M. Tech.

Project (COC6919) Report On

Career Prediction System Using

Machine Learning
SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE AWARD OF THE DEGREE OF

Master of Technology
IN
COMPUTER SCIENCE & ENGINEERING

BY

MOHAMMAD NOUMAN
22CPMEA115

Under the Guidance of

Dr. M.S. UMAR

Department Of Computer Engineering

Zakir Husain College of Engineering & Technology
Aligarh Muslim University
Aligarh (India)-202002, India

2021-2022
 Declaration

The work presented in this report for M.Tech. Project (COC6919) entitled “Career
Prediction System Using Machine Learnig” submitted to the Department of Computer
Engineering, Zakir Husain College of Engineering and Technology, Aligarh Muslim University
Aligarh, for the award of the degree of Master of Technology in Computer Science &
Engineering is my original work. I have neither plagiarized nor submitted the same work for
the award of any other degree.

Date: 31stMay, 2023

Place: Aligarh Muslim University, Aligarh MOHAMMAD NOUMAN
 Certificate

This is to certify that the Report for M.Tech. Project (COC6919) entitled “Career Prediction
System Using Machine Learning”, being submitted by “Mohammad Nouman”, in partial
fulfillment of the requirements for the award of the degree of Master of Technology in
Computer Science & Engineering in the Department of Computer Engineering, Zakir Husain
College of Engineering and Technology, Aligarh Muslim University Aligarh, is a record of
candidate’s own work carried out by him under my (our) supervision and guidance.

Date…31st May, 2023 Dr. M.S. Umar

Professor
Department of Computer Engineering
ZHCET, AMU, Aligarh
 Table of Contents

Abstract V

List of Figures VI

1 Introduction 1

1.1 Error! Bookmark not defined.

1.2 Error! Bookmark not defined.

1.3 Scope 1

2 Literature Review ................................................................................................... 2

2.1 paper 1 2

2.2 paper 2

2.3 paper 3 2

3 Design and Implementation ..................................................................................... 3

3.1 Dataset overview 3

3.2 Requirements 3

3.3 Project pipeline 3

3.4 Model Validation 3

4 Results and Discussion.............................................................................................. 4

5 Conclusions and Future Work ................................................................................ 5

4.1 Conclusions 5

4.2 Recommendation in Future Work 5

References ..............................................................................................................................6
 Abstract

This project proposes a career prediction system using machine learning and a random forest
classifier. The system is trained on a dataset of student records, including their academic
performance, subjects, age, and interests. The system then uses this data to predict the student's
likely career path. The system was evaluated on a test set of students, and it was found to have an
accuracy of 97%. This suggests that the system could be used to help students make informed
decisions about their future careers. The project also discusses the limitations of the system and
suggests ways to improve it in the future.

Overall, this project is a promising step towards developing a more effective career prediction
system. The system has the potential to help students make informed decisions about their future
careers, and it could also be used to help employers identify potential employees.
 List of Figures

Figure 3.1: Dataset used in the proposed system. 13

Figure 3.2: Project pipelining. 18
Figure 3.3: Random Forest Regrassion. 19
Figure 3.4: Confussion Metrics. 20
Figure 3.5: Accuracy and Cross-validation Score. 20
Figure 3.6: Error Score. 21
Figure 4.1: User Interface. 22
Figure 4.2: User INPUT. 22
Figure 4.3: RESULT predicted. 23
 1. INTRODUCTION

In today's rapidly evolving job market, individuals are often faced with the unnerving task of
making career decisions that align with their skills, interests, and aspirations. The traditional approach of relying
completely on personal judgment and parental advice can be irresistible and may not always lead to the most
suitable career choices. This is where the power of machine learning comes into play.

Models of Machine Learning aims to leverage the capabilities of machine learning algorithms to
create a system that can provide data-driven predictions and recommendations for career paths by analysing
various data points.
The project revolves around the idea that machine learning algorithms can sift through vast
amounts of data, identify patterns, and uncover insights that may not be immediately apparent to human
observers. By training these algorithms on historical career data, the model can learn from past trends and patterns
to predict potential career paths for individuals based on their unique attributes and aspirations.

