DATASPIRE

SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS OF MINI

PROJECT 1A

SEMESTER VI

IN

INFORMATION TECHNOLOGY

BY

Himanshu Tiwari

Nithin Nayak

Chandan S Rajpurohit

UNDER THE GUIDANCE OF

Dr. Chhaya Dhavale

(Assistant Professor, Department of Information Technology)

INFORMATION TECHNOLOGY DEPARTMENT XAVIER

INSTITUTE OF ENGINEERING UNIVERSITY OF

MUMBAI

2024-2025
 CERTIFICATE

This to certify that

HIMANSHU TIWARI (202203059)

NITHIN NAYAK (202203045)
CHANDAN S RAJPUROHIT (202203050)

Have satisfactorily carried out the MINI-PROJECT work titled “DATASPIRE ”

in partial fulfillment of the degree of Bachelor of Engineering as laid down by the
University of Mumbai during the academic year 2024-2025.

Internal Examiner/Guide External Examiner

Date:
Place: MAHIM, MUMBAI
 DECLARATION

I declare that this written submission represents my ideas in my own words and where others’
Ideas or words have been included, I have adequately cited and referenced the original
sources.

I also declare that I have adhered to all the principles of academic honesty and integrity and
have not misrepresented or fabricated or falsified any idea/data/fact/source in my submission.

I understand that any violation of the above will be cause for disciplinary action by the Institute
and can also evoke penal action from the sources which thus have not been properly cited or
from whom proper permission have not been taken when needed.

HIMANSHU TIWARI (202203059) -------------------------------

NITHIN NAYAK (202203045) -------------------------------

CHANDAN S RAJPUROHIT (202203050) -------------------------------

Date:
 TABLE OF CONTENTS

SR. TOPIC PAGE

NO NO.

A. LIST OF FIGURES i

B. LIST OF TABLES ii

C. ABSTRACT iii

D. ACKNOWLEDGEMENT iv

E. COURSE OUTCOMES & CO-PO-PSO MAPPING v

INTRODUCTION 1
 Objectives
1 2
 Scope and importance of the project
 Background 3

2 Literature Survey 4
 Introduction
 Summary of Reviewed Papers
 Key Observations
 Literature Conclusion

3 Methodology
 Introduction
 System Architecture / Conceptual Flow
Diagram
 Algorithm Used
 Hardware and Software Requirements

4 Results and Discussion

 Introduction
 Results
 Performance Analysis
 Discussion
 Conclusion and Future work

5 Bibliography 7
 LIST OF FIGURES

SR NO. FIGURE CAPTION PAGE NO.

i
 LIST OF TABLES

SR NO. TABLE TITLE PAGE NO.

ii
 ABSTRACT

Educational institutions face significant challenges in student performance tracking, attendance

monitoring, and faculty management due to the limitations of traditional ERP systems. Existing
solutions lack predictive insights, real-time analytics, and personalized dashboards, making it difficult
for students and faculty to make data-driven decisions.

To address these shortcomings, our project introduces a Machine Learning-powered Enhanced ERP
System that leverages advanced analytics to optimize academic performance tracking and faculty
operations.

By integrating Machine Learning models such as Random Forest and Isolation Forest, the system
predicts academic risks, identifies weak subjects, and assesses students’ placement readiness.
Personalized dashboards provide instant insights into student progress, helping faculty intervene early
to support at-risk students. Additionally, automated attendance analysis enhances efficiency by
detecting patterns and generating real-time alerts, reducing the need for manual tracking.

This intelligent ERP solution transforms academic management by integrating predictive analytics,
automation, and real-time decision-making. It empowers students with personalized insights, assists
faculty in making proactive interventions, and enhances institutional efficiency, ultimately paving the
way for a more data-driven and effective educational ecosystem.
.

iii
 Acknowledgement

We would like to thank Fr. Dr. John Rose S.J (Director of XIE) for providing us with such an
environment so as to achieve goals of our project and supporting us constantly.

We express our sincere gratitude to our Honorable Principal Dr. Y.D.Venkatesh for
encouragement and facilities provided to us.

