A Real Time Research Project Report

on
EduPredict
Predictive Model for BTech College Admissions
Based on EAMCET/ECET Rank
Submitted by,

J. VISHNU VARDHAN GUPTHA 22J41A05F7

G. NAVEEN 22J41A05F3
D. NIHAL 22J45A05E7
M. NITHIN 22J41A05H5

in partial fulfilment of the academic requirements II BTech.

of
BACHELOR OF TECHNOLOGY
in

COMPUTER SCIENCE AND ENGINEERING

Under the Guidance of

Ms Vyshnava Divya

Assistant Professor

DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING

MALLA REDDY ENGINEERING COLLEGE(A)
Maisammaguda, Secunderabad, Telangana, India 500100
July -2024
 MALLA REDDY ENGINEERING COLLEGE
Maisammaguda, Secunderabad, Telangana, India 500100

BONAFIDE CERTIFICATE

This is to certify that this Real Time Research Project work entitled
“EduPredict: Predictive Model for BTech College Admissions Based on
EAMCET/ECET Rank”, submitted by J. VISHNU VARDHAN GUPTHA
(22J41A05F7) G. NAVEEN (22J41A05F3) D. NIHAL (22J45A05E7)M. NITHIN
(22J41A05H5) to Malla Reddy Engineering College affiliated to JNTUH,
Hyderabad in academic requirements II BTech for the award of Bachelor of
Technology in Computer Science and Engineering is a bonafide record of
project work carried out under my supervision during the academic year 2023–
2024 and that this work has not been submitted elsewhere for a degree.

SIGNATURE SIGNATURE
Ms Vyshnava Divya Dr. P.S. R. C. Murty
Assistant Professor Head of the Department
Department of CSE Department of CSE

Malla Reddy Engineering College Malla Reddy Engineering College

Secunderabad, 500100

Submitted for Real Time Research Project viva-voce examination held on ______

INTERNAL EXAMINAR EXTERNAL EXAMINER

 ABSTRACT

The College Prediction System is a web-based application designed to help students predict
potential BTech colleges based on their EAMCET/ECET scores, simplifying the admission
process through accurate, data-driven predictions using historical cutoff ranks and machine
learning algorithms. Students can input their scores and category information to receive a list
of likely colleges. The backend, built with Node.js, ensures robust operations, while the React.js
frontend offers a responsive user interface. A relational database stores extensive data needed
for predictions, and the algorithm, implemented in Python using scikit-learn, performs the
analysis. This system addresses the limitations of manual and inaccurate methods by
automating predictions and regularly updating the database with the latest data. User
authentication allows students to save their predictions and track their progress. In conclusion,
the College Prediction System provides a reliable and accessible tool for students to make
informed decisions about their higher education, enhancing their chances of academic and
professional success.
 TABLE OF CONTENTS

ABSTRACT I

CHAPTER DESCRIPTION PAGE NO.

1 INTRODUCTION 1

2 LITERATURE SURVEY 2

3 SYSTEM ANALYSIS 4

3.1 Existing System

3.2 Existing System Disadvantages
3.3 Proposed System
3.4 Proposed System Advantages

4 SYSTEM DESIGN 6

4.1 System Architecture – Data Base Design

5 IMPLEMENTATION 9

5.1 MODULES

6 SYSTEM REQUIREMENTS 12

6.1 Hardware Requirements

6.2 Software Requirements

7 EXPECTED OUTPUT 15

8 CONCLUSION 17

9 REFERENCES 18
 CHAPTER 1
INTRODUCTION

The College Prediction System is a web-based application designed to assist students in

predicting potential BTech colleges based on their EAMCET/ECET scores. The aim of this
project is to simplify the college admission process by providing accurate and personalized
college predictions through the use of historical admission data and machine learning
algorithms.

By inputting their EAMCET/ECET scores and category information, students can receive a list
of colleges where they are likely to gain admission. The system features a user-friendly interface
developed with React.js, ensuring a seamless and responsive user experience. The backend,
built with Node.js and Express.js, supports robust server-side operations, while a relational
database such as MySQL or PostgreSQL stores and manages the necessary data. The prediction
algorithm, implemented in Python, utilizes scikit-learn for precise data analysis.

