BASAVARAJESWARI GROUP OF INSTITUTIONS

BALLARI INSTITUTE OF TECHNOLOGY & MANAGEMENT

NACC Accredited Institution*
(Recognized by Govt. of Karnataka, approved by AICTE, New Delhi & Affiliated to
Visvesvaraya Technological University, Belgavi)
"Jna Gnana Gangotri" Campus, No.873/2, Ballari-Hospet Road, Allipur,
Ballar1-583 104 (Karnataka) (India)
Ph: 08392 – 237100 / 237190, Fax: 08392 – 237197

DEPARTMENT OF COMPUTER SCIENCE & ENGINEERING

A Project Report On

“PREDICTING EMPLOYEE PROMOTIONS”

A dissertation submitted to the Department of Computer Science and Engineering of
Visvesvaraya Technological University in partial fulfilment for the award of the Degree
of Bachelor of Engineering.

Project Associates:
MONISHA V 3BR20CS106
NAVYA GADIGI 3BR20CS113
RINI SHARON 3BR20CS133
MALLIKA S 3BR20CS139

Under the guidance of

Mr. V. CHANDRASHEKAR
Assistant Professor

Dept of CSE,
BITM, Ballari.

Visvesvaraya Technological University

Belagavi, Karnataka
2023-2024
 BASAVARAJESWARI GROUP OF INSTITUTIONS

BALLARI INSTITUTE OF TECHNOLOGY & MANAGEMENT

NACC Accredited Institution*
(Recognized by Govt. of Karnataka, approved by AICTE, New Delhi & Affiliated to
Visvesvaraya Technological University, Belgavi)
"Jnana Gangotri" Campus, No.873/2, Ballari-Hospet Road, Allipur,
Ballar1-583 104 (Karnataka) (India)
Ph: 08392 – 237100 / 237190, Fax: 08392 – 237197

DEPARTMENT OF COMPUTER SCIENCE & ENGINEERING

CERTIFICATE

This is to certify that the PROJECT WORK entitled “PREDICTING EMPLOYEE

PROMOTIONS” is a bonafide work carried out by MONISHA V, NAVYA GADIGI,
RINI SHARON and MALLIKA S bearing USN 3BR20CS106, 3BR20CS113,
3BR20CS133 and 3BR20CS139 in partial fulfilment of the award of Bachelor’s
Degree in Computer Science & Engineering of the VISVESVARAYA
TECHNOLOGICAL UNIVERSITY, Belagavi during the academic year 2023-2024.It is
certified that all corrections and suggestions indicated for internal assessment have been
incorporated in the report deposited in the library. The project has been approved as it
satisfies the academic requirements regarding project work prescribed for a Bachelor of
Engineering Degree.

Signature of Guide Signature of HOD Signature of Principal

Mr. V Chandrashekar Dr. R. N. Kulkarni Dr. Yadavalli Basavaraj

Name of the Examiner(s) Signature with Date

1)……………………………. ……………………..

2)…………………………… ………………………
 ABSTRACT

Employee promotions are an important aspect of organizational growth and employee career
progression, and understanding the factors that contribute to promotions can help
organizations make informed decisions regarding employee development and performance
management. The project utilizes a dataset containing information about employees,
including their department, region, education, gender, recruitment channel, previous year
rating, length of service, and other relevant attributes. The focus of the analysis is on three
key factors: the number of trainings attended, KPI achievement, and training score. The
analysis begins with exploratory data analysis to gain insights into the distribution and
relationships between the variables. Various statistical techniques, such as data visualization
and summary statistics, are employed to understand the patterns and trends within the dataset.
Next, predictive modelling techniques, specifically machine learning algorithms, are utilized
to build a model that can accurately predict employee promotions. Several algorithms, such
as decision trees, random forests, and logistic regression, are applied to train the model on the
available data. The model is then evaluated using appropriate evaluation metrics to assess its
performance and predictive accuracy. The results of the analysis provide valuable insights
into the factors that influence employee promotions.

Page | I
 ACKNOWLEDGEMENT

Salutations to our beloved and highly esteemed institute, “BALLARI INSTITUTE OF

TECHNOLOGY & MANAGEMENT” for having well qualified staff and labs furnished
with necessary equipment.

We express our sincere thanks to our guide Mr. V. Chandrashekar, Asst. Professor of CSE
dept. for giving us constant encouragement, support and valuable guidance throughout the
course of the project without whose stable guidance this project would not have been
achieved and we would thank our project coordinators Mr. P. Phaniram Prasad and Dr. C K
Srinivas.

We express whole hearted gratitude to Dr. R.N. KULKARNI who is our respectable HOD of
Computer Science Dept. We wish to acknowledge his help who made our task easy by
providing with his valuable help and encouragement.

And also our due thanks to Dr. Yadavalli Basavaraj, the Principal, as we consider ourselves
very lucky to have such excellent computing facilities and their inspiration throughout our
professional course.

We also thank the non-teaching staff of CS Dept. who guided at all the time of difficulties.

Finally, we thank all those who are involved directly and indirectly in completion of our
project.

-Monisha V
-Navya Gadigi
-Rini Sharon
-Mallika S

Page | II
 TABLE OF CONTENTS
Chapter No Chapter Name Page No
Abstract I
Acknowledgement II
Table of Contents III
List of Figures V
1 Chapter 1
Introduction 1
1.1 Vision and Mission 2
1.2 Objectives 3
1.3 Literature Survey 5
1.4 Problem Statement 6
1.5 Scope of the project 6

2 Chapter 2
System Requirements and Specifications
2.1 Hardware Requirements 7
2.2 Software Requirements 7
2.3 Functional Requirements 8
2.4 Non-Functional Requirements 8

3 Chapter 3
Description of Modules 9

4 Chapter 4
Design 11
4.1 Use Case diagram 11
4.2Flow Chart 12
4.3 Activity diagram 14
4.4 Sequence diagram 15
4.5 ER Diagram 16

5 Chapter 5
Implementation 17
5.1 Data Description 17
5.2 Exploratory data analysis 18
5.3 Pre-processing 21
5.4 K-best feature selection method 22
5.5 K-Means clustering 23

6 Chapter 6
System Testing 24
6.1 Unit Testing 24
6.2 Integration Testing 24
6.3 User Acceptance Testing 24
6.4 White box Testing 25
6.5 Black box Testing 25

7 Chapter 7
Page | III
Results 26
Conclusion 31
References 32
CO-PO Matrix 33
Paper publication 34

Page | IV
 LIST OF FIGURES

Sl. No Figure Name Page No

4.1 Use Case diagram 11

4.2.1 Flow chart for workflow of system 12
4.2.2 Flow chart of prediction process 13
4.3 Activity diagram of predicting employee promotion 14
4.4 Sequence diagram of predicting employee promotion 15
4.5 ER Diagram 16
7.1 Home Page 26
7.2 Registration Form 27
7.3 Login Form 27
7.4 Upload Dataset 28
7.5 Data Pre-Processing 28
7.6 Model Training 29
7.7 Enter Input 29
7.8 Result predicting that the employee is promoted 30
7.9 Result predicting that the employee is not promoted 30

Page | V
 PREDICTING EMPLOYEE PROMOTIONS

CHAPTER 1
INTRODUCTION
We aspire to establish a fair and merit-based promotion system that recognizes and rewards talent,
dedication, and professional growth. Our approach aims to enhance transparency in career advancement
processes, enabling organizations to identify and nurture high-potential employees. By leveraging data-
driven insights, we intend to empower companies to make informed, unbiased promotion decisions,
fostering a culture of excellence, innovation, and continuous learning that aligns with organizational goals
and employee aspirations. Our focus is on promoting a culture of meritocracy, where employee
achievements and potential are objectively evaluated. We strive to eliminate biases and subjectivity in
promotion processes, ensuring that every employee has a fair chance at career advancement based on their
contributions and capabilities, ultimately leading to higher employee satisfaction and organizational success.
By examining the relationship between the number of trainings, KPI achievement, and training score,
organizations can identify the key drivers of promotion and develop strategies to enhance employee
performance and career development. This project aims to contribute to the understanding of employee
promotions and provide organizations with valuable insights for effective talent management. The findings
can help organizations make data-driven decisions regarding employee promotions, leading to improved
employee satisfaction, engagement, and organizational performance

BE, 8th SEM, CSE, BITM, Ballari Page | 1

 PREDICTING EMPLOYEE PROMOTIONS

1.1 VISION AND MISSION

VISION

To create a transformative model that accurately predicts employee promotions by integrating key
performance indicators (KPIs), training effectiveness, and training scores.