1. Motivation
The project "Career Prediction Model Using Machine Learning" aims to leverage the capabilities of
machine learning algorithms to create a system that can provide data-driven predictions and recommendations for
career paths. By analysing various data points such as interested subjects, skills or hobbies, marks percentage in
12th, and their age, this model seeks to assist individuals in making informed decisions about their professional
journey.

The project revolves around the idea that machine learning algorithms can sift through vast amounts of
data, identify patterns, and uncover insights that may not be immediately apparent to human observers. By
training these algorithms on historical career data, the model can learn from past trends and patterns to predict
potential career paths for individuals based on their unique attributes and aspirations.

Through the utilization of different machine learning techniques such as classification and regression, the
career prediction model aims to offer personalized recommendations that consider individual strengths,
weaknesses, and interests. By leveraging the power of data analytics, the system can provide valuable insights
into the compatibility between individuals and specific career domains, helping to narrow down choices and guide
career decisions.

The potential impact of this project is significant, as it has the potential to revolutionize the way
individuals approach their career planning. By providing data-driven insights and predictions, the career
prediction model can empower individuals to make well-informed decisions, maximize their potential, and
increase their chances of success in their chosen professions.

1.1.1. Machine Learning

Machine learning is a subfield of artificial intelligence (AI) that focuses on the development of algorithms
and models capable of learning from and making predictions or decisions based on data. It enables computers to
automatically learn and improve from experience without being explicitly programmed. The core idea behind
machine learning is to build mathematical models that can analyse and interpret large volumes of data, identify
patterns, and make predictions or take actions based on those patterns. These models learn by extracting

6
meaningful insights and relationships from the data, enabling them to generalize and make accurate predictions on
new, unseen data.
There are several types of machine learning algorithms, including supervised learning, unsupervised
learning, and reinforcement learning. In supervised learning, the algorithm is trained on labelled data, where each
data point is associated with a known target or outcome. The algorithm learns to map the input features to the
corresponding target values, enabling it to make predictions on new, unseen data.
Unsupervised learning, on the other hand, deals with unlabelled data, where the algorithm learns to
identify patterns, clusters, or hidden structures within the data without any predefined target variable. It discovers
the inherent structure or relationships in the data, allowing for data exploration and discovery of insights.
Reinforcement learning involves training an agent to interact with an environment and learn by receiving
feedback in the form of rewards or penalties. The agent explores the environment, takes actions, and learns to
maximize its cumulative reward over time by trial and error.
However, it is important to note that machine learning is not a magic solution. It requires careful data
preparation, feature engineering, model selection, and continuous monitoring to ensure accurate and reliable
results. Additionally, ethical considerations such as bias and privacy need to be addressed when working with
sensitive data and deploying machine learning models.

1.1.2. MACHINE LEARNING APPLICATIONS

Machine learning algorithms are a set of instructions that allow computers to learn without being
explicitly programmed. Machine learning algorithms are used in a wide variety of applications, including:
1. Image recognition: Machine learning algorithms can be used to identify objects in images. This is used
in applications like facial recognition, object detection, and image classification.
2. Natural language processing: Machine learning algorithms can be used to understand human language.
This is used in applications like speech recognition, machine translation, and text summarization.
3. Predictive analytics: Machine learning algorithms can be used to predict future events. This is used in
applications like fraud detection, risk assessment, and customer churn prediction.
4. Recommendation systems: Machine learning algorithms can be used to recommend products or
services to users. This is used in applications like Netflix, Amazon, and Spotify.
5. Robotics: Machine learning algorithms can be used to control robots. This is used in applications like
self-driving cars, drones, and industrial robots.
These are just a few of the many applications of machine learning algorithms. As machine learning
technology continues to develop, we can expect to see even more applications of machine learning in the future.

Here are some examples of machine learning algorithms:

1. Linear regression: Linear regression is a simple machine learning algorithm that can be used to predict
a continuous value. For example, linear regression can be used to predict the price of a house based on its square
footage, number of bedrooms, and location.

