax[1,2])
sns.countplot(x = 'age', data = df,hue = 'ever_married', ax=ax[1,3])
sns.countplot(x = 'work_type', data = df,hue = 'stroke', ax=ax[2,0])
sns.countplot(x = 'Residence_type', data = df,hue = 'stroke', ax=ax[2,1])
sns.countplot(x = 'smoking_status', data = df,hue = 'stroke', ax=ax[2,2])
sns.lineplot(x = 'bmi', y = 'avg_glucose_level', data = df,hue = 'stroke', ax=ax[2,3])
sns.countplot(x = 'age', data = df,hue = 'smoking_status', ax=ax[3,0])
sns.countplot( x = 'work_type', data = df,hue = 'Residence_type', ax=ax[3,1])
sns.countplot(x = 'work_type', data = df,hue = 'smoking_status', ax=ax[3,2])
sns.countplot(x = 'Residence_type', data = df,hue = 'smoking_status', ax=ax[3,3])
<Axes: xlabel='Residence_type', ylabel='count'>
X_train, X_test, y_train, y_test = train_test_split(df.drop('stroke', axis=1), df['stroke'], test_size=0.2, random_state=42)
lr = LogisticRegression()
lr
LogisticRegression()In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
LogisticRegression()
#training the model
lr.fit(X_train, y_train)
lr.score(X_test, y_test)
C:\Users\DELL\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.11_qbz5n2kfra8p0\LocalCache\local-packages\Python311\site-packages\sklearn\linear_model\_logistic.py:458: ConvergenceWarning: lbfgs failed to converge (status=1):
STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.
Increase the number of iterations (max_iter) or scale the data as shown in:
https://scikit-learn.org/stable/modules/preprocessing.html
Please also refer to the documentation for alternative solver options:
https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression
n_iter_i = _check_optimize_result(
0.9393346379647749
#testing the model
lr_pred = lr.predict(X_test)
accuracy_score(y_test, lr_pred)
0.9393346379647749
from sklearn.svm import SVC
svm = SVC()
svm
SVC()In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
SVC()
#training the model
svm.fit(X_train, y_train)
svm.score(X_test, y_test)
0.9393346379647749
#testing the model
sv_pred = svm.predict(X_test)
accuracy_score(y_test, sv_pred)
0.9393346379647749
from sklearn.tree import DecisionTreeClassifier
dt = DecisionTreeClassifier()
dt
DecisionTreeClassifier()In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
DecisionTreeClassifier()
#training the model
dt.fit(X_train, y_train)
dt.score(X_test, y_test)
0.9099804305283757
#testing the model
dt_pred = dt.predict(X_test)
accuracy_score(y_test, dt_pred)
0.9099804305283757
knn = KNeighborsClassifier()
knn
KNeighborsClassifier()In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
KNeighborsClassifier()
#training the model
knn.fit(X_train, y_train)
knn.score(X_test, y_test)
0.9373776908023483
#testing the model
knn_pred = knn.predict(X_test)
accuracy_score(y_test, knn_pred)
0.9373776908023483
sns.heatmap(metrics.confusion_matrix(y_test, lr_pred), annot=True, fmt='d')
plt.title('Accuracy Score: {}'.format(accuracy_score(y_test, lr_pred)))
plt.ylabel('Predicted')
plt.xlabel('Actual')
plt.show()
print('Logistic Regression Model Accuracy Score:',accuracy_score(y_test, lr_pred))
print('Logistic Regression Model F1 score: ',metrics.f1_score(y_test, lr_pred))
print('Logistic Regression Model Mean Absolute Error: ',metrics.mean_absolute_error(y_test, lr_pred))
print('Logistic Regression Model Mean Squared Error: ',metrics.mean_squared_error(y_test, lr_pred))
print('Logistic Regression Model log loss: ',log_loss(y_test, lr_pred))
Logistic Regression Model Accuracy Score: 0.9393346379647749 Logistic Regression Model F1 score: 0.0 Logistic Regression Model Mean Absolute Error: 0.060665362035225046 Logistic Regression Model Mean Squared Error: 0.060665362035225046 Logistic Regression Model log loss: 2.1866012819229583
sns.heatmap(metrics.confusion_matrix(y_test, sv_pred), annot=True, fmt='d')