MISSION
• To revolutionize the way organizations approach employee promotions by developing an advanced
predictive model that considers training participation, KPI achievements, and training scores.
• To provide a robust tool that assists HR departments in making equitable and data-backed promotion
decisions.

BE, 8th SEM, CSE, BITM, Ballari Page | 2

 PREDICTING EMPLOYEE PROMOTIONS

1.2 OBJECTIVES

1. To develop a comprehensive predictive model.

2. To enhance fairness in promotion decisions.
3. To boost employee motivation and performance.
4. To improve talent management and retention.
5. To facilitate organizational growth and success.

1. To develop a comprehensive predictive model:

Developing a comprehensive predictive model involves utilizing statistical techniques and machine learning
algorithms to analyze historical data and predict future outcomes. This process includes data collection,
preprocessing, model selection, training, validation, and continuous refinement. A robust predictive model
can identify trends, forecast events, and inform strategic decisions, enhancing the ability of organizations to
anticipate market changes, optimize operations, and allocate resources efficiently. By incorporating a wide
range of variables and ensuring model accuracy, organizations can achieve a deeper understanding of
complex patterns and improve their predictive capabilities across various domains.

2. To enhance fairness in promotion decisions:

Enhancing fairness in promotion decisions requires implementing transparent, objective, and equitable
criteria for evaluating employee performance and potential. This can be achieved by standardizing
assessment processes, utilizing performance metrics, and incorporating diverse perspectives to mitigate
biases. Organizations should foster a culture of inclusivity and ensure that all employees have equal access
to development opportunities and clear communication about promotion paths. By focusing on merit-based
evaluations and continuous feedback, companies can build trust, increase morale, and support a more diverse
and equitable workplace, ultimately leading to more effective leadership and a stronger organizational
culture.

3. To boost employee motivation and performance:

Boosting employee motivation and performance involves creating an environment where employees feel
valued, engaged, and supported. This can be achieved through recognition programs, professional

BE, 8th SEM, CSE, BITM, Ballari Page | 3

 PREDICTING EMPLOYEE PROMOTIONS

development opportunities, competitive compensation, and fostering a positive workplace culture. Effective
leadership plays a crucial role in inspiring and guiding employees, while clear communication and setting
achievable goals help align individual efforts with organizational objectives. By addressing intrinsic and
extrinsic motivators, such as personal growth, job satisfaction, and financial rewards, organizations can
enhance productivity, reduce turnover, and cultivate a workforce that is committed to achieving high
performance.

4. To improve talent management and retention:

Improving talent management and retention involves strategic initiatives to attract, develop, and retain top
talent within the organization. This includes creating a supportive onboarding process, offering continuous
learning and career advancement opportunities, and maintaining a positive work-life balance. Regularly
assessing employee satisfaction and addressing concerns promptly can prevent attrition. Implementing
mentorship programs and fostering a culture of recognition and appreciation also contribute to higher
retention rates. By understanding and meeting the evolving needs of employees, organizations can build a
loyal, skilled workforce, reducing the costs associated with turnover and enhancing overall organizational
stability and growth.

5. To facilitate organizational growth and success:

Facilitating organizational growth and success requires a comprehensive strategy that aligns with the
company's vision and goals. This involves effective leadership, innovation, market expansion, and
continuous improvement of processes. Investment in technology and talent development, along with
fostering a culture of agility and adaptability, can drive growth. Strategic planning and execution, customer
focus, and a commitment to sustainability and social responsibility also play vital roles. By leveraging
resources efficiently and responding proactively to market dynamics, organizations can achieve competitive
advantages, increase profitability, and ensure long-term success and resilience in an ever-changing business
environment.

BE, 8th SEM, CSE, BITM, Ballari Page | 4

 PREDICTING EMPLOYEE PROMOTIONS

1.3LITERATURE SURVEY

In the chosen area of employee promotion prediction, several studies have been conducted to explore the
factors and develop predictive models. These studies have contributed valuable insights and methodologies
to enhance the understanding of employee performance and promotion dynamics. This section provides a
background on some key previous work in this area.

In article [1], the author made a notable study conducted by Smith et al. (2017) aimed to predict employee
promotions using a combination of traditional machine learning algorithms and advanced feature
engineering techniques. They analyzed various factors such as performance ratings, education, experience,
and job tenure to develop a predictive model. The study achieved promising results with an accuracy of 85%
in predicting employee promotions.

In paper [2], the author presented a relevant work carried out by Johnson et al. (2018) focused on exploring
the influence of training programs on employee promotion probabilities. They collected data on training
participation, performance metrics, and promotion outcomes. By applying logistic regression and decision
tree algorithms, they identified the significant impact of training performance and completion on promotion
likelihood.

In paper [3] Lee and Kim (2019) investigated the role of Key Performance Indicators (KPIs) in predicting
employee promotions. They conducted a longitudinal study, tracking KPI performance and promotion
outcomes over time. By employing statistical analysis techniques, they identified specific KPIs that
significantly contributed to promotion probabilities in different departments.

In article [4] a research by Chen and Liu (2020) delved into the application of unsupervised learning
techniques in the context of employee promotion. They used K-Means clustering to group employees based
on various attributes such as education, job tenure, and performance ratings. By analyzing the characteristics
of different clusters, they provided valuable insights into employee segmentation and promotion patterns.

BE, 8th SEM, CSE, BITM, Ballari Page | 5

 PREDICTING EMPLOYEE PROMOTIONS

In paper [5] a study by Wang et al. (2021) focused on feature selection methods to enhance the prediction of
employee promotions. They employed the K-Best feature selection algorithm to identify the most relevant
features from a large dataset. By reducing the dimensionality of the data, they achieved improved prediction
accuracy and model performance.

In article [6] several studies have explored the application of supervised machine learning algorithms in
predicting employee promotions. For instance, Li et al. (2018) employed XGBoost, a popular gradient
boosting algorithm, to predict promotion outcomes based on various employee attributes and performance
metrics. Their study demonstrated the effectiveness of XGBoost in accurately predicting promotions.

In paper [7] a different approach by Zhang and Guo (2019) utilized support vector machines (SVM) to
predict employee promotions. They developed a SVM-based model using features such as education,
performance ratings, and job tenure. The study highlighted the potential of SVM in accurately classifying
employees into promotion and non-promotion categories.

1.4 PROBLEM STATEMENT

To design and develop a software application that analyzes and predicts the employee promotions in a
company based on several factors which includes the number of trainings attended, KPI (Key Performance
Indicator) achievements, and training score using machine learning algorithms.

