THE STUDENT FATIGUE INDEX(SFI) PREDICTION MODEL USING ML

A PROJECT REPORT
SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENT FOR
THE AWARD OF THE DEGREE
OF
BACHELOR OF TECHNOLOGY
IN
ELECTRONICS AND COMMUNICATION ENGINEERING

Submitted by:
Anuj Garg (2K21/EC/040)
Aman Kumar Yadav (2K21/EC/024)
Ankit Kumar (2K21/EC/030)

Under the supervision of

Dr. Sonal Singh

DEPARTMENT OF ELECTRONICS AND COMMUNICATION

ENGINEERING
DELHI TECHNOLOGICAL UNIVERSITY
(Formerly Delhi College of Engineering)
Bawana Road, Delhi-110042
DECEMBER, 2024
 DEPARTMENT OF ELECTRONICS AND COMMUNICATION
ENGINEERING
DELHI TECHNOLOGICAL UNIVERSITY
(Formerly Delhi College of Engineering)
Bawana Road, Delhi-110042

CANDIDATE’S DECLARATION

We, Anuj Garg (2K21/EC/040), Aman Kumar Yadav (2K21/EC/024) and Ankit

Kumar (2K21/EC/030) students of B. Tech (Department of Electronics and

Communication Engineering), hereby declare that the report titled “The Student

Fatigue Index(SFI) Prediction model using ML” which is submitted by us to the

Department of Electronics and Communication Engineering, Delhi Technological

University, Delhi in partial fulfilment of the requirement for the award of the degree

of Bachelor of Technology, is original and not copied from any source without proper

citation. This work has not previously formed the basis for the award of any Degree,

Diploma Associateship, Fellowship or other similar title or recognition.

Place: Delhi ​ Anuj Garg (2K21/EC/040)

Date: 13-12-2024 ​ Aman Kumar Yadav (2K21/EC/024)

Ankit Kumar (2K21/EC/030)

1
 DEPARTMENT OF ELECTRONICS AND COMMUNICATION
ENGINEERING
DELHI TECHNOLOGICAL UNIVERSITY
(Formerly Delhi College of Engineering)
Bawana Road, Delhi-110042

CERTIFICATE

We hereby certify that the Project Dissertation titled "The Student Fatigue

Index(SFI) Prediction model using ML” which is submitted by Anuj Garg

(2K21/EC/040), Aman Kumar Yadav (2K21/EC/024) and Ankit Kumar

(2K21/EC/030) of the Electronics and Communication Engineering, Delhi

Technological University, Delhi in partial fulfilment of the requirement for the award

of the degree of Bachelor of Technology, is a record of the project work carried out

by the students under my supervision. To the best of my knowledge this work has not

been submitted in part or full for any Degree or Diploma to this University or

elsewhere.

Place: Delhi ​ ​ Dr. Sonal Singh

SUPERVISOR
Date: 13-12-2024 ​ (Assistant Professor)
Department of Electronics and
​ Communication Engineering

2
 ABSTRACT

The Student Fatigue Index (SFI) Prediction Model aims to develop a measurable and
dynamic metric that quantifies student fatigue based on key lifestyle factors,
including caffeine intake, sleep duration, physical activity, and screen time. Fatigue
among students is a growing concern due to the demanding nature of academic life,
exacerbated by prolonged screen exposure and irregular schedules. This project
introduces the SFI as a health score, calculated using machine learning techniques, to
help students and educational institutions monitor and manage fatigue effectively.

The SFI will serve as a dynamic indicator, updated daily or weekly based on recent
lifestyle inputs. The goal is to provide a comprehensive understanding of fatigue
patterns, enabling students to identify unhealthy habits and make informed
adjustments. Educational institutions can leverage aggregated SFI data to design
well-being programs and intervention strategies. The SFI score will range on a scale
(e.g., 0–100), with defined thresholds signaling varying levels of fatigue. These
thresholds will be mapped to actionable recommendations for improved lifestyle
habits or the need for institutional support.

