AUTONOMOUS

Department of Computer Science and Engineering

AI-Powered Learning
Platform
Revolutionizing Education with Artificial Intelligence
Team Members:
•Y. VemaLatha (21NG1A05C3)
UNDER THE GUIDANCE OF: •T. Vandhana (21NG1A05B4)
Dr. S. M. Roy Choudri •Md. Nizam Farhan (21NG1A0596)
(HOD of CSE department) •M. Charan Teja (21NG1A0595)

Dept of Computer Science and Engineering

URCE
 Abstract Overview:
The AI-Powered Learning Platform (AIPLP) is designed to transform educational
experiences through advanced AI technologies.

•It leverages AI for personalized learning paths, real-time feedback, and adaptive
content tailored to individual student needs.

•The platform incorporates machine learning algorithms, natural language

processing, and data analytics to adjust learning experiences dynamically.

•Key features include intelligent tutoring systems, automated grading, and

predictive analytics to monitor progress, aiming to create an engaging and
effective educational journey.

01
Dept of Computer Science and Engineering
URCE
 Key Components that Make the AI-Powered
Personalized Learning Platform Work
1. Frontend Framework and User Interface:
• HTML, CSS, JavaScript: Forms the foundation of the user interface, ensuring accessibility and
usability across devices.

• React.js (or Angular.js): Provides a dynamic and reactive interface, essential for creating an
engaging, seamless learning experience.

• Bootstrap: Ensures the platform is responsive, adjusting its layout to different screen sizes for
an optimized user experience on all devices.

• Redux: Manages the state of the application, ensuring that user interactions, such as
course progress, test scores, and preferences, are consistently tracked and updated.

• Material UI / Chakra UI: These pre-built components create a modern, interactive, and
user-friendly interface, making it easy for users to navigate and engage with the platform.
02
Dept of Computer Science and Engineering
URCE
 3. Communication Features:
• WebRTC and Twilio APIs: These tools allow the integration of live group meetings,
video conferencing, and real-time communication features. This is key for collaborative
learning environments and virtual classrooms, enhancing the interactive aspect of the
platform.

4. Backend Architecture:
• Java (Spring Boot): Handles server-side logic and core operations such as managing
user authentication, database interaction, and API communication. It defines the key
entities, relationships, and business logic, ensuring smooth platform functionality.

• Python (Flask): Powers AI features that require advanced data processing, such as
recommendation algorithms, adaptive learning paths, and natural language
processing. Python’s flexible AI/ML frameworks are essential for personalized learning
tasks.

• Node.js: Provides additional backend capabilities for handling event-driven, non

blocking operations, ensuring the platform can handle multiple requests and deliver
real time updates. 03
Dept of Computer Science and Engineering
URCE
 5. AI and Machine Learning:
• PyTorch, TensorFlow, SciKit-Learn: These AI libraries are responsible for
personalized learning algorithms, adapting course material to individual
student needs by analyzing their behavior, preferences, and performance.

• NLP (GPT Models and BERT): Natural language processing models help with
automatic grading, personalized feedback, and chatbot interactions. GPT-like
models provide intelligent, context-aware responses, while BERT is used to
evaluate student work and assign grades, reducing the need for manual
grading.
DATABASE
6. Database Management:
• MySQL: Acts as the backbone for data storage, storing everything from user
information, course content, and AI-generated recommendations to
interaction histories and learning analytics. MySQL ensures data integrity and
efficient retrieval during user sessions.

04
Dept of Computer Science and Engineering
URCE
 Literature Survey
Key Findings:
Existing AI Tools: Numerous studies highlight the emergence of AI in education,
with tools providing interactive and personalized learning experiences.

Traditional Approaches: Traditional educational methods often lack flexibility and

responsiveness, leading to disengagement among learners.