1.5 SCOPE OF THE PROJECT

The proposed project is stand-alone application and is suitable only for the employees of any organization
that analyzes the Key performance indicator.

BE, 8th SEM, CSE, BITM, Ballari Page | 6

 PREDICTING EMPLOYEE PROMOTIONS

CHAPTER 2
SYSTEM REQUIREMENTS AND SPECIFICATION

Project requirements are conditions or tasks that must be completed to ensure the success or completion of
the project. They provide a clear picture of the work that needs to be done. They're meant to align the
project's resources with the objectives of the organization. The following section provides enlightenment
into the various requirements of a project and also deals with the requirements pertaining to this work.

2.1 Hardware Requirements:

The most common set of requirements defined by any operating system or software application is the
physical computer resources, also known as hardware.

• RAM : 8 GB (min)

• Hard disc or SSD : 128 GB

• Processor : I5/Intel Processor

• Monitor : any

2.2 Software Requirements

Software Requirements deal with defining software resource requirements and prerequisites that need to be
installed on a computer to provide optimal functioning of an application. These requirements or prerequisites
are generally not included in the software installation package and need to be installed separately before the
software is installed

• Operating system : Windows 7/8/10

• Server side script : HTML ,CSS, Bootstrap and Java script

• IDE : PyCharm/ VSCode

• Framework : Flask, Pandas, Mysql. connector, Numpy

• Technology : Python 3.6+

• Server Deployment : Xampp Server

• Database : MySQL

BE, 8th SEM, CSE, BITM, Ballari Page | 7

 PREDICTING EMPLOYEE PROMOTIONS

2.3 Functional Requirements:

Functional requirement defines a function of a system or its component. A function is described as a set
of inputs, the behaviour, and outputs. They may be calculations, technical details, data manipulation and
processing and other specific functionality that define what a system is supposed.
• Data Collection and Pre-processing: Collect data on employees including department, region, education,
gender, recruitment channel, previous year rating, length of service, number of trainings attended,
KPI achievement, and training scores. Clean and preprocess the data to handle missing values,
normalize numerical values, and encode categorical variables.
• Exploratory Data Analysis (EDA): Perform EDA to understand the distribution and relationships
between variables. Use data visualization tools to identify patterns and trends in the dataset.
• Feature Selection: Implement feature selection techniques to identify the most significant factors
influencing employee promotions.
• Model Building: Develop predictive models using supervised machine learning algorithms including
Naive Bayes, Support Vector Machine (SVM), and XGBoost. Train the models using the processed
dataset.
• Model Evaluation: Evaluate the models using appropriate metrics such as accuracy, precision, recall,
and F1-score. Compare the performance of different models to select the best-performing one.
• Prediction and Analysis: Use the trained model to predict employee promotions based on input data.
Analyze the predictions to identify key drivers of promotions.
• User Interface: Develop a user-friendly interface for HR professionals to input employee data and
receive promotion predictions.

2.4 Non-Functional Requirements:

Non-functional requirement (NFR) is a requirement that specifies criteria that can be used to judge the
operation of a system, rather than specific behaviours. They are contrasted with functional requirements
that define specific behaviour or functions. The plan for implementing functional requirements is
detailed in the system design. The plan for implementing non-functional requirements is detailed in the
system architecture.
• Performance: Ensure the system can handle large datasets efficiently. Optimize the model training and
prediction processes to minimize computation time.
• Scalability: Design the system to be scalable to accommodate increasing amounts of employee data.
• Accuracy: Maintain high accuracy in predictions to ensure reliable and meaningful results.
• Usability: Ensure the user interface is intuitive and easy to use for HR professionals with varying levels
of technical expertise.

BE, 8th SEM, CSE, BITM, Ballari Page | 8

 PREDICTING EMPLOYEE PROMOTIONS

CHAPTER 3

DESCRIPTION MODULES

Module Description

1. Registration.
2. Login.
3. Upload file.
4. View data.
5. Preprocessing.
6. Prediction modules.

1. Registration:
The user should register in the application. The registration process includes entering details
like, user name, user email ID, password and password confirmation. If the user provides
password correctly in the password confirmation along with valid email ID, then registration
process is successful else the application asks to provide details correctly.
2. Login:
The user login page of this application asks for the login credentials. The user should provide
the login details correctly. Login details include user email ID and password. Upon providing
correct credentials the application directs the user home main page. If unsuccessful, the
application will display the login page again, where we need to fill in the correct details.
3. Upload file:
The dataset for employee promotion, which contains various tweets and related information, is
collected from the website. This dataset includes various employee names and other details like
age, department, KPI score, years of service and so on which help us to predict the employee
promotion.
4. View data:
In this module, we can view the dataset that contains all employee names and details that were
uploaded. Additionally, we need to clean and format the data before preprocessing.

5. Preprocessing:
In this module, we have the option to select the model from one of the following algorithms:
Gaussian, XGBoost, SVM or Voting Classifier. During the preprocessing stage, we first give the
split size which determines the amount of data used for training and testing. We commonly give

BE, 8th SEM, CSE, BITM, Ballari Page | 9

 PREDICTING EMPLOYEE PROMOTIONS

the test size as 30 or 40 which is considered optimal for training any neural network.
6. Prediction:
In this module, the user will provide input. The model will predict, whether the employee is
eligible to be promoted or not. It uses various ML algorithms to predict the same.

BE, 8th SEM, CSE, BITM, Ballari Page | 10

 PREDICTING EMPLOYEE PROMOTIONS

CHAPTER 4

DESIGN

The Design consists of a Use-case diagram, Flow Chart, Activity Diagram, Sequence Diagram

4.1 USE-CASE DIAGRAM

A use case diagram in the Unified Modeling Language (UML) is a type of behavioral diagram defined by
and created from a Use-case analysis. Its purpose is to present a graphical overview of the functionality
provided by a system in terms of actors, their goals (represented as use cases), and any dependencies
between those use cases. The main purpose of a use case diagram is to show what system functions are
performed for which actor. Roles of the actors in the system can be depicted.

Figure4.1: use case diagram of predicting employee promotion system

BE, 8th SEM, CSE, BITM, Ballari Page | 11

 PREDICTING EMPLOYEE PROMOTIONS

4.2 WORK FLOW OF PROPOSED SYSTEM

A flowchart is a type of diagram that represents a workflow or process. A flowchart can also be
defined as a diagrammatic representation of an algorithm, a step-by-step approach to solve a task.

The flowchart shows the steps as boxes of various kinds, and their order by connecting the boxes
with arrows. This diagrammatic representation illustrates a solution model to a given problem.
Flowcharts are used in analyzing, designing, documenting, or managing a process or program in
various fields.

Fig 4.2.1: Flow Chart for work flow of this system.

BE, 8th SEM, CSE, BITM, Ballari Page | 12

 PREDICTING EMPLOYEE PROMOTIONS

Flow chart for Prediction Process:

Fig 4.2.2: Flow Chart of Prediction Process

BE, 8th SEM, CSE, BITM, Ballari Page | 13

 PREDICTING EMPLOYEE PROMOTIONS

4.3 ACTIVITY DIAGRAM

Activity diagrams are graphical representations of workflows of stepwise activities and actions with support
for choice, iteration and concurrency. In the Unified Modelling Language, activity diagrams can be used to
describe the business and operational step-by-step workflows of components in a system. An activity
diagram shows the overall flow of control.

Figure4.3: Activity diagram of Predicting Employee Promotion.

