Visa Success Predection

Data set of a study abroad education consultancy is used to train a model in order to predict the visa success rate.

Importing libraries

import pandas as pd
import seaborn as sns
from matplotlib import pyplot
from sklearn.externals import joblib

Read processed data file

df=pd.read_csv('data_final.csv')
df

Finding Mean, Median, Standard Deviation

df.describe().transpose()

Checking for null values

df.apply(lambda x: sum(x.isnull()),axis=0)

Finding Correlation

cor=df.corr()
cor

Ploting Heat Map

pyplot.figure(figsize=(30,30))
sns.heatmap(cor,annot=True,cmap="RdYlGn")

Checking Feature Importance

X = df.iloc[:,0:16]  #independent columns
y = df.iloc[:,-1]    #target column
from sklearn.ensemble import ExtraTreesClassifier
model = ExtraTreesClassifier()
model.fit(X,y)
print(model.feature_importances_) #use inbuilt class feature_importances of tree based classifiers
#plot graph of feature importances for better visualization
feat_importances = pd.Series(model.feature_importances_, index=X.columns)
pyplot.figure(figsize=(10,10))
feat_importances.nlargest(16).plot(kind='barh')
pyplot.show()

Feature Score

X = df.iloc[:,0:16]  #independent columns
y = df.iloc[:,-1]    #target column  best features
from sklearn.feature_selection import SelectKBest
from sklearn.feature_selection import chi2
bestfeatures = SelectKBest(score_func=chi2, k=16)
fit = bestfeatures.fit(X,y)
dfscores = pd.DataFrame(fit.scores_)
dfcolumns = pd.DataFrame(X.columns)
#concat two dataframes for better visualization 
featureScores = pd.concat([dfcolumns,dfscores],axis=1)
featureScores.columns = ['Factors','Score']  #naming the dataframe columns
print(featureScores.nlargest(16,'Score'))  #print 54 best features

Splitting data into train and test

x = df.iloc[:,0:16] 
y = df.iloc[:,-1] 
from sklearn.model_selection import train_test_split
#spliting test and train set.
x_train, x_test, y_train, y_test = train_test_split(x, y, test_size=0.20, random_state=42)
print(x_train)
print(y_train)

Applying Model

Logistic Regression

from sklearn.linear_model import LogisticRegression
#Make an instance of the Model
logisticRegr = LogisticRegression()

logisticRegr.fit(x_train, y_train)
#Predict labels for new data
y_predict = logisticRegr.predict(x_test)

from sklearn.metrics import accuracy_score
score = accuracy_score(y_test, y_predict, normalize = True)
print("Accuracy : ",score)

Creating Confusion Matrix

from sklearn import metrics
cm = metrics.confusion_matrix(y_test, y_predict)
pyplot.figure(figsize=(9,9))
sns.heatmap(cm, annot=True, fmt=".3f", linewidths=.5, square = True, cmap = 'Blues_r');
pyplot.ylabel('Actual label');
pyplot.xlabel('Predicted label');
all_sample_title = 'Accuracy Score: {0}'.format(score)
pyplot.title(all_sample_title, size = 15)

ROC Curve

from sklearn.metrics import roc_curve
from sklearn.metrics import roc_auc_score
auc = roc_auc_score(y_test, y_predict)
print('Accuracy: %.3f' % auc)
# calculate roc curve
fpr, tpr, thresholds = roc_curve(y_test, y_predict)
pyplot.figure(figsize=(5,5))
# plot no skill
pyplot.plot([0, 1], [0, 1], linestyle='--')
# plot the roc curve for the model
pyplot.plot(fpr, tpr, marker='.')
#show the plot
pyplot.show()

Naive-Bayes

from sklearn.naive_bayes import GaussianNB
#create an object of the type GaussianNB
gnb = GaussianNB()
#train the algorithm on training data and predict using the testing data
gnb.fit(x_train, y_train)
y_predict_nb = gnb.predict(x_test)
#Import scikit-learn metrics module for accuracy calculation
from sklearn import metrics
# Model Accuracy, how often is the classifier correct?
print("Accuracy:",metrics.accuracy_score(y_test, y_predict_nb))