The model will employ supervised learning algorithms trained on collected set of
data, correlating lifestyle variables with reported fatigue levels. Feature engineering
will ensure the accurate representation of input parameters, while robust validation
techniques will assess the model’s reliability. By providing interpretable outputs akin
to credit scores, the SFI will empower stakeholders with a user-friendly and practical
tool for promoting student health and productivity.

The novelty of this project lies in its ability to transform subjective experiences of
fatigue into an objective, quantifiable measure. As a scalable and automated system,
the SFI has the potential to enhance student well-being, reduce burnout, and foster a
healthier academic environment.

3
 ACKNOWLEDGEMENT

After doing our project, we were helped by some honourable people through their

guidelines. We take great pleasure after completing this report. I would like to thank

for the help of my mentor Dr. Sonal Singh on this project. Having given us an

excellent guideline to follow in writing throughout many consultations.

Besides, we would like to extend our heartfelt thankfulness to everyone who has

directed and indirectly assisted us in writing this report. On the other hand, many

people including some of our friends have commented and suggested valuable things

that inspired us a lot to improve our work. To various individuals who aided directly

or not directly in making sure that our task was done on time, we are grateful.

Moreover, we wish to acknowledge with gratitude the Department of Electronics and

Communication Engineering at Delhi Technological University, Delhi for allowing

us to work on this subject.

Lastly, special thanks are due to all the participants and contributors who gave their

time and thoughts toward making this research successful.

Anuj Garg ​ Aman Kumar Yadav​ Ankit Kumar

(2K21/EC/040)​ (2K21/EC/024)​ (2K21/EC/030)

4
 CONTENTS

Candidate’s Declaration ………………………………………………………….. 1

Certificate …………………………………………………………………………. 2

Abstract ……………………………………………………………………………. 3

Acknowledgement ……………………………………………………………….... 4

Contents ………………………………………………………………………….... 5

CHAPTER 1 INTRODUCTION ………………………………………………… 6

1.1​Understanding Fatigue in Students ………………………………… 6

1.1.1​ Definition of Student Fatigue…………………………... 6
1.1.2​ Causes of Fatigue in Students………………………….. 6
1.1.3​ Issues caused by Fatigue in Students…………………… 8
1.1.4​ The Concept of the Student Fatigue Index(SFI)………… 9
CHAPTER 2 LITERATURE REVIEW ………………………………………... 11

CHAPTER 3 WORK DONE TILL NOW ………………………………………15

CHAPTER 4 FUTURE WORK……………………………………………….. 20

References …………………………………………………………………............ 22

5
 CHAPTER-1: INTRODUCTION

1.1. Understanding Fatigue in Students

1.1.1 Definition of Student Fatigue

Student fatigue is a widespread issue in academic environments, significantly

affecting the physical and mental well-being of students. Fatigue refers to a state of
physical and mental exhaustion that diminishes one’s ability to perform tasks
efficiently. In students, this condition can arise due to various factors such as long
study hours, inadequate sleep, poor nutrition, stress, and the pressures of academic
performance. Unlike normal tiredness, fatigue can impair cognitive functions,
leading to difficulties in concentration, memory retention, and decision-making.

The impact of student fatigue extends beyond just academic performance. It can
cause increased absenteeism, reduced participation in class, and contribute to mental
health issues such as anxiety and depression. Furthermore, fatigue affects students'
social interactions and their overall quality of life, leading to long-term consequences
if left unaddressed.

Several studies have shown that fatigue in students can be attributed to irregular
sleep patterns, excessive screen time, high workloads, and lifestyle imbalances.
These factors disrupt the body's natural rhythms, further exacerbating the feelings of
tiredness. As the demands on students continue to rise, it is crucial to find ways to
measure and predict fatigue levels accurately to allow timely interventions.

By identifying the root causes and signs of fatigue early, educational institutions and
students can take preventive measures to manage workload effectively, enhance
student performance, and improve overall health. The challenge, however, lies in
developing reliable tools that can assess and predict student fatigue in a data-driven,
objective manner.