BE, 8th SEM, CSE, BITM, Ballari Page | 14

 PREDICTING EMPLOYEE PROMOTIONS

4.4 SEQUENCE DIAGRAM

A sequence diagram in Unified Modelling Language (UML) represents the interaction between
objects in a particular scenario of a use case. It illustrates the flow of messages exchanged between objects
and the order in which these messages are sent. It is a construct of a Message Sequence Chart. Sequence
diagrams are sometimes called event diagrams, event scenarios, and timing diagrams.

Figure4.4: Sequence diagram of Predicting Employee Promotion.

BE, 8th SEM, CSE, BITM, Ballari Page | 15

 PREDICTING EMPLOYEE PROMOTIONS

4.5 ER Diagram:

An Entity–relationship model (ER model) describes the structure of a database with the help of a diagram,
which is known as Entity Relationship Diagram (ER Diagram). An ER model is a design or blueprint of a
database that can later be implemented as a database. The main components of E-R model are: entity set and
relationship set.

An ER diagram shows the relationship among entity sets. An entity set is a group of similar entities and
these entities can have attributes. In terms of DBMS, an entity is a table or attribute of a table in database, so
by showing relationship among tables and their attributes, ER diagram shows the complete logical structure
of a database. Let’s have a look at a simple ER diagram to understand this concept.

Fig 4.5: ER diagram.

BE, 8th SEM, CSE, BITM, Ballari Page | 16

 PREDICTING EMPLOYEE PROMOTIONS

Chapter 5:
IMPLEMENTATION

5.1 Data Description:

The dataset collected from Kaggle contains information about employees in a company, including attributes
such as department, region, education, gender, recruitment channel, previous year rating, length of service,
and performancerelated metrics like the number of trainings attended, Key Performance Indicator (KPI)
achievement, and training score. And we have 54808 records are there with 14 features. The detail feature
description is as below
• employee_id: A unique identifier for each employee.
• department: The department or functional area within the company where the employee works.
• region: The region or location where the employee is based.
• education: The level of education attained by the employee (e.g., Bachelor's, Master's, etc.).
• gender: The gender of the employee (Male/Female).
• recruitment_channel: The channel through which the employee was recruited (e.g., sourcing,
referred, etc.).
• no_of_trainings: The number of training programs the employee has attended.
• age: The age of the employee.
• previous_year_rating: The performance rating given to the employee in the previous year.
• length_of_service: The length of time the employee has been working in the company.
• KPIs_met>80%: A binary indicator (1/0) indicating whether the employee met KPIs (Key
Performance Indicators) at a level greater than 80%.
• awards_won?: A binary indicator (1/0) showing whether the employee has won any awards.
• avg_training_score: The average score attained by the employee in training programs.
• is_promoted: The target variable, indicating whether the employee was promoted (1) or not (0).
This dataset provides valuable insights into various employee attributes and performancerelated metrics,
allowing for datadriven analysis and prediction of employee promotions. It is a valuable resource for
understanding the factors that influence promotion decisions and for developing effective talent management
strategies within the organization.
Here are the first five rows of the dataset

Figure 5.1 first five rows of the dataset

BE, 8th SEM, CSE, BITM, Ballari Page | 17

 PREDICTING EMPLOYEE PROMOTIONS

5.2 Exploratory Data Analysis (EDA)

5.2.1 Null values:

Figure 5.2.1 Null value of the dataset

Missing Education Data: The 'education' attribute has 2409 missing values, indicating that some employees'
education information is not available. To perform analysis or modeling, handling these missing values is
necessary. Strategies like imputation or creating a separate category for missing values can be considered.
• Missing Previous Year Rating Data: The 'previous_year_rating' attribute has 4124 missing values.
This indicates that some employees' performance ratings from the previous year are not available.
Similar to handling missing education data, appropriate imputation methods can be applied to deal
with missing values in this attribute.
• Diverse Departments and Regions: The 'department' and 'region' attributes show no missing values,
suggesting that data is available for all employees in various departments and regions within the
company. This diversity can be valuable for analyzing promotion patterns across different parts of
the organization.
• Balanced Gender Representation: The 'gender' attribute displays no missing values, indicating that
gender data is available for all employees. Additionally, the absence of missing data suggests a
balanced representation of gender in the dataset, which can enable unbiased analysis of promotion
trends based on gender.
• Training and Age Distribution: The attributes 'no_of_trainings' and 'age' have no missing values,
suggesting complete data for the number of training programs attended by employees and their ages.
This allows for investigating potential relationships between training, age, and promotion prospects.
• Performance Metrics Availability: The attributes 'KPIs_met>80%' and 'awards_won?' exhibit no
missing values, indicating complete data for whether employees met KPIs and won awards. These
performancerelated metrics can be vital factors in predicting employee promotions.
• Average Training Score: The 'avg_training_score' attribute has no missing values, indicating
complete data for the average training scores of employees. This metric can provide insights into
how employee training performance influences promotions.
• Promotion Labels: The 'is_promoted' attribute has no missing values, as expected, since this is the
target variable indicating whether employees were promoted or not. This attribute is crucial for
building and evaluating predictive models for employee promotions.

BE, 8th SEM, CSE, BITM, Ballari Page | 18

 PREDICTING EMPLOYEE PROMOTIONS

In summary, the dataset provides valuable insights into various attributes related to employee promotions.
While there are missing values in the 'education' and 'previous_year_rating' attributes, the available data can
be utilized to gain meaningful understanding of promotion patterns, performance metrics, and training
influences. By appropriately handling missing data and conducting thorough analysis, organizations can
make informed decisions regarding talent management and career progression for their employees.
5.2.2 Information about the data

Figure 5.2.2 Information about the data set

The dataset contains 54,808 entries with 14 columns. The 'education' attribute has 2,409 missing values, and
'previous_year_rating' has 4,124 missing values. The data includes diverse departments and regions with
balanced gender representation. Employees attended various training programs, with ages and length of
service available. Performance metrics like 'KPIs_met>80%' and 'awards_won?' are complete. The dataset
provides valuable insights into employee attributes and performance related metrics, allowing analysis of
promotion trends and talent management strategies.

BE, 8th SEM, CSE, BITM, Ballari Page | 19

 PREDICTING EMPLOYEE PROMOTIONS

5.2.3 Statistical Description

Figure 5.2.3 Statistical Description of the dataset

From the above statistical information, we can derive the following insights:

• Employee ID: The employee IDs range from 1 to 78,298, with a mean of 39,195.8. This suggests a
wide range of employee IDs in the dataset, indicating a large number of employees in the
organization.

• Number of Trainings: On average, employees have attended approximately 1.25 training programs,
with a standard deviation of 0.61. The majority of employees (25th to 75th percentile) have attended
only 1 training program. The maximum number of trainings attended by an employee is 10.

• Age Distribution: The average age of employees is approximately 34.8 years, with a standard
deviation of 7.66. The age ranges from 20 to 60 years, with most employees falling between 29 and
39 years old.

• Previous Year Rating: The average previous year rating is 3.33, indicating that, on average,
employees have received a satisfactory rating. The ratings range from 1 to 5, with the majority of
employees having a rating of 3 or 4.

• Length of Service: The average length of service for employees is approximately 5.87 years, with a
standard deviation of 4.27. Most employees (25th to 75th percentile) have a service length between 3
and 7 years, while the maximum length of service is 37 years.

• KPI Achievement: On average, approximately 35.2% of employees have met or exceeded their Key
Performance Indicators (KPIs). The data is binary, with 0 indicating not meeting the KPI and 1
indicating meeting or exceeding the KPI

• Awards Won: Only a small percentage of employees (approximately 2.3%) have won awards within
the organization, based on the binary data. Most employees (75th percentile) have not won any
awards

BE, 8th SEM, CSE, BITM, Ballari Page | 20

 PREDICTING EMPLOYEE PROMOTIONS

• Average Training Score: The average training score is 63.39, with a standard deviation of 13.37. The
scores range from 39 to 99, indicating a varying level of performance in training programs among
employees

• Promotion Rate: The overall promotion rate is low, with only approximately 8.5% of employees
being promoted. The data is binary, with 0 indicating not being promoted and 1 indicating
promotion.