Personalization Importance: Research indicates that personalized education

significantly enhances learner engagement, motivation, and retention of
information.
Technological Trends: The integration of AI, ML, and NLP in educational platforms
is a growing trend, providing innovative solutions to address diverse learning
needs.
Dept of Computer Science and Engineering 05
URCE
 Algorithm Paper/Author Research Paper

Natural Language Processing Brown et al. “Language models are few

(NLP) shot learners” (GPT-3) arXiv

“Matrix Factorization
Collaborative filtering (for Koren et al. techniques for recommender
recommendations) system” Research Paper

Knowledge Tracing (student Peich et al. “Deep Knowledge Tracing”

modelling) arXiv

“Automated Question
Automated Question Zhang et al. Generation from Educational
Generation
Texts” (GPT-3) Research Paper

06
Dept of Computer Science and Engineering
URCE
 Motivation to Our Problem Statement

•Increased Demand: The rise in online education has highlighted the necessity for
personalized solutions that cater to individual learning preferences and styles.

•Challenges in Traditional Education: Many learners struggle to keep pace in

traditional settings, which can lead to frustration and poor academic performance.

•Transformational Potential of AI: AI has the capacity to revolutionize how education

is delivered, making it more adaptive and engaging.

•Need for Scalable Solutions: The shift towards scalable, interactive, and data-driven
educational tools is essential for meeting the needs of a diverse learner population.

07
Dept of Computer Science and Engineering
URCE
 Problem Statement

•Inefficiencies in Traditional Systems: Conventional education systems often

struggle to adapt to the varying needs of students, resulting in missed learning
opportunities.

•Lack of Personalization: A one-size-fits-all approach fails to account for individual

learning styles, leading to disengagement and suboptimal learning outcomes.

•Difficulty Tracking Progress: Educators often find it challenging to monitor and

adapt to the unique learning paths of each student, hindering timely intervention.

•Timeliness of Feedback: Traditional grading and feedback mechanisms can be

slow, preventing students from receiving immediate assistance when they need it.

08
Dept of Computer Science and Engineering
URCE
 Overview:

The feasibility study assesses the viability of the AIPLP through various
lenses, including technical, operational, and economic aspects.

This evaluation is crucial for understanding the project’s potential

impact, resource requirements, and the overall implementation
strategy.

Feasibility Study

Dept of Computer Science and Engineering 09

URCE
 Technical Feasibility

 Technologies Used: The frontend is developed using HTML, CSS,

JavaScript, and React.js.

 Backend Frameworks: Java with Spring Boot handles core business

logic, while Python (Flask) supports AI functionalities, ensuring efficient
data processing.

 AI and ML Integration: Libraries like PyTorch and TensorFlow are

utilized for developing machine learning models that enhance
personalized learning experiences.

 Database Management: A MySQL database stores user data,

learning materials, and analytics, ensuring quick and reliable data
retrieval. 10
Dept of Computer Science and Engineering
URCE
 System Specifications:

A decent GPU is crucial for training AI

models because it handles large amounts of
CPU: Intel Core i5 (or) i7 parallel computations efficiently, speeding
RAM: 16GB DDR4 Memory up tasks like matrix operations and deep
STORAGE: 512 GB NVMe SSD learning tasks. GPUs are optimized for
handling the vast amounts of data and
computations involved in model training,
GPU: nVIDIA GeForce RTX 2050 and above leading to faster results compared to CPUs,
*Extra Cloud Power is used for
especially for large datasets and complex
training the AI Model models.

11
Dept of Computer Science and Engineering
URCE
 Operational Feasibility

•User Requirements: The platform is designed to meet the needs of students and
educators by providing personalized learning paths and real-time feedback.

•Real-Time Data Processing: The architecture supports the processing of user

interactions and data analytics, ensuring that the platform can adapt quickly to user
behavior.

•Communication Features: Integration of WebRTC and Twilio for live communication

enhances collaboration and interactivity in virtual classrooms.

•Training and Support: The project includes training for educators to effectively utilize
the platform, ensuring they can leverage AI features to support student learning.

12
Dept of Computer Science and Engineering
URCE
Use Case Diagram

Dept of Computer Science and Engineering

URCE 13
 System Use Case Diagram

Dept of Computer Science and Engineering

URCE 14
 Data Flow Diagram

15
Dept of Computer Science and Engineering
URCE