1.1.2 Causes of Fatigue in Students

1. Academic Pressure

Long Study Hours: Students frequently dedicate extended hours to prepare for
exams, complete assignments, or work on projects. This continuous focus strains
their mental capacity, leaving little room for relaxation and recovery.

6
Tight Deadlines: Multiple assignments or tasks often overlap, requiring students to
prioritize and manage their time effectively. Constantly working under pressure can
lead to mental fatigue, particularly when combined with other responsibilities.
High Expectations: The desire to meet personal and external expectations, such as
securing good grades or excelling in competitions, adds to the stress, further
exhausting students mentally and emotionally.

2. Poor Sleep Habits

Irregular Sleep Cycles: Staying up late to complete academic tasks or engaging in

recreational activities disrupts natural sleep rhythms. The lack of consistency in sleep
cycles confuses the body’s internal clock, resulting in poor-quality sleep.
Sleep Deprivation: Many students sleep fewer than the recommended 7-8 hours per
night due to their packed schedules. This deprivation reduces the body’s ability to
repair itself and leads to long-term cognitive impairments, such as slower reaction
times and reduced problem-solving skills.

3. Excessive Screen Time

Academic Use: Digital devices are integral to modern education. However,

prolonged exposure to screens for online classes, research, and assignment
completion leads to digital eye strain and mental fatigue.
Recreational Use: Many students spend additional hours on social media, gaming, or
streaming videos, further increasing their screen exposure. This excessive usage
contributes to poor sleep quality and diminished overall energy.

4. Lack of Physical Activity

Sedentary Lifestyle: Students often spend hours sitting during classes, studying, or
engaging in digital activities. This lack of movement decreases energy levels and
reduces the body’s ability to manage stress.
Missed Opportunities for Exercise: Physical activity releases endorphins, which
improve mood and energy. The absence of regular exercise can lead to sluggishness,
low stamina, and mental dullness.

5. Unhealthy Dietary Habits

Caffeine Overuse: Many students rely on caffeinated drinks like coffee or energy
drinks to stay awake. While caffeine provides a temporary boost, overconsumption
leads to energy crashes and increased dependency.

7
Poor Nutrition: Skipping meals or eating unhealthy snacks deprives the body of
essential nutrients, affecting energy levels and overall health. Proper nutrition is
critical to combating fatigue, yet it is often overlooked in busy student schedules.

6. Emotional and Social Stress

Social Isolation: A lack of time for socializing due to academic demands can lead to
loneliness and increased mental fatigue. Social interactions are crucial for mental
rejuvenation and stress relief.
Personal Challenges: Family issues, peer pressure, or anxiety about the future can
weigh heavily on students, contributing to emotional exhaustion. These stressors
often compound other causes of fatigue, making it harder to recover.

1.1.3. Issues Caused by Fatigue in Students

1. Academic Challenges

Difficulty Concentrating: Fatigue reduces focus and makes it harder to absorb and
retain information, which directly affects academic performance.
Slower Cognitive Processing: Mental fatigue slows down thinking and
problem-solving, making complex tasks more time-consuming and frustrating.
Decreased Productivity: Prolonged tiredness often leads to procrastination or
incomplete tasks, causing students to fall behind in their work.

2. Health Consequences

Physical Health Issues: Chronic fatigue can weaken the immune system, making
students more susceptible to illnesses like colds and infections. Other symptoms
include headaches, muscle pain, and digestive problems.
Mental Health Problems: Persistent fatigue is closely linked to conditions like
anxiety, depression, and burnout. Students experiencing these issues may feel
overwhelmed, hopeless, or unmotivated.

3. Quality of Life Impacts

Reduced Social Engagement: Students often withdraw from social activities due to a
lack of energy, which can lead to feelings of loneliness and isolation.
Emotional Frustration: Constant fatigue creates a cycle of dissatisfaction, where
students feel incapable of achieving their goals, further diminishing their confidence
and happiness.