5.3 Preprocessing
5.3.1 Data transformation

Figure 1. label encoding

Here we are using the label encoder to convert the data from object data type to numeric data type and here
is the data after applying the label encoder.

Figure 2. After conversion of data

5.3.2 Correlation matrix

The provided data represents the correlation matrix between different features/columns in the dataset. Here
are the insights that can be derived from the correlation matrix:

➢ `Correlation`: The values in the matrix range from 1 to 1, indicating the strength and direction of the
correlation between the features. Positive values indicate a positive correlation, negative values
indicate a negative correlation, and values closer to 0 indicate a weak or no correlation.

➢ `Education and Previous Year Rating`: The feature "education" has a positive correlation of 0.34
with the feature "previous_year_rating." This suggests that there is some degree of correlation
between an employee's education level and their performance rating from the previous year.

➢ `Age and Length of Service`: The feature "age" has a relatively strong positive correlation of 0.66
with the feature "length_of_service." This indicates that as the age of employees increases, their
length of service within the organization also tends to increase.

BE, 8th SEM, CSE, BITM, Ballari Page | 21

 PREDICTING EMPLOYEE PROMOTIONS

Figure 3. Correlation matrix

➢ `Promotion and KPIs Met`: The feature "is_promoted" has a positive correlation of 0.22 with the
feature "KPIs_met>80%." This implies that employees who meet or exceed their Key Performance
Indicators have a higher likelihood of being promoted.

➢ `KPIs Met and Previous Year Rating`: The feature "KPIs_met>80%" has a positive correlation of
0.34 with the feature "previous_year_rating." This implies that employees who achieved higher
performance ratings in the previous year are more likely to meet or exceed their Key Performance
Indicators.

➢ `Average Training Score and Department`: The feature "avg_training_score" has a negative
correlation of 0.25 with the feature "department." This suggests that certain departments may have
lower average training scores compared to others.

➢ `Awards Won and Average Training Score`: The feature "awards_won?" has a positive correlation of
0.07 with the feature "avg_training_score." This indicates that employees with higher average
training scores are more likely to have won awards within the organization.

➢ These insights highlight the relationships and associations between different features in the dataset,
providing valuable information for further analysis and understanding of the data.

5.4 K best feature selection method

Mutual Information: MI Estimate mutual information for a discrete target variable. Mutual information (MI)
between two random variables is a nonnegative value, which measures the dependency between the
variables. It is equal to zero if and only if two random variables are independent, and higher values mean
higher dependency. The function relies on nonparametric methods based on entropy estimation from
knearestneighbors distances.

BE, 8th SEM, CSE, BITM, Ballari Page | 22

 PREDICTING EMPLOYEE PROMOTIONS

In short A quantity called mutual information measures the amount of information one can obtain from one
random variable given another. The mutual information between two random variables X and Y can be
stated formally as follows:

I(X ; Y) = H(X) – H(X | Y) Where I(X ; Y) is the mutual information for X and Y, H(X) is the entropy
for X and H(X | Y) is the conditional entropy for X given Y. The result has the units of bits.

Figure 4. Implementing the kbest feature selection

5.5 K Means clustering:

Figure 5. kmeans Clustering

Here we can see that most of the employee is falling under the 0th cluster and as per the promotion point of
view 1 and 3rd cluster employee has high promotion rate as compared to the other clusters.

BE, 8th SEM, CSE, BITM, Ballari Page | 23

 PREDICTING EMPLOYEE PROMOTIONS

CHAPTER6
SYSTEM TESTING

System testing ensures that the entire integrated software system meets requirements. It tests a configuration
to ensure known and predictable results. An example of system testing is the configuration oriented system
integration test. System testing is based on process descriptions and flows, emphasizing pre-driven process
links and integration points.

6.1 Unit Testing

Unit testing is usually conducted as part of a combined code and unit test phase of the software lifecycle,
although it is not uncommon for coding and unit testing to be conducted as two distinct phases.

Test strategy and approach:

Field testing will be performed manually and functional tests will be written in detail.
Test objectives
• All field entries must work properly.
• Pages must be activated from the identified link.
• The entry screen, messages and responses must not be delayed.
Features to be tested
• Verify that the entries are of the correct format
• No duplicate entries should be allowed
• All links should take the user to the correct page.

6.2 Integration Testing

Software integration testing is the incremental integration testing of two or more integrated software
components on a single platform to produce failures caused by interface defects.

The task of the integration test is to check that components or software applications, e.g. components in a
software system or – one step up – software applications at the company level – interact without error.

Test Results: All the test cases mentioned above passed successfully. No defects encountered.

6.3 User Acceptance Testing

User Acceptance Testing is a critical phase of any project and requires significant participation by the end
user. It also ensures that the system meets the functional requirements.

Test Results: All the test cases mentioned above passed successfully. No defects encountered.
BE, 8th SEM, CSE, BITM, Ballari Page | 24
 PREDICTING EMPLOYEE PROMOTIONS

6.4 WhiteBox Testing

White Box Testing is a testing in which in which the software tester has knowledge of the inner workings,
structure and language of the software, or at least its purpose. It is purpose. It is used to test areas that cannot
be reached from a black box level.

6.5 BlackBox Testing

Black Box Testing is to sting the software without any knowledge of the inner workings, structure or
language of the module being tested. Black box tests, as most other kinds of tests, must be written from a
definitive source document, such as specification or requirements document, such as specification or
requirements document. It is a testing in which the software under test is treated, as a black box. you cannot
“see” into it. The test provides inputs and responds to outputs without considering how the software works.

BE, 8th SEM, CSE, BITM, Ballari Page | 25

 PREDICTING EMPLOYEE PROMOTIONS

CHAPTER7
RESULTS

Fig7.1:Homepage

BE, 8th SEM, CSE, BITM, Ballari Page | 26

 PREDICTING EMPLOYEE PROMOTIONS

Fig7.2: Registration Form

Fig7.3: Login page.

BE, 8th SEM, CSE, BITM, Ballari Page | 27

 PREDICTING EMPLOYEE PROMOTIONS

Fig7.4: Upload dataset.

Fig7.5: Data Preprocessing.

BE, 8th SEM, CSE, BITM, Ballari Page | 28

 PREDICTING EMPLOYEE PROMOTIONS

Fig 7.6: Model training.

Fig 7.7: Enter Input.

BE, 8th SEM, CSE, BITM, Ballari Page | 29

 PREDICTING EMPLOYEE PROMOTIONS

Fig 7.8: Result predicting that the employee is promoted.

Fig 7.9: Result predicting that the employee is not promoted.

BE, 8th SEM, CSE, BITM, Ballari Page | 30

 PREDICTING EMPLOYEE PROMOTIONS

CONCLUSION

In this research we have developed a web application where it results an employee is eligible for promotion
or not. We have calculated this based on few factors i.e., Experience, Previous year ratings, KPI’s
performed, Average score of trainings and No. of trainings attended by the employee. This web application
also considered basic factors like gender, age, region, education and department of an employee to calculate
and show whether employee is eligible for promotion or not. To perform this, we have used machine
learning models of Decision Tree, Gaussian, random forest and Support Vector Machine (SVM).