plt.title('Accuracy Score: {}'.format(accuracy_score(y_test, sv_pred)))
plt.ylabel('Predicted')
plt.xlabel('Actual')
plt.show()
print('SVM Model Accuracy Score:',accuracy_score(y_test, sv_pred))
print('SVM Model F1 score: ',metrics.f1_score(y_test, sv_pred))
print('SVM Model Mean Absolute Error: ',metrics.mean_absolute_error(y_test, sv_pred))
print('SVM Model Mean Squared Error: ',metrics.mean_squared_error(y_test, sv_pred))
print('SVM Model log loss: ',log_loss(y_test, sv_pred))
SVM Model Accuracy Score: 0.9393346379647749 SVM Model F1 score: 0.0 SVM Model Mean Absolute Error: 0.060665362035225046 SVM Model Mean Squared Error: 0.060665362035225046 SVM Model log loss: 2.1866012819229583
sns.heatmap(metrics.confusion_matrix(y_test, dt_pred), annot=True, fmt='d')
plt.title('Accuracy Score: {}'.format(accuracy_score(y_test, dt_pred)))
plt.ylabel('Predicted')
plt.xlabel('Actual')
plt.show()
print('Decision Tree Model Accuracy Score:',accuracy_score(y_test, dt_pred))
print('Decision Tree Model F1 score: ',metrics.f1_score(y_test, dt_pred))
print('Decision Tree Model Mean Absolute Error: ',metrics.mean_absolute_error(y_test, dt_pred))
print('Decision Tree Model Mean Squared Error: ',metrics.mean_squared_error(y_test, dt_pred))
print('Decision Tree Model log loss: ',log_loss(y_test, dt_pred))
Decision Tree Model Accuracy Score: 0.9099804305283757 Decision Tree Model F1 score: 0.2459016393442623 Decision Tree Model Mean Absolute Error: 0.09001956947162426 Decision Tree Model Mean Squared Error: 0.09001956947162426 Decision Tree Model log loss: 3.2446341602727773
sns.heatmap(metrics.confusion_matrix(y_test, knn_pred), annot=True, fmt='d')
plt.title('Accuracy Score: {}'.format(accuracy_score(y_test, knn_pred)))
plt.ylabel('Predicted')
plt.xlabel('Actual')
plt.show()
print('KNN Model Accuracy Score:',accuracy_score(y_test, knn_pred))
print('KNN Model F1 score: ',metrics.f1_score(y_test, knn_pred))
print('KNN Model Mean Absolute Error: ',metrics.mean_absolute_error(y_test, knn_pred))
print('KNN Model Mean Squared Error: ',metrics.mean_squared_error(y_test, knn_pred))
print('KNN Model log loss: ',log_loss(y_test, knn_pred))
KNN Model Accuracy Score: 0.9373776908023483 KNN Model F1 score: 0.0 KNN Model Mean Absolute Error: 0.06262230919765166 KNN Model Mean Squared Error: 0.06262230919765166 KNN Model log loss: 2.2571368071462796
models = ['Logistic Regression', 'SVM', 'Decision Tree', 'KNN']
accuracy = [accuracy_score(y_test, lr_pred), accuracy_score(y_test, sv_pred), accuracy_score(y_test, dt_pred), accuracy_score(y_test, knn_pred)]
plt.figure(figsize=(10,5))
plt.bar(models, accuracy, color = 'Maroon', width = 0.4)
plt.xlabel('Models')
plt.ylabel('Accuracy')
plt.title('Model Accuracy')
plt.show()
The model accuracies of Logistic Regression, SVM and KNN are quite similar i.e. 93.8 %. The accuracy of Decision Tree Classifier is 91.8 %. So, we can use any of these models to predict the heart stroke.
According to the graphs age v/s hypertension, heart disease showing chances of stroke, the number of person having a stroke shows dependece upon heart disease and hypertension. But when we plot the graph of heart disease and hypertension against the stroke, the persons with lower chances of hypertension and heart disease has increased chances of stroke. This is a peculiar thing and needs to be investigated further. In addition to that non somkers have higher chances of stroke than smokers. This is also a peculiar thing and needs to be investigated further. However person having BMI between20 to 50 have higher chances of stroke.
Last but not least other features such as martial status, residence type as well as work type are showing effect on the chances of stroke.