Name – Preet Desai

Internship Program - Data Science with Machine Learning and Python

Batch - Jan 2022 - Mar 2022
Certificate Code - TCRIB2R123
Date of submission – 18/04/2022

Technical Coding Research Innovation, Navi Mumbai,

Maharashtra, India-410206

HR EMPLOYEE ATTRITION ANALYSIS

A Case-Study Submitted for the requirement of
Technical Coding Research Innovation
For the Internship Project work done during
DATA SCIENCE WITH MACHINE LEARNING AND PYTHON
INTERNSHIP PROGRAM
by
Ayush Sanghavi (TCRIB2R180)
Preet Deshi (TCRIB2R123)
Somil Doshi(TCRIB2R122)

Rutuja Doiphode
CO-FOUNDER & CEO
TCR innovation.
Name – Preet Desai
Internship Program - Data Science with Machine Learning and Python
Batch - Jan 2022 - Mar 2022
Certificate Code - TCRIB2R123
Date of submission – 18/04/2022

Abstract - This paper gives your insight in a classification algorithm to a dataset.

Index Terms –
• Aim.
• Introduction to Dataset.
• Exploratory data analysis on dataset.
• Training & Prediction of data
• Conclusion

I. AIM
The aim is to predict whether an employee wants to continue to do work or not. For this, I am
using a dataset named “HR Employee Attrition”.

II. INTRODUCTION TO DATASET

The “HR EMPLOYEE ATTRITION DATASET” consists of the details of an employee like
gender, age, business travel, department, education, relationship satisfaction, and many others.
Basically, the dataset consists of exactly 2940 employees' data, and employee has 34 features.
The dataset consists of both numerical and categorical data. Below is an image of the dataset.

III. EXPLORATORY DATA ANALYSIS

Basically, Exploratory Data Analysis, or EDA, is an important step in any Data

Analysis or Data Science project. EDA is the process of investigating the dataset
to discover patterns. EDA involves generating summary statistics for numerical
data in the dataset and creating various graphical representations to understand the
data better. We will understand EDA with the help of an example dataset. Using
EDA, we try to find out the frequency distribution of all categorical values w.r.t.
Attrition. In the "HR EMPLOYEE ATTRITION DATASET", there are no NAN
and outliers’ values. The libraries used were pandas, numpy, matplotlib, sklearn.
Name – Preet Desai
Internship Program - Data Science with Machine Learning and Python
Batch - Jan 2022 - Mar 2022
Certificate Code - TCRIB2R123
Date of submission – 18/04/2022

IV. TRAINING & PREDICTION OF DATA

After analysing the complete dataset, I apply the Random Forest Classification
algorithm to train the machine learning model because this algorithm can work
effectively with a large number of features. Firstly, I split the dataset as 83.88 %
for training and 16.12% for testing. Below is the implementation of training and
predicting the machine learning model to achieve the aim:
Name – Preet Desai
Internship Program - Data Science with Machine Learning and Python
Batch - Jan 2022 - Mar 2022
Certificate Code - TCRIB2R141
Date of submission – 18/04/2022
Name – Preet Desai
Internship Program - Data Science with Machine Learning and Python
Batch - Jan 2022 - Mar 2022
Certificate Code - TCRIB2R141
Date of submission – 18/04/2022

V. CONCLUSION

After applying the Random Forest Classification algorithm, the machine learning
model will be able to predict employee attrition with an accuracy of 95.46%. This
is not the only method to train the model for the prediction of employee attrition.
Using various other algorithms, it can be possible to predict but I found that this
Name – Preet Desai
Internship Program - Data Science with Machine Learning and Python
Batch - Jan 2022 - Mar 2022
Certificate Code - TCRIB2R141
Date of submission – 18/04/2022

algorithm works better than all other classification algorithms in the "HR
EMPLOYEE ATTRITION DATASET".

Final output of the accuracy of model:

VI. REFERENCES

[1] Book: Machine Learning for Absolute Beginners by Oliver Theobald.

[2] Book: Python for Data Analysis by Wes McKinnie

WWW.TCRINNOVATION.CO.IN INTERNSHIP PROGRAM NAME