We would like to place on record our deep sense of gratitude to Dr. Jaychand Upadhyay, Head
of Dept Of Information Technology, Xavier Institute of Engineering, Mahim, Mumbai, for her
generous guidance help and useful suggestions.

With deep sense of gratitude we acknowledge the guidance of our project guide Guide Name.
The time-to-time assistance and encouragement by her has played an important role in the
development of our project.

We would also like to thank our entire Information Technology staff who have willingly
cooperated with us in resolving our queries and providing us all the required facilities on time.

Himanshu Tiwari -----------------------------

Nithin Nayak -----------------------------

Chandan S Rajpurohit -----------------------------

iv
 COURSE OUTCOMES

CO No. COURSE OUTCOME

ITM 601.1 Identify problems based on societal /research needs.

ITM 601.2 Apply Knowledge and skill to solve societal problems in a group.

ITM 601.3 Demonstrate project management principles and develop interpersonal skills to
work as member of a group or leader.
ITM 601.4 Draw the proper inferences from available results through theoretical/
experimental/simulations and excel in written and oral communication.
ITM 601.5 Identify problems and analyze the impact of solutions in societal and environmental
context for sustainable development.
ITM 601.6 Use standard norms of engineering practices to Demonstrate capabilities of self-
learning in a group, which leads to life-long learning.

CO-PO-PSO MAPPING

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

ITM
601.1 3 3 3 3 3 3 3 3 3 3 3 3 3 3

ITM 3 3 3 3 3 3 3 3 3 3 3 3 3 3
601.2
ITM 3 3 3 3 3 3 3 3 3 3 3 3 3 3
601.3
ITM 3 3 3 3 3 3 3 3 3 3 3 3 3 3
601.4
ITM 3 3 3 3 3 3 3 3 3 3 3 3 3 3
601.5
ITM 3 3 3 3 3 3 3 3 3 3 3 3 3 3
601.6
Chapter 1: Introduction
1.1 Introduction
In today's educational institutions, existing ERP systems face several challenges in
efficiently managing student performance and faculty operations. Some key issues include:
 Inefficient student performance tracking
 Lack of insightful dashboards
 No quick insights for quizzes
 No personalized academic tracking
 Absence of automated attendance analysis
To address these challenges, our project introduces an Enhanced ERP System powered by
Machine Learning (ML). This next-generation ERP system is designed to streamline student
management, improve academic tracking, and provide insights for both students and faculty.

1.2 Objectives
Our project aims to develop a next-generation ERP system integrated with Machine
Learning (ML) to enhance student management and academic tracking in educational
institutions.

Key Objectives:
o Enhance Student Performance Tracking – Provide real-time insights into
academic progress.
o Develop Dashboards – Offer data-driven analytics for students and faculty.
o Automate Attendance Analysis – Predict and track attendance trends using ML.
o Improve UI & User Experience – Create a responsive and interactive interface.
o Enable Quick Quiz Insights – Generate instant analysis of quiz performance.
o Provide Personalized Student Dashboards – Offer tailored insights for
academic growth.
o Predict Academic Risks – Identify defaulters, weak subjects.
o Assess Placement Readiness – Analyze students' preparation for job
placements.

Expected Outcome:
By achieving these objectives, our smart and efficient ERP system will:
o Optimize student success through personalized insights.
o Streamline faculty management with predictive analytics.
o Make education more data-driven and insightful.
o

1.3 Scope of the Project

The ML-powered Enhanced ERP System is designed to improve student and faculty
management in educational institutions. Below is a detailed breakdown of the project scope:
What the Project Will Deliver:
 A Smart ERP System integrated with Machine Learning (ML) for advanced
academic tracking.
 Predictive Analytics for:
o Defaulter detection.
o Weak subject identification.
o Placement readiness assessment.
 Automated Attendance Analysis for streamlined management.
 Personalized Dashboards providing quick insights on:
o Quizzes and assessments.
o Academic progress.
o Performance tracking.