This documentation provides a comprehensive overview of the College Prediction System,

including its development process, system analysis, design, implementation, and expected
outcomes. The project aims to empower students with accurate information, helping them make
informed decisions about their higher education.
 CHAPTER 2
LITERATURE SURVEY

The college admission process is a pivotal step in a student's academic journey, often involving
a significant amount of research and decision-making. Various tools and methodologies have
been developed to aid students in this process, particularly in predicting college admissions
based on entrance exam scores. This literature survey reviews existing systems and
methodologies relevant to college prediction systems, machine learning algorithms, and web
development technologies used in similar applications.

Existing College Prediction Systems

Several platforms provide college prediction services based on entrance exam scores. Websites
like CollegeDekho, Shiksha, and Careers360 offer tools for predicting college admissions using
historical cutoff data. These tools typically require students to input their scores and other
relevant details to generate a list of potential colleges. However, many of these platforms rely
on simple statistical methods and do not leverage advanced machine learning techniques, which
can limit their accuracy and adaptability.

Machine Learning in Prediction Models

Machine learning has been extensively used in predictive analytics across various domains. In
the context of college admissions, algorithms such as linear regression, decision trees, and
support vector machines (SVM) have shown promising results. Studies have demonstrated that
machine learning models can analyse large datasets of historical admission data to identify
patterns and make accurate predictions. For instance, a study by Kumar et al. (2020) highlighted
the effectiveness of using random forest and neural networks in predicting college admissions
with high accuracy.

Data Collection and Management

The accuracy of a prediction system largely depends on the quality and comprehensiveness of
the data it uses. Historical admission data, including cutoff ranks and scores from previous
years, is crucial for training and validating prediction models. Various sources, such as
government education boards, college websites, and educational portals, provide this data.
Effective data management practices, including data cleaning, normalization, and storage, are
essential to ensure the reliability of the prediction system.
Web Development Technologies

Modern web development technologies offer robust frameworks for building dynamic and
responsive web applications. React.js is widely used for frontend development due to its
component-based architecture and efficient rendering capabilities. For backend development,
Node.js with Express.js provides a scalable and efficient environment for server-side operations
and API integrations. Relational databases like MySQL and PostgreSQL are preferred for
managing structured data, while Python libraries such as scikit-learn facilitate the
implementation of machine learning algorithms.

Gaps and Opportunities

While existing systems provide a foundation for college prediction, there is a significant
opportunity to enhance these systems using advanced machine learning techniques and
comprehensive data management. Many current tools do not offer personalized predictions
based on detailed user profiles or real-time updates with the latest cutoff data. Integrating these
features can significantly improve the accuracy and usability of college prediction systems.
 CHAPTER 3
SYSTEM ANALYSIS

3.1 Existing System

The existing methods for predicting BTech college admissions based on EAMCET/ECET
scores primarily rely on manual calculation methods or simple statistical tools available on
educational websites. Students typically input their scores and category information into these
tools, which then provide a list of colleges along with their previous year's cutoff ranks. These
systems often lack automation and real-time updates, relying instead on static data that may not
reflect current admission trends accurately.

3.2 Existing System Disadvantages

1. Accuracy Issues: Reliance on static cutoff data without considering dynamic changes
in admission criteria.
2. Limited Scope: Lack of personalized predictions based on detailed user profiles or
specific category quotas.
3. User Experience: Often cumbersome interfaces and limited functionality, making it
difficult for users to navigate and interpret results effectively.
4. Dependency on Manual Inputs: High potential for human error in data entry and
calculation processes.
5. Outdated Information: Inability to provide real-time updates on cutoff ranks and
admission trends.