BE, 8th SEM, CSE, BITM, Ballari Page | 31

 PREDICTING EMPLOYEE PROMOTIONS

REFERENCES
[1] Y. Asim, B. Raza, A. K. Malik, S. Rathore and A. Bilal,” Improving the Performance of Professional
Blogger’s Classification”, 2018. International Conference on Computing, Mathematics and Engineering
Technologies (iCOMET), Sukkur, 2018, pp. 1-6.

[2]T.W Ramdhani, B. Purwandari and Y. Ruldeviyani, ”The Use of Data Mining Classification Technique to
Fill in Structural Positions in Bogor Local Government” 2016 International conference on Advanced
computer Science and Information Systems (ICACSIS), Malang 2016.

[3] V. Mathew, A.M. Chacko and A. Udhayakumar, “Prediction of suitable human resource for replacement
in skilled job positions using Supervised Machine Learning “2018 8th International Symposium on
Embedded Computing and System Design (ISED), Cochin, India, 2018.

[4] A. D. Hartanto, E. Utami, S. Adi and H.S.Hudnanto,” Job Seeker Profile Classification of Twitter Data
Using the Naïve Bayes Classifier Algorithm Based on the DISC Method”, 2019 4th International Conference
on Information Technology, Information Systems and Electrical Engineering(ICITISEE), Yogyakarta,
Indonesia, 2019.

[5] T. Tarusov and O. mitrofanova, “Risk Assessment in Human Resource Management Using Predictive
Staff Turnover Analysis”, 2019 1st International Conference on Control System, Mathematical Modelling,
Automation and Energy Efficiency (SUMMA), Lipetsk, Russia, 2019.

[6] Qiangwei Wang, Boyang Li and Jinglu Hu, “Feature Selection for Human Resource Selection Based on
Affinity Propagation and SVM Sensitivity Analysis,” 2009 World Congress on Nature & Biologically
Inspired Computing (NaBIC), Coimbatore, 2009.

[7] M. Eminagaoglu and S. Eren, “Implementation and comparison of machine learning classifiers for
information security risk analysis of a human resources department,’ 2010 International Conference on
Computer Information System and Industrial Management Applications (CISIM), Krakow, 20010.

[8] L.I.F Dutsinm and P. Temdee,”VARK Learning Style Classification Using Decision Tree with
physiological signals,” 2020 Wireless personal Communication, 2020.

[9] Q.Guohao, W. Bin and Z. Baoil,”Competency Analysis in Human Resources Using Text Classification
Based on Deep Neural Network,” 2019 IEEE Fourth International Conference on Data Science in
Cyberspace (DSC).

[10] N. Aottiwerch and U. Kokaew, “The analysis of matching learners in pair programming using K-
means,” 2018 5th International Conference on Industrial and Applications (ICIEA), Singapore, 2018.

BE, 8th SEM, CSE, BITM, Ballari Page | 32

 PREDICTING EMPLOYEE PROMOTIONS

BALLARI INSTITUTE OF TECHNOLOGY AND MANAGEMENT

DEPARTMENT OF COMPUTER SCIENCE & ENGINEERING

Project CO-PO Mapping

ACADEMIC YEAR 2023-24

U.S.N Student Guide Name Project

Name Title
3BR20CS106 MONISHA V Mr.V.CHANDRASHEKAR PREDICTING
3BR20CS113 NAVYA GADIGI (AsstProf) EMPLOYEE
PROMOTION
3BR20CS133 RINI SHARON
3BR20CS139 MALLIKA S

COURSEOUTCOMES(CO'S)

Course Description of Course Outcomes

Outcomes
Cox
CO1 Identify the problem in the field of educational organizations.
CO2 Analyze the problem of Manual attendance in educational schools, colleges, and
offices.
CO3 Design the solution methodologies for the fetch of attendance in the educational sector.

CO4 Write technical Project reports by following professional ethics.

CO5 Create and publish the outcome of the thesis into an article.

CO-POMAPPING

CO-
PO PO1 PO2 PO3 PO4 PO5 PO6 PO7 PO8 PO9 PO10 PO11 PO12 PSO1 PSO2

CO1 L H L
CO2 H
CO3 L H L L
CO4 M M L
CO5 L L L

Signature of Guide

BE, 8th SEM, CSE, BITM, Ballari Page | 33

 PREDICTING EMPLOYEE PROMOTIONS

PROGRAMOUTCOMESANDPROGRAMSPECIFICOUTCOMES
(POs& PSOs)
PO1. Engineering knowledge: Apply the knowledge of mathematics, science, engineering fundamentals,
and an engineering specialization to the solution of complex engineering problems.

PO2. Problem analysis: Identify, formulate, review research literature, and analyze complex
engineeringproblemsreachingsubstantiatedconclusionsusingfirstprinciples of mathematics, natural
sciences, and engineering sciences.

PO3. Design/development of solutions: Design solutions for complex engineering problems and design
system components or processes that meet the specified needs with appropriate consideration for public
health and safety, and cultural, societal, and environmental considerations.

PO4. Conduct investigations of complex problems: Use research-based knowledge and research
methods including design of experiments, analysis and interpretation of data, and synthesis of the
information to provide valid conclusions.

PO5. Modern tool usage: Create, select, and apply appropriate techniques, resources, and modern
engineering and IT tools including prediction and modelling to complex engineering activities with an
understanding of the limitations.

PO6. The engineer and society: Apply reasoning informed by the contextual knowledge to assess
societal, health, safety, legal, and cultural issues and the consequent responsibilities relevant to the
professional engineering practice.

PO7. Environment and sustainability: Understand the impact of professional engineering solutions in
societal and environmental contexts, and demonstrate the knowledge of, and need for sustainable
development.

PO8. Ethics: Apply ethical principles and commit to professional ethics and responsibilities and norms of
the engineering practice.

PO9. Individual and teamwork: Function effectively as an individual, and as a member or leader in
diverse teams, and in multidisciplinary settings.

PO10. Communication: Communicate effectively on complex engineering activities with the engineering
community and with society at large, such as being able to comprehend and write effective reports and
design documentation, make effective presentations, and give and receive clear instructions.

PO11. Project management and finance: Demonstrate knowledge and understanding of the engineering
and management principles and apply these to one’s work, as a member and leader in a team, to manage
projects and in multidisciplinary environments.

PO12. Life-long learning: Recognize the need for, and have the preparation and ability to engage in
independent and life-long learning in the broadest context of technological change.

PSO1 Demonstrate the principles, architecture, and organization of computers, embedded systems, and
computer networks.

PSO2 Develop software applications using advanced technologies to cater to the growing needs of the
industry

BE, 8th SEM, CSE, BITM, Ballari Page | 34

BE, 8th SEM, CSE, BITM, Ballari Page |
42
[Type here] [Type here] [Type here]
 International Journal of Research Publication and Reviews, Vol (5), Issue (5), May (2024), Page –5646-5650

International Journal of Research Publication and Reviews

Journal homepage: www.ijrpr.com ISSN 2582-7421

A NOVAL APPROACH FOR PREDICTING EMPLOYEE

PROMOTION

Mr. V Chandrashekar1, Monisha V2, Navya Gadigi3, Rini Sharon4, Mallika S5

1
Asst. professor, Department of CSE, Ballari Institute of Technology & Management, Ballari. [email protected]
2
Department of CSE, Ballari Institute of Technology & Management, Ballari [email protected]
3
Department of CSE, Ballari Institute of Technology Ballari [email protected]
4
Department of CSE, Ballari Institute of Technology & Management, Ballari [email protected]
5
Department of CSE, Ballari Institute of Technology Ballari [email protected]@gmail.com

ABSTRACT:

This piece of work investigates the analysis and prediction of employee promotions within a company, focusing on various factors including the number of
trainings attended, Key Performance Indicator (KPI) achievement, and training score. Employee promotions are crucial for organizational growth and employee
career advancement, necessitating a comprehensive understanding of the contributing factors for informed decision-making in employee development and
performance management. Employing exploratory data analysis techniques, including data visualization and summary statistics, insights into variable
distributions and relationships are gained. Subsequent application of predictive modelling techniques, notably machine learning algorithms like decision trees,
random forests, and logistic regression, facilitates the development of a model capable of accurately predicting employee promotions. Evaluation metrics are then
employed to assess the model's performance and predictive accuracy. The findings offer valuable insights into the determinants of employee promotions, enabling
organizations to identify promotion drivers and formulate strategies for enhancing employee performance and career development. This work contributes to the
understanding of employee promotions, providing organizations with actionable insights for effective talent management and data-driven decision-making. Such
insights are poised to improve employee satisfaction, engagement, and overall organizational performance.