Stakeholders of this project:

 Students:
o To track academic performance and receive actionable insights for
improvement.
 Faculty/Teachers:
o To monitor student progress and take timely, data-driven actions.

Limitations & Exclusions:

 The project focuses exclusively on academic and performance-related data
(excluding finance, payroll, or other non-academic functions).
 The accuracy of ML predictions depends on the quality and availability of data.

Impact of the Project:

 Enhances student success through early intervention strategies and
personalized insights.
 Optimizes faculty workload by automating repetitive tasks like attendance
analysis and performance tracking.
 Improves decision-making for administrators with data-driven insights and
predictive analytics.
 Creates an efficient and modern ERP system tailored to the needs of
educational institutions.

This project ensures the development of an intelligent, efficient, and data-driven ERP system
that transforms student performance tracking and faculty management, paving the way for a more
effective and insightful educational ecosystem.

1.4 Background
Enterprise Resource Planning (ERP) systems play a crucial role in managing academic and
administrative processes in educational institutions. Traditional ERP systems focus on record-
keeping and process automation but often lack advanced data analytics and real-time decision-
making capabilities. These limitations create inefficiencies in student
 performance tracking, academic monitoring, and faculty management.
Existing Systems and Limitations
Most educational institutions rely on traditional ERP systems that primarily handle:
 Student record management
 Attendance tracking
 Examination and grading systems
 Fee and finance management
 Timetable scheduling
However, these systems face several challenges:
1. Inefficient Student Performance Tracking – Most ERP systems lack real-time
analytics to provide insights into student academic progress.
2. Absence of Predictive Insights – Traditional ERPs do not use machine learning (ML) to
forecast student performance, identify weak subjects, or predict academic risks like backlogs
(KTs).
3. Lack of Personalized Dashboards – Students and faculty have limited access to
personalized academic insights, making academic decision-making less effective.
4. No Automated Attendance Analysis – While ERP systems track attendance, they do not
analyse attendance patterns to predict academic performance risks.

Need for a Machine Learning-Powered ERP System

To overcome these challenges, integrating Machine Learning (ML) into ERP systems can
significantly enhance data-driven decision-making. ML-powered ERP systems offer:
 Defaulter Prediction – Identifying students at risk of failing or underperforming based on
attendance, grades, and engagement.
 Weak Subject Prediction – Analyzing student performance trends to highlight subjects where
students need additional support.
 Placement Readiness Prediction – Assessing students' academic records, quiz scores,
and aptitude test performance to gauge their employability.

Summary
The integration of Machine Learning in ERP systems presents a significant opportunity to
enhance student performance tracking, optimize faculty workload distribution, and improve
overall academic management. By leveraging predictive analytics, institutions can proactively
address academic challenges and enhance educational outcomes. This project aims to build a
next-generation ML-powered ERP system that provides intelligent insights, real-time
notifications, and predictive analysis for better student and faculty management.
Chapter 2: Literature Survey
2.1 Introduction
A literature survey is an essential part of any research project as it provides insights into existing
studies, technologies, and methodologies related to the topic. For our project on Machine
Learning-powered ERP systems, reviewing previous research helps us understand the current
limitations of traditional ERP systems in educational institutions and explore how ML can
enhance their efficiency.
By analyzing previous studies, we can:
 Identify challenges in traditional ERP systems, such as limited predictive analytics,
inefficient student performance tracking, and lack of real-time insights.
 Evaluate existing ERP models and their integration with AI/ML to enhance
decision-making and automation.
 Identify research gaps, such as the absence of personalized dashboards,
predictive academic tracking, and automated attendance analysis.
This literature survey provides a strong foundation for our project by highlighting areas where
ML can improve ERP functionalities.