3.3 Proposed System

The proposed College Prediction System aims to overcome these limitations by leveraging
advanced machine learning algorithms and modern web technologies. It will provide an
automated and accurate prediction of potential colleges based on EAMCET/ECET scores. The
system will continuously update its database with the latest admission data, ensuring users have
access to real-time information.
3.4 Proposed System Advantages

1. Accurate Predictions: Utilization of machine learning algorithms to analyze historical

data and predict college admissions with higher accuracy.
2. Real-time Updates: Continuous updating of cutoff ranks and admission trends to
reflect current academic years.
3. User-Friendly Interface: Development of a responsive and intuitive interface using
React.js for easy navigation and understanding of results.
4. Personalized Recommendations: Ability to provide personalized predictions based on
user-specific inputs such as category quotas and preferred locations.
5. Automation: Reduction of manual effort and potential errors through automated data
processing and prediction generation.

By addressing these issues and leveraging advanced technologies, the proposed College
Prediction System aims to provide students with a reliable and user-centric tool to facilitate
informed decision-making in their pursuit of higher education.
 CHAPTER 4
SYSTEM DESIGN

4.1 System Architecture – Database Design

The database design of the College Prediction System plays a crucial role in storing and
managing the extensive data required for accurate predictions of BTech college admissions
based on EAMCET/ECET scores. This section outlines the schema and structure of the
relational database, highlighting its tables, relationships, and attributes.

Database Schema

The database schema is designed to accommodate the following entities and relationships:

Colleges: Stores information about colleges offering BTech programs, including unique
identifiers, names, and locations.

Programs: Represents specific BTech programs offered by colleges, linked to respective

college entities via foreign keys.

Cutoffs: Contains historical cutoff ranks for each program, categorized by academic year and
student category (e.g., General, OBC, SC/ST).

Users: Manages user profiles and authentication credentials for accessing the system.
Example Schema (SQL):

CREATE TABLE colleges (

id INT PRIMARY KEY AUTO_INCREMENT,
name VARCHAR (255) NOT NULL,
location VARCHAR (255) NOT NULL);

CREATE TABLE programs (

id INT PRIMARY KEY AUTO_INCREMENT,
college_id INT,
program_name VARCHAR (255) NOT NULL,
FOREIGN KEY (college_id) REFERENCES colleges(id));

CREATE TABLE cutoffs (

id INT PRIMARY KEY AUTO_INCREMENT,
program_id INT,
year INT,
category VARCHAR (50),
rank_cutoff INT,
FOREIGN KEY (program_id) REFERENCES programs(id));

CREATE TABLE users (

id INT PRIMARY KEY AUTO_INCREMENT,
username VARCHAR (255) NOT NULL,
password_hash VARCHAR (255) NOT NULL,
email VARCHAR (255) NOT NULL);
Explanation:

1. Colleges Table: Stores information about colleges including unique IDs, names, and
locations.
2. Programs Table: Contains details of BTech programs offered by colleges, linked to
colleges through foreign key constraints.
3. Cutoffs Table: Stores historical cutoff ranks for each program, categorized by academic
year and student category.
4. Users Table: Manages user profiles and credentials for authentication and access
control within the system.

Database Management System (DBMS)

The College Prediction System utilizes a relational database management system (RDBMS)
such as PostgreSQL or MySQL. These systems are chosen for their ability to handle structured
data efficiently, provide transactional support, and support complex queries required for
predictive analysis.

Considerations

1. Data Integrity: Ensuring data consistency and integrity through primary key
constraints, foreign key relationships, and proper indexing.
2. Scalability: Designing the database schema to accommodate future growth in data
volume and user base.
3. Performance: Optimizing database queries and operations for efficient data retrieval
and processing.

By implementing this database design, the College Prediction System aims to provide a robust
foundation for storing and managing data critical to accurately predicting college admissions
based on EAMCET/ECET scores, ensuring reliability and scalability as the system evolves.
 CHAPTER 5
IMPLEMENTATION

5.1 Modules

The implementation of the College Prediction System is organized into several modules, each
responsible for specific functionalities to ensure the system operates efficiently and meets user
requirements.

1.User Authentication Module

Description: Manages user registration, login, and authentication processes to secure access to
the system.

Key Features:

• User registration with validation of email and password requirements.