7
 2. Logistic regression: Logistic regression is a machine learning algorithm that can be used to predict a
binary value. For example, logistic regression can be used to predict whether a customer will click on an ad based
on their demographics and browsing history.
3. Decision trees: Decision trees are a machine learning algorithm that can be used to make decisions
based on a set of rules. For example, a decision tree can be used to decide which treatment is best for a patient
based on their symptoms.
4. Random forests: Random forests are a machine learning algorithm that combines the predictions of
multiple decision trees. This approach helps to reduce the variance of the predictions and improve the overall
accuracy of the model.
5. Support vector machines: Support vector machines are a machine learning algorithm that can be used
to classify data into two or more classes. For example, support vector machines can be used to classify email
messages as spam or not spam.

These are just a few of the many machine learning algorithms that are available. The best algorithm for a
particular application will depend on the specific problem that is being solved. Machine learning is a powerful
tool that can be used to solve a wide variety of problems. As machine learning technology continues to develop,
we can expect to see even more applications of machine learning in the future.

1.1.3. RANDOM FOREST CLASSIFIER

Random forest is an ensemble learning method for classification, regression and other tasks that operates
by constructing a multitude of decision trees at training time. Random forests are a popular choice for
classification problems because they can achieve high accuracy and are relatively easy to interpret.
Random forests work by constructing a large number of decision trees, each of which is trained on a
random subset of the training data. The predictions of the individual trees are then combined to produce a final
prediction. This approach helps to reduce the variance of the predictions and improve the overall accuracy of the
model. Random forests are a powerful tool that can be used to solve a wide variety of classification problems.
They are relatively easy to implement and can be trained on large datasets. Random forests are also relatively
robust to overfitting, which can be a problem with other machine learning algorithms.

Here are some of the advantages of using random forest classifiers:

1. Accuracy: Random forest classifiers are known for their high accuracy. They can achieve high
accuracy even when the data is noisy or imbalanced.
2. Robustness: Random forest classifiers are robust to overfitting. This means that they can still perform
well even when the training data is not representative of the test data.
3. Interpretability: Random forest classifiers are relatively easy to interpret. This is because each
decision tree in the forest can be examined individually.

Here are some of the disadvantages of using random forest classifiers:

1. Computational complexity: Random forest classifiers can be computationally expensive to train.

This is because they require training a large number of decision trees.

8
 2. Sensitivity to hyper-parameters: The performance of random forest classifiers can be sensitive to the
hyper-parameters that are used. This means that it can be difficult to find the optimal hyper-
parameters for a particular dataset.

1.1.4. MODEL IN PROPOSED SYSTEM

As a project introduction, machine learning offers a powerful set of tools and techniques that can be used to train
complex models that will be difficult or impossible to solve with traditional programming methods. By using
Random Forest algorithm to analyse large amount of data, we can identify patterns and make predictions with a
high degree of accuracy in the given project.
So for successfully completing this project based on machine learning approach, it is important to have a clear
understanding of the problem we are trying to solve, the data we have available, and the goals we want to achieve.
We choose the right machine learning algorithm and optimise it for our specific problem by taking into account
factors like data quality, feature selection and model performance.

2. Objective
The proposed system is a "Web Enabled Application" that extracts important information using the Machine
Learning technique. Users with similar qualities and habits will have similar preferences, according to the
proposed approach. Students are divided into groups and offered counsel based on their commonalities in the
proposed system.

3. Scope
The scope of the project is to develop a career prediction system using machine learning and a random forest
classifier. The system will be trained on a dataset of student records, including their academic performance,
interested subjects, age, and interests. The system will then use this data to predict the student's likely career path.

4. Problem Statement
Most students throughout the world are always perplexed about completing schooling and reaching the point
where they must choose a job route. Students under the age of 18, lack the maturity to understand what steps must
be taken to pursue a rewarding professional path.
As we progress through phases, we will find that every student has question or thought about what to do after 12 th
grade, which is the single most difficult question. This becomes our first problem statement.
Another problem is that then there’s question of whether they have the necessary talent for the career path they
have selected.