2.2 Summary of Reviewed Papers

The following table presents a summary of six research papers related to this project:
Table 2.1: Literature Review Summary

3 Table 2.1: Literature Review Summary

Sr.
Paper Title Author(s) Year Methodology Key Findings Limitations
No.
Identifies
challenges in
ERP System in Review of Limited
ERP adoption,
Higher existing ERP discussion on
such as high
1 Education: Ahmad Jamal 2018 implementations solutions and
cost, user
Prospects and in higher lacks empirical
resistance,
Challenges education data
and
integration
issues
Educational
Data Mining: Application of
ML algorithms
Prediction of machine learning Requires large
(e.g., Random
Students' models to predict datasets for
Mehmet Forests, SVM)
2 Academic 2022 student higher
Yağcı effectively
Performance performance accuracy; lacks
predict student
Using based on midterm real-time data
performance
Machine grades analysis
Learning
Algorithms
Yash
Suggests an
Kothawade, Proposal of an
integrated ERP
ERP Model for Kartik ERP framework No real-world
May model to
3 Education Parmar, tailored for implementation
2021 streamline
System Abhishek educational details provided
administration
Gupta, institutions
 Prof.
and academics
Jyotsna More
Sr.
Paper Title Author(s) Year Methodology Key Findings Limitations
No.
ML enhances
Machine
Zainab ERP
Learning-Driven Comprehensive Lacks case
Nadhim efficiency by
Optimization of analysis of ML studies on
4 Jawad, 2024 automating
ERP Systems: A applications in practical
Villányi processes and
Comprehensive ERP optimization implementation
Balázs improving
Review
decision-
making
3.1 Key Observations
The literature survey provided valuable insights into the existing research on ERP systems in education
and the role of Machine Learning (ML) in academic performance prediction. The key takeaways from
the reviewed studies are summarized below:
Common Methodologies in Previous Research
 ERP System Analysis: Studies reviewed ERP adoption in educational institutions, highlighting
challenges such as high costs, system integration issues, and resistance to change.
 Machine Learning in Student Performance Prediction: Various ML algorithms like Decision
Trees, Support Vector Machines (SVM), and Random Forests were applied to student
datasets to predict academic performance.
 Framework Proposals: Some studies proposed ERP frameworks tailored for education but
lacked real-world implementation details.
 ML-Driven ERP Optimization: Research explored the potential of ML in enhancing ERP
functionalities, focusing on process automation and decision-making improvements.
Research Gaps Identified
Despite significant advancements in ERP and ML applications, several gaps remain in existing studies:
 Limited Integration of ML in ERP Systems: Most ERP systems lack predictive
analytics for student performance tracking, defaulter prediction, and placement
readiness assessment.
 Absence of Personalized Dashboards: Existing ERPs provide generic academic reports
but do not offer personalized insights for students and faculty.
 Lack of Automated Attendance Analysis: While attendance tracking exists, its impact on
academic performance is not analyzed using ML models.
 Delayed or No Real-Time Notifications: Most ERPs fail to provide instant updates on
academic progress, attendance shortfalls, or upcoming deadlines.
How Our Project Builds on Previous Studies
Our ML-powered ERP system addresses these research gaps by:
 Integrating Predictive Analytics to forecast weak subjects, defaulters, etc
 Providing Personalized Dashboards for students and faculty, offering real-time
attendance insights into academic performance (Placement).
 Enhancing Attendance Analysis by identifying attendance patterns and correlating them
with student success rates.
 Offering Real-Time Notifications for students and faculty regarding academic
progress, deadlines, and attendance warnings.
By incorporating advanced ML techniques and real-time analytics, our project goes beyond traditional
ERP functionalities to create a data-driven, intelligent academic management system that optimizes
student success and faculty efficiency.
3.2 Literature Conclusion
The literature survey played a crucial role in shaping our methodology and approach for developing a
Machine Learning-powered ERP system. By analyzing previous studies, we identified limitations in
existing ERP systems and leveraged insights to design a more intelligent, efficient, and student-centric
solution.

Key Influences on Our Approach:

1. Adoption of Machine Learning for Predictive Analytics

o Previous research demonstrated the effectiveness of Decision Trees, SVM, and

Random Forest in predicting student performance.

o This influenced our decision to incorporate ML-based prediction models for

defaulter identification, weak subject prediction, and placement readiness
assessment.

2. Implementation of Personalized Dashboards

o Existing ERP systems lacked real-time academic tracking and personalized

insights.

o Our project incorporates individualized dashboards for students and faculty,

offering AI-driven academic progress tracking.