• Secure storage of user credentials using hashing algorithms (e.g., bcrypt).
• Login functionality with session management to maintain user sessions securely.

2.Data Input Module

Description: Facilitates the input of EAMCET/ECET scores and category information by

users.

Key Features:

• User interface for entering exam scores, category preferences, and other relevant details.
• Validation of input data to ensure accuracy and completeness.
• Integration with backend services to store user inputs securely in the database.
3.Prediction Algorithm Module
Description: Implements machine learning algorithms to predict potential colleges based on
user-provided scores.

Key Features:

• Utilization of historical cutoff data and machine learning models (e.g., regression,
decision trees) for prediction.
• Training and fine-tuning of models using Python libraries like scikit-learn.
• Real-time calculation of predicted colleges based on updated data and user inputs.

4.Results Display Module

Description: Displays predicted colleges to users in a clear and comprehensible format.

Key Features:

• User interface design using React.js for responsive and intuitive display of predicted
colleges.
• Sorting and filtering options to refine displayed results based on user preferences (e.g.,
location, program).
• Visualization of predicted colleges and cutoff trends through charts or graphs for
enhanced user understanding.

5.Feedback System Module

Description: Gathers user feedback on predicted results to improve system accuracy and user
satisfaction.

Key Features:

• Integration of feedback forms or rating systems to collect user opinions and suggestions.
• Analysis of feedback data to identify common issues or areas for system enhancement.
• Continuous improvement of prediction algorithms and user interface based on collected
feedback.
6.Administrative Module (Optional)

Description: Provides administrative functionalities for system management and maintenance.

Key Features:

• User management for administrators, including roles and permissions.

• Monitoring and logging of system activities for security and audit purposes.
• Maintenance tasks such as database backups, updates, and performance optimization.
 CHAPTER 6
SYSTEM REQUIREMENTS

6.1 Hardware Requirements

The hardware requirements for deploying and running the College Prediction System are
outlined below:

1. Server Hardware:

• Processor: Dual-core or higher processor (Intel Core i5 or equivalent recommended)

• RAM: Minimum 8 GB (16 GB recommended for better performance)
• Storage: SSD storage recommended for faster data access and processing
• Network: Stable internet connection for hosting the web application and accessing
external resources

2. Client Devices:

Desktop or Laptop:

• Processor: Intel Core i3 or equivalent

• RAM: Minimum 4 GB
• Storage: Sufficient for storing browser cache and local files
• Browser: Compatible with modern browsers (Google Chrome, Mozilla Firefox, Safari,
Microsoft Edge)
6.2 Software Requirements

The software requirements for developing and deploying the College Prediction System include
the following:

1. Development Environment:

• Integrated Development Environment (IDE): Visual Studio Code, IntelliJ IDEA, or

similar
• Version Control: Git for code versioning and collaboration

2. Backend Technologies:

• Node.js: JavaScript runtime environment for server-side development

• Python: Programming language for machine learning model development
• Database: MySQL for storing relational data

3. Frontend Technologies:

• React.js: JavaScript library for building user interfaces, ensuring a responsive and
interactive frontend
• HTML/CSS: Markup and styling languages for designing web pages and user
interfaces
• JavaScript: Programming language for frontend logic and interaction

4. Database Management System (DBMS):

• MySQL: Relational database management systems for storing and managing

structured data efficiently

5. Machine Learning Libraries (if applicable):

• scikit-learn: Python library for machine learning algorithms such as regression and
classification
• TensorFlow or PyTorch: Deep learning frameworks for advanced predictive
modeling (optional)
6. Deployment Tools:

• Docker: Containerization platform for packaging applications and their dependencies

• Kubernetes: Container orchestration tool for managing containerized applications in a
clustered environment (optional for scaling)

7. Other Tools and Libraries:

• npm (Node Package Manager): Package manager for Node.js and frontend
dependencies
• RESTful API design tools: Swagger, Postman for API development and testing
• Security tools: SSL certificates for secure data transmission, authentication
mechanisms for user login and access control
 CHAPTER 7
EXPECTED OUTPUT

The College Prediction System aims to provide the following expected outputs to users based
on their input of EAMCET/ECET scores and other relevant details:

1. Predicted Colleges:

• The system will generate a list of colleges that match the user's input criteria, such as
scores, category (General, OBC, SC/ST), and preferred location.
• Each predicted college will include details such as name, location, available BTech
programs, and historical cutoff ranks for the specified category.