8
1.1.4 The Concept of the Student Fatigue Index (SFI)

1. What is the SFI?

The Student Fatigue Index (SFI) is a measurable score designed to quantify a

student’s fatigue level based on key lifestyle factors. By providing a dynamic health
score, the SFI allows students to track their fatigue over time and understand the
impact of their habits on their well-being.

The SFI bridges the gap between subjective experiences of tiredness and objective
health data, enabling targeted interventions and informed lifestyle changes.

2. Key Features of SFI

Dynamic Updates: The SFI score is updated regularly (daily or weekly) to reflect
changes in lifestyle habits, providing real-time insights.

Personalized Insights: Each score comes with tailored recommendations to improve

well-being, such as sleep adjustments or reduced screen time.

Actionable Thresholds: Scores are categorized into ranges (e.g., low, moderate, high
fatigue), making it easy to identify when action is needed.

3. Data Collection for SFI

The SFI relies on self-reported data collected from students using surveys or mobile
applications. Key inputs include:

Sleep Duration: Average number of hours slept per night.

Screen Time: Total hours spent on devices for academic and recreational purposes.
Caffeine Intake: Amount of caffeinated beverages consumed daily.
Physical Activity: Frequency and duration of exercise or physical activity.

9
4. Benefits of the SFI Model

For Students

Improved Awareness: Students can identify how their habits impact their energy
levels and make necessary adjustments.

Actionable Guidance: Tailored recommendations help students take control of their

health and prevent burnout.

For Educational Institutions

Well-Being Insights: Aggregated SFI scores provide a snapshot of student health,

enabling institutions to implement targeted well-being programs.

Preventative Support: Early identification of high-fatigue students allows for timely

interventions.

For Long-Term Impact

Encourages healthier lifestyle choices, reducing the risk of chronic health issues.
Promotes a culture of self-care and mental health awareness in academic settings.

10
 CHAPTER -2: LITERATURE REVIEW

We synthesize the findings of 5 studies addressing student fatigue, focusing on key

factors affecting fatigue and exploring predictive approaches using machine learning
models to identify and manage fatigue effectively. These findings are mentioned as
follows

1. Lockdown fatigue among college students during the COVID‐19 pandemic:

Predictive role of personal resilience, coping behaviors, and health

Leodoro J. Labrague and Cherry Ann Ballad in their paper published in National
Library of Medicine explores the growing concern of college student’s fatigue during
the permanent lockdown period of COVID-19, which is linked to poor academic
performance, physical health issues, and psychological stress. The study highlights
various contributing factors, such as irregular sleep patterns, excessive screen time,
and academic pressure, which disrupt students' well-being.Their DataSet mainly used
pre-defined 10 statements for evaluating the SFI. The 10‐item scale was answered by
the participants on a five‐point Likert‐type scale that ranged from 1 (never) to 5
(always). The Authors used the multiple Linear Regression Model to evaluate the
SFI. The authors suggest that understanding these factors and utilizing predictive
models could help in early identification of fatigue and better management.

Furthermore, the study emphasizes the potential of integrating technology, such as

wearable devices, and machine learning algorithms to predict fatigue levels
accurately. This could help institutions intervene in a timely manner, offering
students a more balanced approach to academic life.

2. Sleep Quality and Fatigue during Exam Periods in University Students:

Prevalence and Associated Factors

11
Fatigue among university students is a prevalent issue exacerbated by academic
pressures, poor sleep quality, and unhealthy lifestyle choices. During high-stress
periods, such as examinations, students often experience disrupted sleep patterns,
reduced physical activity, and increased reliance on stimulants like caffeine, all
contributing to heightened fatigue levels. Studies using tools like the Pittsburgh
Sleep Quality Index (PSQI) and Fatigue Severity Scale (FSS) have highlighted the
strong correlation between poor sleep quality and fatigue, which negatively impacts
concentration, memory retention, and overall academic performance. Factors such as
physical inactivity, emotional stress, and depressive symptoms further amplify the
problem, making it essential for universities to implement strategies promoting sleep
hygiene, regular physical activity, and stress management. Innovative solutions like
the Student Fatigue Index (SFI), which uses real-time data to quantify fatigue, offer a
promising approach to monitoring and addressing this critical issue effectively.