3. Enhancement of Attendance Monitoring

o Studies highlighted attendance tracking issues but lacked automated analysis

of attendance trends.

o We decided to integrate ML-based attendance pattern analysis, correlating

attendance with academic performance (weak subjects).

By addressing critical gaps in ERP functionality and utilizing ML-driven insights, our methodology
ensures that the system is more adaptive, efficient, and data-driven, ultimately enhancing student
performance tracking and faculty management.
 Chapter 3: Methodology
3.1 Introduction
This chapter provides an overview of the methodology used in developing our ML-powered College
ERP system. Our approach is designed to address the limitations of existing ERP systems by
integrating Machine Learning for better student performance tracking, academic insights, and
faculty management.
The methodology is based on findings from the literature survey, which highlighted issues such as lack
of predictive analytics, absence of real-time notifications, and inefficient dashboards. To overcome
these limitations, we have designed an enhanced ERP system with the following features:
 Automated Attendance Tracking with real-time notifications.
 AI-powered Dashboards for student and faculty insights.
 Web-based application, accessible on all devices.
 ML-based student performance analysis, including defaulter prediction, weak
subject identification, KT prediction, and placement readiness assessment.
The following sections detail the ML models, algorithms, and data processing techniques
used in the project.

3.2 System Architecture / Conceptual Flow Diagram

The proposed ML-powered College ERP system is designed to provide automated student
performance tracking, predictive analytics, and enhanced faculty management. It integrates
machine learning algorithms with an interactive web-based dashboard to streamline administrative
and academic processes.
The working of the system can be understood through the conceptual flow diagram (Figure 3.1), which
illustrates the interaction between different system components.

Figure 3.1: Block Diagram / Conceptual Flow of the System

Components of the System
1. User Interface (Web Application & Dashboard)
 A responsive web-based ERP system accessible to students, faculty, and
administrators.
 Provides personalized dashboards for students and faculty.
 Displays academic insights, notifications, attendance reports, and predictive
analytics.
2. Student & Faculty Database
 Stores academic records, attendance data, quiz results, and student history.
 Data is securely stored and updated in real-time.
 Used for training machine learning models.
3. Attendance Monitoring & Notification System
 Automated attendance tracking using manual entry or biometric integration.
 Machine Learning-based attendance pattern analysis predicts defaulters.
 Real-time notifications are sent to students & faculty in case of low attendance.
4. Machine Learning Models for Student Performance Analysis
The following ML models are implemented for different predictive analytics tasks:
(a) Defaulter Prediction (Random Forest Classifier & Isolation Forest Algorithm)
Inputs:
 Semester Number
 Total Classes Held & Attended
 Attendance % (Current & Previous Semester)
 Drop in Attendance %
 CGPA & Backlogs (KTs)
Functionality:
 Predicts students at risk of falling below minimum attendance criteria.
 Sends early alerts to students & faculty for corrective action.
(b) Weak Subject Identification (Random Forest Classifier)
Inputs:
 Semester Number
 Quiz Marks, IAT Marks
 Subject-wise Attendance %
 Final Marks
Functionality:
 Identifies subjects in which a student is underperforming.
 Suggests areas for improvement based on historical trends.
(c) KT (Backlog) Prediction (Random Forest Classifier)
Inputs:
 Semester Number
 Quiz Marks, IAT Marks, Final Exam Marks
  Attendance %
 CGPA History
 Quiz Trend
Functionality:
 Predicts probability of a student failing a subject.
 Helps faculty intervene early with additional support.
(d) Placement Readiness Prediction (Random Forest Classifier)
Inputs:
 Student_ID
 Aptitude, Verbal, Logical, and Technical Scores
 Overall CGPA
 Backlogs (KTs)
Functionality:
 Evaluates student readiness for placements based on academic & aptitude
performance.
 Helps in career guidance and personalized recommendations.
5. Data Processing & ML Model Training
 Pandas & NumPy are used for data preprocessing.
 Scikit-Learn, TensorFlow, and PyTorch are used for training ML models.
 Matplotlib & Seaborn generate insights and visualizations.
6. Real-time Insights & Notifications
 AI-powered dashboards display trends, predictions, and academic alerts.
 Students receive personalized recommendations for academic improvement.
 Faculty can monitor student progress and intervene when needed.