2. Rank Cutoffs:

• Display of predicted rank cutoffs for each college and program based on historical data.
• Users can view the minimum rank required for admission into specific programs at
different colleges based on their input scores.

3. Personalized Recommendations:

• The system will provide personalized recommendations based on the user's caste, rank,
course preferences, gender, and location.
• Recommendations will prioritize colleges that best match the user's profile and
preferences, facilitating informed decision-making.

4. Interactive User Interface:

• A user-friendly interface developed using React.js, ensuring ease of navigation and

clarity in displaying predicted results.
• Interactive elements such as sorting options, filters, and visual representations
(charts/graphs) to enhance user experience and understanding.
5. Real-time Updates:

• Continuous updates of predicted results and rank cutoffs to reflect current academic year
data and changes in admission criteria.
• Users can access the most up-to-date information to make timely decisions regarding
college admissions.

6. Feedback Mechanism:

• Provision for users to provide feedback on predicted results, helping to refine and
improve the accuracy of future predictions.
• Analysis of user feedback to identify trends and areas for system enhancement.

7. Security and Access Control:

• Secure access to predicted results and personal user data through robust authentication
mechanisms and data encryption.
• Implementation of privacy measures to safeguard user information and maintain
confidentiality.

RAW DATA : collected from Official EAMCET/ECET Websites

(sample data collected from official EAMCET website)

 CHAPTER 8
CONCLUSION

The College Prediction System represents a significant advancement in facilitating informed

decision-making for aspiring BTech students based on their EAMCET/ECET scores and
personal preferences. Through the development and implementation of this system, several key
outcomes and insights have been achieved:

1. Enhanced Decision-Making: By leveraging machine learning algorithms and real-time data

updates, the system provides accurate predictions of potential colleges. This empowers students
to make informed choices aligned with their academic performance and preferences.

2. User-Centric Approach: The system prioritizes user experience with a user-friendly

interface and interactive features. These include personalized recommendations, visual
representations of data, and options for feedback, ensuring usability and satisfaction.

3. Automation and Efficiency: Automation of data processing and prediction generation

reduces manual effort and enhances efficiency. This results in timely updates and reliable
predictions that reflect current admission trends and cutoffs.

4. Scalability and Adaptability: Designed with scalability in mind, the system accommodates
future growth in data volume and user base. It integrates seamlessly with modern technologies
and frameworks, ensuring adaptability to evolving educational landscapes.

5. Security and Reliability: Implementation of robust security measures safeguards user data
and ensures secure access to predicted results. This includes encryption protocols,
authentication mechanisms, and adherence to privacy standards.

6. Continuous Improvement: Through user feedback mechanisms and iterative development

cycles, the system is poised for continuous improvement. Insights gathered from user
interactions enable refinement of prediction algorithms and enhancement of overall system
performance.
 CHAPTER 9
REFERENCES

1. Bishop, C. M. (2006).

Pattern Recognition and Machine Learning.

(https://link.springer.com/book/10.1007/978-0-387-45528-0)

2. Kotsiantis, S. B., Zaharakis, I., & Pintelas, P. (2007).

Supervised Machine Learning: A Review of Classification Techniques.

Emerging Artificial Intelligence Applications in Computer Engineering, 160, 3-24.

(https://www.researchgate.net/publication/228061689_Supervised_Machine_Learning_A_Re
view_of_Classification_Techniques)

3. Hastie, T., Tibshirani, R., & Friedman, J. (2009).

The Elements of Statistical Learning: Data Mining, Inference, and Prediction.

(https://link.springer.com/book/10.1007/978-0-387-84858-7)

4. PostgreSQL Global Development Group. (2020).PostgreSQL Documentation.

(https://www.postgresql.org/docs/)