3. The effect of exercise on academic fatigue and sleep quality among university
students

This research article by Sophie M. M. C. de Lima, José C. S. dos Santos, and Daniel
L. S. da Silva, published in Frontiers in Psychology (2022), investigates the critical
relationship between physical activity, academic fatigue, and sleep quality in
university students. The authors highlight that regular physical exercise not only
reduces fatigue but also enhances sleep quality, which is essential for cognitive
functioning and academic performance.

The study emphasizes that academic fatigue, often characterized by physical and
mental exhaustion, is a growing issue among students, primarily due to intense
academic demands and poor lifestyle habits. The research findings suggest that
incorporating exercise into daily routines helps mitigate these adverse effects by
improving overall physical health, reducing stress levels, and enhancing sleep
patterns.

This study contributes to existing literature by providing actionable insights into how
lifestyle changes like physical activity can combat fatigue, promoting better

12
academic outcomes and well-being. The article also underscores the importance of
further research on integrating such interventions into educational and healthcare
systems to address academic fatigue effectively.

4. An Explainable Student Fatigue Monitoring Module with Joint Facial

Representation

The article authored by Mohammed Naeem Khan et al., was published in the journal
Sensors by MDPI. This research proposes an innovative fatigue monitoring module
leveraging advanced deep learning methods and explainable artificial intelligence
(XAI). The system primarily uses joint facial representation to analyze fatigue levels,
enabling accurate, real-time monitoring.

A distinguishing feature of this module is its focus on explainability, achieved

through XAI techniques that provide insights into how fatigue predictions are made.
This addresses a significant limitation of conventional AI systems, which are often
criticized for being black-box models with limited interpretability. By ensuring both
accuracy and transparency, the proposed system enhances user trust and applicability
in real-world settings.

The module also incorporates multimodal data, such as variations in facial

expressions, to ensure adaptability and robustness across diverse environments. Its
potential applications are extensive, ranging from improving student well-being in
educational institutions to monitoring fatigue in workplaces. The research
underscores the importance of timely detection and intervention, aiming to reduce
the adverse impacts of fatigue on academic and professional performance.

This study contributes significantly to the domain of student fatigue prediction by

integrating state-of-the-art AI techniques with practical implementation
considerations.

13
5. Tracking Sleep Times to Reduce Tiredness and Improve Sleep in College
Students

The article authored by Daniel J. Prestwich, Linda L. Rankin, and Jeff Housman,
published in the Californian Journal of Health Promotion (2007), explores the impact
of sleep health education and sleep tracking on college students. The study aimed to
assess whether these interventions could improve sleep duration and reduce daytime
sleepiness. Participants were assigned to different groups, including one receiving
both sleep education and a sleep log. The study found that students in this group
increased their sleep by over 50 minutes per night.

This research highlights the effectiveness of combining educational strategies with

behavioral tracking to improve sleep habits in students. Despite some limitations,
such as self-reported data and a small sample size, the study provides valuable
insights into the potential benefits of these interventions in addressing sleep
deprivation among college students. The findings suggest that incorporating sleep
education and monitoring into student health programs could lead to improved sleep
health, ultimately enhancing academic performance and overall well-being.

The reviewed studies collectively enhance the understanding of student fatigue,

focusing on its contributing factors and the use of machine learning for predictive
modeling. While each study offers a unique perspective, they collectively underscore
the importance of data-driven approaches to improve fatigue detection and
management. Future research may refine predictive methodologies and address
existing challenges, paving the way for practical solutions to mitigate fatigue and
promote student well-being.