9
 2. LITERATURE REVIEW

1. PAPER 1

2.1.1. TITLE
Tanya V Yadalam, Vaishnavi M Godwa, Vandita Shiva Kumar, Disha Girish; titled “Career recommendation
system using content based filtering” -ICCES 2020

2.1.2. DATASET
Data set contains 17 columns and 20,000 entries.

2.1.3. METHODOLOGY
Recommendation takes a form based input from the user and recommends an accurate Career
It recommends 3 best career options
Career prediction is based on data filled by the user and perform cosine similarity on that. Model apply cosine
similarity function to get the similarity between previous user preference and the available jobs and finally get the
top recommended jobs according to the scores of the similarity.
The aim is to implement the feedback with a comment and perform NLP on the given feedback and determine
whether it’s a positive and negative or a neutral comment to provide better results to the students using the
recommendation system.

2.1.4. ALGORITHMS
 NLP
 Cosine similarity
 Content based filtering

2.1.5. ADVANTAGES
 Introduce features of security, reliability and transparency.
 Allow users to read the jobs based on their experience on different parameters

2.1.6. DISADVANTAGES
 It is mainly for engineering students and has no data encryption mechanism.

10
 2. PAPER 2

2.2.1. TITLE
Vignesh S, Shivani Priyanka C, Shree Manju H, Mathili K; titled “An intelligent career guidance system using
machine learning”- ICACCS 2021

2.2.2. DATASET
Dataset contains 500 rows and hence 500 unique values attached with each target variables and features.

2.2.3. METHODOLOGY
The framework totally consists of 3 modules where the whole process take place.
First module is skill set assessment module.
Second module is the prediction module where; with the help of the scores obtained by the candidate the
assessment module; prediction takes place.
The 3rd module is the result analysis module, in this module a detailed analysis of the candidate’s performance
will be represented in various forms.

2.2.4. ALGORITHMS
 KNN
 SVM
 K Means Clustering
 Naïve Bayes

2.2.5. ADVANTAGES
 Model has low chances of selecting a department by a candidate where candidate has higher chance of
failure rate.

2.2.6. DISADVANTAGES
 The system is for plus 2 students and they have only knowledge with their academics.

11
 3. PAPER 3

2.3.1. TITLE
Manar Qamheih, Haya Sammaneh, Mona Nabil Demaidi; titled “PCRS: Personalized Career Recommendation
System for Engineering Students” -Under Grant ANNU-2020

2.3.2. DATASET
The MBTI personality test consists of 21 questions randomly chosen from set of 70 questions.

2.3.3. METHODOLOGY
There are more 4 main phases:
Obtaining students personal information including gender, high school grades in main courses and a list of
extracurricular interest.
Determining student’s personality type based on self-administered personality test
Processing input data to construct a personality and academic profile for each student
Build a fuzzy recommended system to provide students with personalised and user specific ranking of engineering
disciplines.

2.3.4. ALGORITHM
 Fuzzy logic

2.3.5. ADVANTAGES
 Helpful for the high school students in developing countries where the education professional guidance
school is limited

2.3.6. DISADVANTAGES
 Extremely time consuming
 This is repeated for all the 7 engineering disciplines considered in PCRS.

12
 3. Design and Implementation

3.1. Dataset overview

Fig 2.1
Dataset we worked on has 20865 rows * 5 columns
# Column Non-Null Count Data type
0 subjects 20865 non-null object
1 hobbies 20865 non-null object
2 marks_percentage 20865 non-null float64
3 age 20865 non-null int64
4 career_interest 20865 non-null object
Data types : float64(1), int64(1), object(3)

3.1.1. Features
Features are the input variables that are used to train a model. Features can be categorical, continuous, or a
combination of both.
Features in the dataset= [‘subjects’, ‘hobbies’, marks_percentage’, ‘age’]

3.1.2. Target
The target is the output variable that the model is trying to predict. The target can be categorical or continuous.
Target variable= [‘career interest’]

13
3.2. Requirements

3.2.1. Software Requirements

1. Python-
Python is a general-purpose programming language that is used for a wide variety of tasks, including:

 Web development: Python is a popular choice for web development, as it is easy to learn and use.
There are many Python frameworks available for web development, such as Django and Flask.
 Data science: Python is a popular choice for data science, as it has a large number of libraries
available for data analysis and machine learning.
 Machine learning: Python is a popular choice for machine learning, as it has a large number of
libraries available for training and evaluating machine learning models.
 Automation: Python is a popular choice for automation, as it can be used to automate tasks that would
otherwise be done manually.
 Scientific computing: Python is a popular choice for scientific computing, as it has a large number of
libraries available for numerical analysis and scientific computing.
Python is a powerful and versatile language that can be used for a wide variety of tasks. It is easy to learn
and use, and it has a large community of users and developers.
 Easy to learn: Python is a relatively easy language to learn, even for beginners. The syntax is simple
and straightforward, and there are many resources available to help you learn Python.
 Versatile: Python can be used for a wide variety of tasks, including web development, data science,
machine learning, automation, and scientific computing.

