Loan Status Prediction

A Support Vector Machine model to predict whether loan can be sanctioned or not

Prerequisites

• dataset
• Numpy
• Pandas
• Seaborn
• Sklearn

Process

1. Loading the modules and reading the csv file

2. Checking and clearing the null values in the dataframe loan_data.dropna()

3. Label encoding using dictionary loan_data.replace({'Loan_Status':{'N':0,'Y':1}},inplace=True)

   • 1 for yes
   • 0 for no

4. Checking column values of 'Dependents'

5. Here we got 3+ values as count of 41 so we can't have that many, so

6. Replacing value of 3+ to 4 by loan_data=loan_data.replace(to_replace='3+' , value=4)

7. Visualising the data using seaborn

   • Education vs Loan_status
   • Marriage status vs Loan_status
   • Self employed vs Property area

8. Convert categorical text into numerical like

                                'Self_Employed':{'No':0,'Yes':1},'Property_Area':{'Rural':0,'Semiurban':1,'Urban':2},
                                'Education':{'Graduate':1,'Not Graduate':0}})

9. Separating the data and the label
10. Splitting data into train and test`
11. Train SVM classifier model by classifier = svm.SVC(kernel='linear')
12. Fit the data by classifier.fit(x_train,y_train)
13. Model evaluation by comparing accuracy score of training and test data
14. Prediction part

input_data = (1,0,2,0,0,3000,250.0,200.0,180.0,0.0,2)

input_data_as_numpy_array = np.asarray(input_data)

#model predicting only one instance
input_data_reshaped= input_data_as_numpy_array.reshape(1,-1)  
prediction = classifier.predict(input_data_reshaped)

print(prediction)
if(prediction[0] == 0):
  print('Not Sanctioned')
else:
  print('Loan Sanctioned')