Key words: Employee Promotion, Predictive Analysis, Machine Learning, Supervised Machine Learning.

INTRODUCTION:

Employee promotions are vital for organizational growth and employee career development. This research project aims to predict and analyse
employee promotions based on three key factors: the number of trainings attended, KPI achievement, and training scores. Understanding these factors'
influence can help organizations make informed decisions for talent management and performance improvement. The study utilizes a dataset with
employee attributes, and through exploratory data analysis and machine learning algorithms, aims to build a predictive model. The model's performance
will be evaluated to ensure accuracy in predicting promotions. The project's significance lies in providing valuable insights into promotion drivers,
enabling data-driven decisions for employee development and fostering a culture of growth and engagement. By reducing biases and subjectivity in the
promotion process, the research aims to contribute to a fair and inclusive work environment. Ultimately, organizations can enhance employee
satisfaction, engagement, and overall performance by leveraging the predictive model's insights to make strategic talent management decisions.

Background:

The background study for this project focuses on the significance of employee promotions in organizational growth and employee career development.
Promotions are essential for recognizing and rewarding exceptional employee performance, motivating individuals to excel in their roles, and fostering
a culture of growth and loyalty within the organization. Employee promotions play a pivotal role in talent retention, as they provide employees with a
sense of career progression and personal achievement. Promotions also act as a strong incentive for employees to continually improve their skills and
contribute to the company's success.
However, the promotion process can be subjective and influenced by various biases, leading to potential disparities and inefficiencies. Hence, there is a
growing need for objective and data-driven approaches to identify the most deserving candidates for promotions. By analysing the factors that influence
promotions, such as the number of trainings attended, KPI achievement, and training scores, organizations can develop a more transparent and fair
promotion process. The background study also highlights the growing significance of data analytics and machine learning in talent management
practices. By leveraging these techniques, organizations can gain valuable insights from employee data, enabling them to make more informed and
strategic decisions regarding promotions, employee development, and performance improvement. Ultimately, this research project aims to contribute to
 International Journal of Research Publication and Reviews, Vol (5), Issue (5), May (2024), Page –5646-5650 5647

the understanding of employee promotions and provide organizations with valuable insights for effective talent management, leading to enhanced
employee satisfaction, engagement, and overall organizational performance.

Research Objectives:

This research project endeavours to enhance the accuracy and efficacy of employee promotion prediction models while offering valuable insights for
human resource management and decision-making processes. The primary objectives encompass a comprehensive exploration of advanced techniques
and algorithms in machine learning and feature engineering. Initially, the project aims to mitigate class imbalance within the dataset through feature
engineering methodologies such as the Synthetic Minority Oversampling Technique (SMOTE). By rectifying imbalances, model performance is
expected to improve significantly, ensuring fair and accurate predictions of promotion outcomes.
In addition, the project seeks to harness the power of unsupervised machine learning techniques, specifically K-Means clustering, to uncover
underlying patterns within employee attributes. This approach facilitates the identification of distinct employee groups based on shared characteristics,
thereby providing insights into performance trends and promotion potential. Furthermore, the application of the K-Best feature selection method aims
to streamline the dataset by identifying the most influential features contributing to employee promotion. By reducing dimensionality and focusing on
critical predictors, this step enhances model efficiency and interpretability, laying the groundwork for more robust predictive models.
Subsequently, the project progresses to the implementation of supervised machine learning algorithms, including MLP classifier, linear discriminant
analysis, and k-nearest neighbours, to construct predictive models for employee promotion based on the selected features. Rigorous evaluation and
comparison of these models using appropriate metrics such as accuracy, precision, recall, and F1 score provide insights into their respective strengths
and weaknesses. Ultimately, the project aims to translate model predictions into actionable insights and recommendations for human resource
management. By identifying key factors influencing employee promotion and proposing strategies for performance improvement and career growth,
this research contributes to the advancement of talent management practices and decision-making processes within organizations.

Problem Statement:

The problem addressed in this project is the prediction of employee promotions in a company. Employee promotions play a vital role in recognizing
and rewarding exceptional performance, motivating employees, and ensuring career growth within the organization. However, the promotion process is
often subjective and lacks a clear and standardized methodology. The challenge is to identify the key factors that contribute to employee promotions
and develop a predictive model that can accurately forecast the likelihood of an employee being promoted. The project aims to address this problem by
analysing the influence of three important factors: the number of trainings attended, KPI achievement, and training score.
By exploring the relationships between these factors and employee promotions, the project seeks to uncover patterns and insights that can guide
decision making processes related to promotions. The development of a predictive model will enable organizations to make more informed and data-
driven decisions regarding employee promotions, thereby promoting fairness, transparency, and objectivity in the promotion process. Ultimately, this
project aims to enhance employee satisfaction, engagement, and retention by establishing a more effective and efficient system for promoting deserving
employees

Motivation :

The motivation behind the development and research on "Employee promotion prediction" stems from the realization that fairness and transparency in
employee promotion should not be compromised in the digital age. By analysing historical promotion data and performance metrics, organizations can
gain valuable insights into what factors contribute to employee success and advancement. This information can be used to refine recruitment, training,
and performance management strategies.

Objectives of the Proposed work:

In the existing system, the process of predicting employee promotions relies heavily on manual evaluation and subjective judgment. HR professionals
and managers assess employees based on their experience, performance reviews, and personal observations. They consider factors such as educational
qualifications, previous work experience, performance ratings, length of service, and potential for growth within the organization. The main objective
of this system is to overcome these problems.
1. High prediction accuracy and performance on various types of datasets.
2. Flexibility in handling both numerical and categorical features.
3. Capability to handle high-dimensional data and feature interactions.
4. Interpretability and transparency, allowing for insights into the decision-making process.
As we delve deeper into the intricacies of "Employee Promotion Prediction" in subsequent sections, it is imperative to recognize its potential in
redefining the narrative of personal promotion and growth in the digital age. Through a nuanced exploration of its features, development process, and
implementations, this work aims to position "Employee Promotion Prediction" as a beacon of hope in the quest for a fairer, unbiased and more
transparent solution for the promotion of an employee in the future.
 International Journal of Research Publication and Reviews, Vol (5), Issue (5), May (2024), Page –5646-5650 5648

METHODOLOGY:

The proposed methodology to predict the employee promotion is accomplished in following steps:-
Feature selection using K-Best
Applying K-Means clustering method
Choosing a Multilayer Perceptron MLP Classifier
Implementing the Linear Discriminant Analysis (LDA)
Applying the K-Nearest Neighbours (KNN)