2. HTML-
HTML, or Hypertext Mark-up Language, is the standard mark-up language for creating web pages. HTML
describes the structure of a web page by using a series of elements.
HTML elements are the building blocks of web pages, and they tell the browser how to display the
content. HTML elements can also have attributes, which provide additional information about the element.
HTML documents are made up of a series of elements, which are nested inside each other:
 The <html> element is the root element of an HTML document, and it contains the entire page.
 The <head> element contains information about the page, such as the title and the Meta tags.
 The <body> element contains the visible content of the page.
HTML is a powerful language that can be used to create a wide variety of web pages. It is easy to learn,
and there are many resources available to help you get started. If you are interested in creating web pages,
HTML is a great place to start.

3. CSS-
CSS, or Cascading Style Sheets, is a style sheet language used for describing the presentation of a
document written in a mark-up language such as HTML or XML. CSS is a cornerstone technology of the
World Wide Web, alongside HTML and JavaScript.
 CSS is designed to enable the separation of content and presentation, including layout, colours, and
fonts.
 CSS allows the separation of presentation from structure.

14
  CSS can define colour, font, text alignment, size, borders, spacing, layout and many other
typographic characteristics, and can do so independently for on-screen and printed views.
CSS is a powerful language that can be used to create a wide variety of web pages.

3.2.2. LIBRARIES REQUIRED

1. Flask-
Flask is lightweight and flexible Python web framework that is designed to make it easy to build web
applications quickly and with minimal code.

Flash provides a simple and intuitive interface for the handling HTTP request and responses, routing, and
creating views and templates.

One of the key benefits of flash is its simplicity and flexibility, it does not impose any particular
architecture structure on one’s application allowing to organise the code and design the application in a
way that makes the most sense for one’s needs.

2. NumPy-
Numpy is a Python library for numerical computing that provides support for large, multidimensional
arrays and matrices, as well as a wide range of mathematical functions to operate on them. Numpy is built
on top of the C programming language and provides fast and efficient array operations that are optimised
for performance.

NumPy is a powerful and flexible library that provides a solid foundation for numerical computing in
Python.

3. Pandas-
Pandas is a Python library that provides high-performance, easy-to-use data structures and data analysis
tools for working with structured and time series data.
 Load data from a variety of sources, including CSV files, JSON files, and SQL databases.
 Manipulate data in a variety of ways, including filtering, sorting, and grouping.
 Analyse data using statistical methods, such as aggregation, correlation, and regression.
 Visualize data using a variety of charts and graphs.
 Export data to a variety of formats, including CSV files, JSON files, and HTML tables.
Pandas is a powerful tool for working with data, and it can be used to solve a wide variety of problems.

4. Matplotlib-
Matplotlib is a Python library for creating static, animated, and interactive visualizations. It is a
comprehensive library that can be used to create a wide variety of plots, including line plots, bar charts,
and scatter plots.

5. LabelEncoder and OrdinalEncoder-

15
 Classes in the scikit-learn library that can be used to convert categorical data into numerical data. This is
useful for machine learning models, which typically only work with numerical data.

3.2.3. ENVIROMENTS USED

1. Jupyter Notebook-

A Jupyter Notebook environment is a way to run Jupyter Notebooks in a isolated environment. This can
be useful for a variety of reasons, such as:

 To keep your Jupyter Notebooks separate from your system Python installation.
 To install specific versions of Python packages that are not available in your system Python
installation.
 To share Jupyter Notebooks with others without having to worry about them installing the same
Python packages on their system.