14
 CHAPTER -3: WORK DONE TILL NOW

Data Collection

We are currently in the process of collecting data from students across different
academic years to build the Student Fatigue Index (SFI) Prediction Model. This
data collection focuses on capturing detailed information about students'
demographics, academic routines, and lifestyle habits, which influence fatigue levels.
By gathering responses directly from students, the dataset will reflect real-world
scenarios and diverse perspectives. Below are the key parameters being collected:

1. Demographic Information

1.1 Name: To uniquely identify each respondent for tracking and analysis purposes.​
1.2 Age: To understand variations in fatigue across different age groups, as physical
and mental energy levels can vary with age.​
1.3 Gender: To explore potential differences in fatigue patterns and contributing
factors among different genders.​
1.4 Year of Study: To assess how academic year impacts fatigue levels, recognizing
that first-year students may face challenges adapting to college life, while final-year
students might experience stress related to projects and placements.

2. Academic and Lifestyle Factors

2.1 Study Hours: To quantify the daily time students dedicate to academic activities
and how excessive study hours contribute to fatigue.​
2.2 Sleep Duration: To analyze the relationship between sleep quality, sleep
patterns, and fatigue, as insufficient or inconsistent sleep is a known contributor to
mental and physical exhaustion.​
2.3 Physical Activity: To measure the amount of daily exercise or active routines
and their role in mitigating or exacerbating fatigue levels.​
2.4 Screen Time: To evaluate the effect of prolonged exposure to digital
screens—whether for academic purposes or recreation—on eye strain, cognitive
fatigue, and overall well-being.​

15
2.5 Caffeine Intake: To explore the reliance on caffeine (e.g., coffee, tea, or energy
drinks) as a stimulant to combat fatigue and its long-term implications on energy
levels.

This data collection process is pivotal for building a robust dataset that will be
cleaned and preprocessed to ensure accuracy and reliability. Once the dataset is
finalized, it will serve as the foundation for training machine learning models to
predict the Student Fatigue Index (SFI). This innovative model will provide a
dynamic health metric to quantify fatigue levels and offer actionable
recommendations for students and educational institutions to enhance well-being and
productivity.

We have received 400+ responses as of now.

Fig- Snapshots of the form circulated to collect data

16
Analysis of some of the fields of data collected

17
Challenges Faced During Data Collection

Collecting data for the Student Fatigue Index (SFI) Prediction Model posed
several challenges due to the busy schedules and varying priorities of the students
involved. These challenges affected participation rates and the quality of responses,
necessitating additional efforts to ensure the dataset's reliability. Below are the key
challenges faced during this process:

1. Lack of Student Participation

1.1 Preoccupation with Exams: Many students were busy preparing for or
appearing in exams, making it difficult for them to prioritize completing the survey.​
1.2 Placement Season: Final-year students, in particular, were focused on placement
activities, leaving little time for extracurricular tasks such as filling out forms.​
1.3 Low Response Rates: These factors collectively led to delayed or incomplete
responses, requiring frequent reminders and follow-ups to encourage participation.

2. Errors in Form Design

2.1 Confusion Due to Ambiguous Questions: An error in the phrasing of one of the
questions caused confusion among students, leading to misinterpretation.​
2.2 Invalid Data Entries: The ambiguous question resulted in irrelevant or unusable
data in a specific column, reducing the overall quality of the dataset.

3. Recirculation of the Form

3.1 Form Revision and Resending: To correct the error, the form was revised and
redistributed with clear instructions, but this introduced additional complications.​
3.2 Decreased Motivation: Students who had already completed the form during the
first round were less motivated to respond again, impacting response rates in the
second iteration.​
3.3 Data Integration Challenges: Combining data from the first and second
iterations added complexity to the data cleaning and preprocessing stages.

4. Lessons Learned

4.1 Thorough Form Review: Ensure that all survey questions are reviewed for
clarity and tested with a small pilot group before distribution.​

18
4.2 Timing of Data Collection: Avoid distributing the survey during peak academic
periods such as exams or placement activities.​
4.3 Effective Communication: Provide clear instructions and highlight the
importance of participation to motivate students to respond promptly and accurately.

