International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 12 Issue: 12 | Dec 2025
p-ISSN: 2395-0072
www.irjet.net
ENGINEERING EDUCATION SUPPORT SYSTEM FOR ONLINE COURSE SELECTION USING PYTHON: A SUPERVISED LEARNING APPROACH Dr. Vinay V. Kuppast1, Meghana S. Biradar2, Shivakumar S. Shidaraddi 3, Sudeep Shettar4, Shanmukhagouda M. Karakanagoudra5 1Professor, Dept. of Mechanical Engineering, Basaveshwar Engineering College, Bagalkote, Karnataka, India 2,3,4,5 Students, Dept. of Mechanical Engineering, Basaveshwar Engineering College, Bagalkote, Karnataka, India
------------------------------------------------------------------------***-------------------------------------------------------------------------
Abstract - Course selection plays a crucial role in
multiple suitable options and reducing uncertainty in the decision-making process.
determining a student’s academic growth and career trajectory. Traditional manual selection processes often lack personalization, leading to confusion and misalignment of student skills with course requirements. This paper presents an intelligent Engineering Education Support System that uses supervised machine learning, specifically a Random Forest Classifier, to recommend suitable courses to engineering students. The system analyzes student attributes such as GPA, mathematical proficiency, coding skill, interest areas, academic stream, and career goals to generate ranked course recommendations. A Python-based Tk inter GUI ensures user-friendly interaction, while an SQLite backend maintains structured records for administrative analysis. Experimental results demonstrate that the model provides accurate, transparent, and personalized recommendations. The system addresses critical challenges in engineering education by enhancing decision-making, reducing selection errors, and supporting digital academic planning.
2. LITERATURE REVIEW The growth of online learning platforms and MOOCs has created a strong need for intelligent course-selection support systems. Traditional advising often lacks personalization, clear prerequisite guidance, and structured pathways, leading to poor decision-making among learners. Hour et al. [1] highlighted major challenges in MOOCs, including unclear prerequisites and information overload. Jena et al. [2] showed that students struggle to match their skills and interests to suitable courses, and that collaborative filtering improves recommendations but suffers from sparsely and cold-start issues. Content- based models, though widely used, lack adaptability. Wang et al. [3] demonstrated that they often miss semantic relationships between concepts. To address this, Altars et al. [4] introduced Concept GCN, a graphbased model that integrates semantic and behavioral data for improved accuracy.
Key Words: Machine Learning, Course Recommendation System, Random Forest, Python, Tk inter GUI, Educational Data Mining, SQLite
Prerequisite- and sequence-aware models also enhance decision-making. Chanaa and El Faddily [5] proposed a prerequisite-matching framework, while Wong [6] developed a sequence-based planner using historical enrollment patterns. Khan [7] showed that session-based methods capture temporal learning behavior and improve short-term recommendations.
1. INTRODUCTION Course selection in engineering education has become increasingly complex due to diversified electives, interdisciplinary learning pathways, and evolving skill requirements. Students often rely on informal advice or incomplete information, resulting in suboptimal course choices. With the rise of data-driven academic systems and AI-assisted learning tools, there is a strong need for an intelligent, automated support mechanism that assists students in making informed academic decisions.
Overall, literature trends show a shift toward hybrid, semantic, and context-aware models that overcome limitations of traditional content-based and collaborative filtering methods. These advanced approaches offer better accuracy, adaptability, and personalization—essential for effective course recommendation in engineering education.
This study proposes a Python-based recommendation system that uses machine learning techniques to analyze student profiles and recommend appropriate courses. The system is equipped with graphical user interface (GUI) and database storage to ensure transparency, accessibility, and long-term usability. The model leverages historical training data to identify patterns and generate ranked course recommendations, providing students with
© 2025, IRJET
|
Impact Factor value: 8.315
|
ISO 9001:2008 Certified Journal
|
Page 113