2. Virtual Studio Code-

Visual Studio Code (VS Code) is a free and open-source code editor developed by Microsoft. It is built on
top of the Electron framework and can run on Windows, macOS, and Linux. VS Code is a popular choice
for web development, but it can also be used for a variety of other programming languages, including
Python, Java, C/C++, and Typescript.
VS Code has a number of features that make it a good choice for developers, including:
 Syntax highlighting and IntelliSense support for a variety of programming languages
 Code completion and refactoring tools
 Built-in debugger
 Support for extensions
 Lightweight and fast

VS Code is a great choice for developers who are looking for a powerful and flexible code editor. It is
easy to use and has a wide range of features that can help you with your development work.

16
3.2.4. Hardware Requirements
RAM: 4 GB
Processor: core i3
Operating System: Window XP and above

17
3.3. PROJECT PIPELINE

• {from the
Data Collection
database}

• {removing any
Data Pre-Processing invalid or null
value}

• {Random
Training the model
Forest
using ML Algorithm
Classifier}

Predict

Fig: 3.2

18
3.4. MODEL VALIDATION

For model training, we have used Random Forest Algorithm.

Model is trained using the dataset file we have and due to a lot of categorical values we apply LabelEncoding to
convert strings to floats.
The classifier is trained using the fit () method by providing the features and the target variable from the dataset.
Once trained, the classifier can make predictions using predict () method based on the user input.

3.4.1. REGRESSION GRAPH

Fig: 3.3

19
3.4.2. CONFUSION METRICS

Fig: 3.4

3.4.3. ACCURACY AND CROSS VALIDATION SCORE

Fig: 3.5

20
3.4.4. ERROR SCORE

Fig: 3.6

21
4. Results and Discussion

Fig:4.1

Fig: 4.2

22
Fig: 4.3

23
 5. Conclusion and Future Work

5.1. Conclusion

In this project, we developed a machine learning-based solution to predict career interests and assist individuals in
making informed decisions about their professional paths.
By leveraging the Random Forest algorithm and a Flask web application, we were able to create a user-friendly
system that predicts career interests based on user input.
Through this project, we aimed to provide individuals with personalized career guidance and enhance their
understanding of the subjects relevant to their predicted careers.

24
5.2. Scope of Improvements

The proposed system can be much improved in the future:

 The UI design need to be more detailed and user friendly.

 Stronger ML algorithm need to be used instead of Random forest algorithm.

 Dataset can be updated with more entries.

 NLP based feedback be provided with the result.

25
 6. REFERENCES

[1] Tanya V Yadalam, Vaishnavi M Gowda, Vanditha Shiva Kumar, Disha Girish "Career
Recommendation Systems Using Content Based Filtering" -ICCES 2020
[2] Vignesh S, Shivani Priyanka C, Shree Manju H, Maithili K " An Intelligent CareerGuidance Systems
Using Machine Learning" - ICACCS 2021
[3] Manar Qamheih, Haya Sammaneh, Mona Nabil Demaidi" PCRS: PersonalisedCareer Path
Recommendation System for Engineering Student"- Under Grant ANNU 1920-SC004 (2020).
[4] Gamage, T. N., & Wijayarathna, C. (2021). An Intelligent Career Guidance System Using Machine
Learning. In 2021 IEEE 17th International Conference on Industrial and Information Systems
(ICIIS) (pp. 67-72). IEEE.
[5] Chakraborty, S., Basu, S., & Chowdhury, S. (2021). An Intelligent Career Recommendation
System Using Hybrid Filtering Techniques. In 2021 IEEE International Conference on Systems,
Man, and Cybernetics (SMC) (pp. 1996-2001). IEEE.
[6] Zhang, Y., & Wen, X. (2020). An Intelligent Career Guidance System Based on Deep Learning. In
2020 5th International Conference on Control and Robotics Engineering (ICCRE) (pp. 129-134).
IEEE.
[7] Jayaweera, A. D. (2020). An Intelligent Career Recommendation System Based on Machine
Learning Techniques. In 2020 IEEE 13th International Conference on Human System Interaction
(HSI) (pp. 455-460). IEEE.
[8] Choudhury, D., Mishra, D. P., & Dandapat, S. (2020). Intelligent Career Guidance System Using
Machine Learning. In 2020 IEEE Region 10 Symposium (TENSYMP) (pp. 266-271). IEEE.

26