By addressing these challenges and implementing corrective measures, we are

currently in the process of collecting a more reliable dataset . These experiences
emphasize the importance of careful planning, flexibility, and strong communication
in the data collection process.

19
 CHAPTER – 4: FUTURE WORK

The next phase of this project focuses on building the Student Fatigue Index (SFI)
Prediction Model, leveraging the collected dataset to design a robust and dynamic
system for quantifying fatigue among students. This phase will involve several key
steps to ensure that the model is accurate, interpretable, and scalable for real-world
application. Below are the primary tasks that will be undertaken in the development
of the SFI model:

1. Data Cleaning and Preprocessing

1.1 Handling Missing Data: Impute or remove missing values in critical columns
like sleep duration, screen time, and caffeine intake to ensure the dataset is complete
and reliable.​
1.2 Standardizing Inputs: Convert inconsistent entries (e.g., different units or
ranges in screen time and caffeine intake) into a standardized numeric format for
analysis.​
1.3 Feature Engineering: Create meaningful variables, such as a cumulative score
for physical activity or stress levels, to improve the model’s predictive capability.

2. Exploratory Data Analysis (EDA)

2.1 Identifying Patterns: Analyze the relationships between variables like sleep,
study hours, and screen time to understand their collective impact on fatigue.​
2.2 Correlation Analysis: Identify highly correlated features to reduce redundancy
and focus on the most impactful variables.​
2.3 Visualization: Use graphical techniques to convey insights and trends, such as
how fatigue varies across academic years or gender.

3. Model Development

3.1 Model Selection: Experiment with various supervised machine learning

algorithms, such as linear regression, decision trees, and random forests, to find the
best-fit model for predicting fatigue.​
3.2 Parameter Tuning: Optimize hyperparameters to enhance the model’s
performance and ensure accurate predictions.​
3.3 SFI Score Calculation: Define a systematic approach to calculate the Student

20
Fatigue Index, normalizing the output on a scale (e.g., 0-100) with clear thresholds
for low, moderate, and high fatigue levels.

4. Ethical and Privacy Considerations

4.1 Data Anonymization: Ensure all collected data is anonymized to protect student
privacy and comply with data protection regulations.​
4.2 Transparency: Clearly communicate how the SFI score is calculated and how
the data is used to build trust among users.​
4.3 Inclusive Design: Validate the model across diverse groups to avoid biases and
ensure fair recommendations.

This future work plan outlines a clear roadmap for the creation of the SFI model,
paving the way for an innovative and impactful tool to monitor and improve student
well-being.

21
 REFERENCES

1.​ Prestwich, Daniel J., Linda L. Rankin, and Jeff Housman. "Tracking sleep
times to reduce tiredness and improve sleep in college students." California
Journal of Health Promotion5.2 (2007): 148-156.

2.​ Li, Xiaomian, et al. "An Explainable Student Fatigue Monitoring Module
with Joint Facial Representation." Sensors23.7 (2023): 3602.

3.​ Labrague LJ, Ballad CA. Lockdown fatigue among college students during
the COVID-19 pandemic: Predictive role of personal resilience, coping
behaviors, and health. Perspect Psychiatr Care. 2021 Oct;57(4):1905-1912.
doi: 10.1111/ppc.12765. Epub 2021 Mar 17. PMID: 33728666; PMCID:
PMC8251079.

4.​ Li, Wenjing, et al. "The effect of exercise on academic fatigue and sleep
quality among university students." Frontiers in Psychology 13 (2022):
1025280.

5.​ Li, X.; Lin, J.; Tian, Z.; Lin, Y. An Explainable Student Fatigue Monitoring
Module with Joint Facial Representation. Sensors 2023, 23, 3602.
https://doi.org/10.3390/s23073602

6.​ Bouloukaki, Izolde, et al. "Sleep quality and fatigue during exam periods in
university students: prevalence and associated factors." Healthcare. Vol. 11.
No. 17. MDPI, 2023.

22