Conclusion
The proposed system enhances traditional ERP functionalities by integrating Machine Learning for real-
time insights, predictive analytics, and automated academic tracking. By implementing this intelligent
ERP system, we aim to optimize student success, improve faculty decision-making, and enhance
institutional efficiency.

3.3 Algorithm Used

Algorithms Implemented
The proposed system integrates Machine Learning (ML) models to predict student performance, defaulters,
backlogs (KTs), and placement readiness. The primary algorithm used is the Random Forest Classifier,
along with the Isolation Forest Algorithm for anomaly detection.
1. Random Forest Classifier
Random Forest is a supervised learning algorithm that builds multiple decision trees and merges them for a
more accurate and stable prediction. It is used for:
 Defaulter Prediction
 Weak Subjects Prediction
 Placement Readiness Analysis
2. Isolation Forest Algorithm
The Isolation Forest Algorithm is an unsupervised learning model used to detect outliers
(students at high risk of failure or dropout).
Step-by-Step Explanation
1. Start
2. Input Student Attendance & Performance Data
o Attendance % per subject
o Drop in attendance compared to previous semester
o CGPA drop rate
o Number of backlogs
3. Train the Model
o Create decision trees where anomalies are isolated faster than normal data points
4. Detect Anomalies
o If attendance drop is significant → Flag as high-risk defaulter
o If CGPA is dropping continuously → Flag as academic risk
5. Output the Results
o Generate alerts for faculty & students
6. End

Figure 3.2: Flowchart of the Proposed Algorithm

Random Forest Classifier

Isolation Forest Algorithm

3.4 Hardware and Software Requirements

3.4.1 Hardware Requirements
The following hardware components are required for the successful implementation and execution of the
project:
 Processor: Intel Core i5 (or higher) / AMD Ryzen 5 (or higher)
 RAM: Minimum 8GB (Recommended: 16GB for better performance)
 Storage: Minimum 256GB SSD (Recommended: 512GB SSD or higher)
 Graphics Card: Integrated GPU (Dedicated GPU optional for ML model training)
 Sensors: Not applicable
 Other Peripherals: Standard keyboard, mouse, and monitor
3.4.2 Software Requirements
The software tools, frameworks, and programming languages used in this project are:
 Operating System: Windows 10/11, Ubuntu 20.04+, or macOS
 Programming Language: Python, JavaScript (for web development), Dart (for mobile app)
 Development Environment:
o VS Code – For Python and JavaScript development
o PyCharm – For machine learning model development
o Android Studio / FlutterFlow – For mobile app development
 Libraries/Frameworks:
o Machine Learning: Scikit-Learn, TensorFlow, Pandas, NumPy
o Visualization: Matplotlib, Seaborn
o Backend: Node.js, Express.js
o Frontend: React.js (Web), Flutter (Mobile)
o Database: MongoDB (NoSQL) or MySQL (SQL)
This setup ensures smooth development and execution of the ML-powered ERP System with optimized
performance.
Chapter 4 – Results and Discussion
4.1 Introduction
Provide an overview of the results obtained from the project and their significance.

4.2 Results
● Present the outputs of the project in the form of tables, graphs,
screenshots, or images.
● Label figures properly (e.g., "Figure 4.1: System Output").
● If applicable, compare expected vs. actual results.

4.3 Performance Analysis

● Evaluate system performance using quantitative (accuracy, efficiency, speed)
and
qualitative (usability, reliability) measures.
● If applicable, compare with existing systems.

4.4 Discussion
● Interpret the results and discuss their implications.
● Explain if the project meets the objectives defined in Chapter 1.
● Mention any unexpected findings or challenges encountered.

4.5 Conclusion
Summarize the key findings and their significance.

Bibliography