Classification Algorithms - Attrition Prediction

Use Case: This dataset gives information about the employees in the organization and target variable (Y) is Attrition, which tells whether an employee was replaced or not. We will look at the variables and different factors that affects the employee performance, including factors like work from home situation, job role, monthly income etc.

Importing the basic required libraries

#Basics Required Packages:
import matplotlib.pyplot as plt
import  pandas as pd
import numpy as np

#Visualization
import matplotlib.pyplot as plot
import seaborn as sns

#Sklearn ML General
from sklearn import preprocessing
from sklearn.model_selection import train_test_split
# from sklearn.metrics import plot_confusion_matrix
from sklearn.metrics import ConfusionMatrixDisplay


from sklearn import metrics

#Sklearn ML Algorithms
from sklearn.naive_bayes import GaussianNB
from sklearn.neighbors import KNeighborsRegressor
from sklearn.tree import DecisionTreeClassifier
from sklearn.ensemble import RandomForestClassifier
from sklearn.svm import SVC
# from sklearn import SVM
from sklearn.svm import SVC, SVR

Reading and exploring the datasets using pandas :

#Data Exploration
employee_data = pd.read_csv("/Users/Library/CloudStorage/Course Files - Classification/3 - Classification Algorithms/HREmployeeData.csv")
print(employee_data)

#Data Exploration
print(employee_data.describe())

Data Visualization of the original dataset and ensure whether all the ‘NA’ values are dropped:

#Plot a histogram of the target variable
employee_data.Attrition.value_counts().nlargest(40).plot(kind = 'bar',figsize = (10,5))
plt.title('Attrition of the employees')
plt.ylabel('Occurence of each in the dataset')
plt.xlabel('Attrition')

Predictive Modelling : X will contain our independent feature variables and y will be our target and outcome variable. Then the split the test and train the data with 50% ratio.

#Predictive Modelling :
x_inputs = employee_data.drop(['Attrition'], axis = 1)
y_target = employee_data['Attrition']


#Spliting test and train dataset:
x_train,x_test, y_train, y_test = train_test_split(x_inputs, y_target,
                                                   test_size=0.5,random_state=0)

Initialize each of the classifiers from the original dataset:

#Define the classifiers :
classifier = {
    "classifier_NB": GaussianNB(),
    "classifier_SVM": SVC(gamma=2, C = 1),
    "classifier_KNN": KNeighborsRegressor(n_neighbors=3),
    "classifier_DT": DecisionTreeClassifier(random_state=0),
    "classifier_RT": RandomForestClassifier(max_depth=2, random_state=0)
}
results = {}
print("Number of mislabeled points out of total %d points:" % x_test.shape[0])

Now train the classifiers on the training set and test classifiers on unseen test data:

for clf_name, clf_model in classifier.items():
    clf_model.fit(x_train,y_train)
    clf_model.predict(x_test)
    print(clf_name + ": %d" % (y_test != clf_model.predict(x_test)).sum())
    results [clf_name] = clf_model.predict(x_test)