This work is justified based on their specific advantages and potential contributions to predicting employee promotions.
Step-1: K-Best Feature selection:
K-Best feature selection method is justified for this project as it helps identify the most relevant features from a large dataset, allowing us to focus on
the key factors influencing employee promotions. This approach reduces dimensionality, improves model performance, and enhances interpretability by
selecting the most informative features.
Step-2: K-Means clustering;
K-Means clustering is a popular unsupervised learning technique that can group similar data points together based on their attributes. By applying K-
Means clustering in this project, we can identify distinct employee segments or clusters based on their characteristics such as education, job tenure, and
performance metrics. This can provide valuable insights into different employee profiles and their promotion probabilities, enabling a more targeted
and customized approach to promotion prediction.
Step-3: MLP Classifier (Multilayer Perceptron):
MLP is a type of artificial neural network known for its ability to learn complex patterns. It consists of multiple layers of interconnected nodes
(neurons) with nonlinear activation functions. Through backpropagation, MLP adjusts the weights to minimize errors in predictions during training. It's
effective for tasks like classification, where it learns to map input data to output classes. MLP excels in handling large datasets and can capture intricate
relationships between features. However, it requires careful tuning of parameters and can be prone to overfitting without proper regularization.
Step-4: Linear Discriminant Analysis (LDA):
LDA is a statistical technique used for dimensionality reduction and classification. It finds linear combinations of features that best separate different
classes in the data. By modelling the distribution of classes, LDA determines decision boundaries to classify new instances. It assumes that the features
are normally distributed and that classes have identical covariance matrices. LDA is particularly useful when classes are well separated and the
assumptions hold. It's computationally efficient and provides insight into the discriminatory power of features. However, it may perform poorly if the
assumptions are violated or classes overlap significantly.
Step-5: K-Nearest Neighbours (KNN):
KNN is a simple and intuitive classification algorithm that operates based on similarity measures. It classifies a new data point by a majority vote of its
k nearest neighbours in the feature space. The choice of k influences the algorithm's decision boundary and sensitivity to noise. KNN doesn't involve
explicit training; instead, it memorizes the entire training dataset. It's effective for both linear and nonlinear decision boundaries and handles multiclass
classification naturally. However, its performance can degrade with high-dimensional data and requires efficient data structures for fast querying.
Choosing an optimal k value is crucial for optimal performance.
 International Journal of Research Publication and Reviews, Vol (5), Issue (5), May (2024), Page –5646-5650 5649

Figure 1 :Architecture diagram for

RESULT:

This system, after being put into action and thoroughly checked, showed some important positive outcomes. Firstly, we made sure the system is easy
for everyone to use. We tackled the issue like biased or selective promotion of personnel. The proposed system is a stand-alone application and is
suitable only for employees of any organisation that analyses the key performance indicator. This system is proved to be providing more accuracy,
ensuring that it can be useful for any organisation. This system provides improvement in employee satisfaction and retention rates. It helps to reduce in
turnover and associated cost savings. This system also indulges in enhancement of the performance and productivity levels. This system promotes
fairness and transparency in the promotion process. Thus, strengthening of succession planning and strategic talent management. Also enables benefits
of data-driven decision-making in promotion-related matters.

DISCUSSION:

These positive results show that the system is doing well in meeting our goals. Making it more accurate means more people can use it. This
methodology was chosen to focus on the following:
 Perform feature engineering techniques, such as SMOTE (Synthetic Minority Oversampling Technique), to address class imbalance in the
dataset and improve the performance of the models.
 Utilize unsupervised machine learning techniques, specifically K-Means clustering, to identify patterns and group employees based on their
attributes, which can provide insights into employee performance and promotion potential.
 Apply the K-Best feature selection method to identify the most relevant and influential features that contribute to employee promotion,
thereby reducing the dimensionality of the dataset and improving model efficiency.
 Implement supervised machine learning algorithms, including MLP classifier, linear discriminant analysis, k-neighbours to build predictive
models for employee promotion based on the selected features.
 Evaluate and compare the performance of the different models using appropriate evaluation metrics such as accuracy, precision, recall, and
F1score.
 Provide insights and recommendations based on the model predictions, identifying key factors that contribute to employee promotion and
suggesting strategies for improving employee performance and career growth within the organization.

CONCLUSION:

The proposed methodology has been implemented and tested for correctness and completeness of the system by considering several case studies. The
system presents a comprehensive analysis of the impact of training quality, KPI attainment, and training evaluation on employee promotion prospects,
employing a range of machine learning classifiers including KNearest Neighbors (KNN), Linear Discriminant Analysis (LDA), and Multilayer
Perceptron (MLP). Through thorough data exploration, feature engineering, and rigorous model evaluation, we have gained valuable insights into the
factors influencing employee promotions. Additionally, we employed SelectKBest feature selection technique to identify the most relevant features for
predicting employee promotions, enhancing model interpretability and performance. By integrating hyperparameter tuning to finetune our models, we
ensured optimal performance across various algorithms. This multialgorithm approach, coupled with feature selection and hyperparameter tuning,
ensures the robustness and reliability of our predictive model, facilitating more accurate decisionmaking in talent management. Ultimately, this project
contributes to the advancement of HR practices by offering a versatile and datadriven tool for organizations to optimize employee career development
and organizational growth.

REFERENCE:

1. Smith, A., Johnson, B., & Williams, C. (2017). Predicting employee promotions using machine learning and feature engineering. Journal of
Applied Data Science, 5(2), 123136.
2. Johnson, C., Thompson, L., & Davis, M. (2018). The impact of training programs on employee promotion probabilities: A logistic
regression and decision tree analysis. International Journal of Human Resource Management, 29(7), 13231342.
3. Lee, H., & Kim, J. (2019). Predicting employee promotions using key performance indicators: A longitudinal study. Journal of
Organizational Behaviour, 40(5), 587602.
4. Chen, X., & Liu, Y. (2020). Unsupervised learning for employee promotion analysis: A case study using Kmeans clustering. Expert Systems
with Applications, 151, 113358.
5. Khadar, A. A., Hayath, T. M., & Nagraj, M. S. (2017, August). Design and implementation of ZigBee based smart grid system for power
management. In 2017 International Conference On Smart Technologies For Smart Nation (SmartTechCon) (pp. 1339-1343). IEEE.
6. Wang, Z., Zhang, Q., & Li, C. (2021). Feature selection for employee promotion prediction using KBest algorithm. International Journal of
Machine Learning and Cybernetics, 12(6), 13791390.
7. Li, J., Zhao, L., & Zhang, G. (2018). Employee promotion prediction using XGBoost algorithm. In Proceedings of the International
Conference on Advanced Data Mining and Applications (pp. 682695). Springer.
 International Journal of Research Publication and Reviews, Vol (5), Issue (5), May (2024), Page –5646-5650 5650

8. Zhang, Y., & Guo, L. (2019). Predicting employee promotions using support vector machines. In Proceedings of the International
Conference on Machine Learning and Data Mining in Pattern Recognition (pp. 452464). Springer.
9. Park, S., Kim, D., & Lee, J. (2020). Employee promotion prediction using random forest algorithm: A case study in a manufacturing
company. Journal of Industrial Engineering and Management, 13(4), 709725.
10. Liu, X., Liu, L., & Wang, X. (2021). Deep learning for employee promotion prediction: A comparative study of neural network models.
Expert Systems with Applications, 175, 114810.
11. Wang, Y., Wang, Y., & Li, H. (2022). Employee promotion prediction based on ensemble learning methods. Journal of Business Research,
137, 179190.
Powered by TCPDF (www.tcpdf.org)