FYP – Grad,io

Project Title

Final Year Project: Grad.io – University Recommendation System

Project Description

Grad.io is an innovative university recommendation system designed to align students’ academic paths with their unique personalities, interests, and career aspirations. The system assesses user personalities through the RIASEC personality test & user’s text-based input for their interests. Utilizing a comprehensive dataset gathered from various university websites via web scraping, Grad.io provides tailored recommendations for academic programs according to their academic grades.

Key Responsibilities

Team Leadership: Led a team of 4 in developing Grad.io, a platform that assists students in selecting universities and degree programs based on their grades, preferences, and interests.
Backend Development: Designed and developed Backend APIs, incorporating RESTful architecture, to process & handle the scrapped data of universities, user’s academic grades, RIASEC Personality assessment results & user interests.
Personality Assessment Implementation: Implemented the RIASEC model for personality assessment, using Cosine similarity & Sentence Transformer to match students’ interests and personalities with suitable university programs.
Natural Language Processing: Utilized NLP libraries along with Porter Stemmer and word lemmatization for preprocessing user interests.
Security: Implemented authentication and tokenization to ensure secure access to the system.
Model Optimization: Optimized the model to increase the accuracy of university program recommendations by 30%.
Web Scraping: Implemented web scraping using Selenium to gather data from around 200 leading universities in Pakistan. Ensured the data was up-to-date and accurate for reliable recommendations.

Technologies Used

Frontend: React.js
Backend: Node.js
Database: MySQL
Web Scraping: Beautiful Soup, Selenium
Model Training: Flask, NLTK, scikit-learn
Deployment: Heroku & Digital Ocean

Challenges Faced

Ensuring accurate and up-to-date data collection from university websites.
Balancing multiple factors such as academic records, personality traits, and interests to provide reliable and relevant recommendations.
Designing an intuitive and user-friendly interface for diverse user inputs.

Results Achieved

Successfully developed a fully functional university recommendation system that provides personalized university recommendations based on academic records, personality traits, and interests.
Received positive feedback from users for the system’s accuracy and relevance in university recommendations.

Key Features

User registration and authentication
Academic input and preferences
RIASEC personality tests and personalized recommendations
User’s text-based input for Interests
Scrapped data from 200 universities of Pakistan
User-friendly interface and responsive design
Algorithms for accurate recommendations

Conclusion and Future Work

Grad.io aims to simplify the decision-making process for students, helping them identify the best educational opportunities that align with their individual profiles and goals. Future work includes expanding the dataset to include more universities, enhancing the recommendation algorithms, and incorporating more advanced AI and ML techniques for even more accurate recommendations.