Intelligent
Code Grading
& Personalized Programming Feedback
A Comprehensive Research Initiative
Exploring the intersection of artificial intelligence, automated assessment, and pedagogical innovation in computer science education.
Project Scope
A comprehensive architectural and pedagogical overview of our research initiative, encompassing system design, educational frameworks, and technical implementation strategies.
Core System Components
A modular ecosystem designed to handle every aspect of modern programming education and assessment with AI precision.
Code Integrity & Provenance Analysis
“Current tools lack integrated detection of AI-generated code alongside traditional syntactic/semantic clones, failing to identify complex provenance shifts in modern hybrid codebases.”
- Syntactic Code Clone detection (Type 1-3)
- Semantic Similarity detection
- AI Likelihood & Plagiarism scoring
Automated Code Analysis and Feedback System
“Existing autograders primarily focus on binary pass/fail results, lacking real-time, Socratic-style pedagogical feedback that guides students through logic errors without providing direct answers.”
- Intelligent Autograder (Semantic analysis)
- Socratic Style pedagogical Chatbot
- Real-time actionable feedback
Behavior & Learning Analytics and IDE Integration
“There is a critical lack of integrated coding environments that link biometric-authenticated liveness with granular behavior analytics to verify the authenticity of the learning journey.”
- Isolated Cloud Coding Platform
- Biometric Authentication & Liveness
- Detailed behavior & provenance analysis
Intelligent Viva Voce System
“Manual oral assessments are unscalable for large cohorts, and current automated systems lack the depth to verify conceptual understanding through adaptive, context-aware dialogue.”
- Automated oral assessment generation
- Intelligent response evaluation
- Scalable student verification
Project Milestones
Tracking our journey from inception to launch with structured research milestones and deliverables.
Research Project Initiation
Core Research Gap Analysis
Progress Presentation 1 (PP1)
Progress Presentation 2 (PP2)
Research Paper Submission
System Integration & Portfolio Launch
Final Presentation & Viva
Downloads
Access our comprehensive research documentation, presentation materials, and project checklists.
Research Paper
Full research paper on "Scalable Hybrid LSH-Based Framework for Structural Code Clone Detection" accepted at ICHORA 2026.
TAF (Topic Application Form)
The initial Topic Application Form (TAF) defining the research problem and proposed area of study.
ViewProposal Documentation
Comprehensive project proposal and detailed scope documentation submitted for academic approval.
ViewFinal Reports
Complete project documentation, final research findings, and comprehensive system implementation reports.
ViewResearch Achievements
Our work has been recognized by international academic communities and peer-reviewed publications.
Scalable Hybrid LSH-Based Framework for Structural Code Clone Detection
Our research on advanced code clone detection using Locality Sensitive Hashing (LSH) has been accepted for presentation at ICHORA 2026.
The People Behind GradeLoop
A multidisciplinary team of researchers and developers driven by a passion for educational technology and AI.
Academic Supervision
Prof. Nuwan Kodagoda
nuwan.k@sliit.lkLeading academic expert in Computer Science education with extensive research in learning technologies.
Dr. Kalpani Manathunga
kalpani.m@sliit.lkExpert in technical systems and educational technology with a focus on AI-driven learning solutions.
Developers
Wickramasooriya W.T.D.
Backend architecture & automated assessment logic
Contact Us
Have questions about our research or want to collaborate? We'd love to hear from you.
