VNU-HUS MAT1206E - Mini Projects


About

This page explains how to prepare and register a mini-project topic for “Introduction to Artificial Intelligence (VNU-HUS MAT1206E)” during Semester 1 of the 2026-2027 academic year. Proposed topics and updates are recorded on the private MAT1206E topic board.

About the Mini Projects

Use the student project template. Read the detailed submission guide and study the worked topic-proposal example before starting. The project is graded manually according to the mini-project rubric; Classroom50 does not grade this assignment.

How to Register a Mini-Project Topic

The course topic board is available to course members. Browse topics, then use the Project topic proposal form after your group has its private Classroom50 repository and an agreed proposal commit. Sign in with your registered GitHub account; access requires membership of the MAT1206E Classroom50 student team. If GitHub shows a 404, ask the lecturers to check your course-team membership.

The Classroom50 final-project acceptance link is published in step 2 below and starts accepting on September 11, 2026 at 13:00 ICT (UTC+7). Before you accept, form your group, read the guide and existing topics, and prepare your ideas. Do not open a topic-registration issue in the student-template repository, and do not submit a placeholder issue without the required proposal commit.

  1. Form the group and choose one founder. Agree on one to five students from the same course and verify everyone’s GitHub username. Review existing proposals on the course topic board before settling on a precise problem.
  2. Only the founder accepts final-project in Classroom50. Use this course’s acceptance link. Other members must not accept separately: this can create duplicate project repositories.
    Classroom50: Final Examination Mini-Project
  3. Initialize the group’s private repository and add members. Follow the bootstrap instructions to copy the starter into the empty repository created by Classroom50. Do not create a second project repository or push to the starter. Add the other agreed members as collaborators, then complete team.json and the private root README with each member’s full name, student ID, and GitHub username.
  4. Prepare the proposal together. Study the completed proposal example, then complete your own proposal/proposal.md, including a precise selected problem, scope and non-goals, method, and expected output. Mini-Project Ideas offers optional inspiration. Edit the real proposal and membership files, not the example files; do not submit the example verbatim.
  5. Review, optionally check, then commit and push. Every listed member must agree to the proposal and submitted version. You may run python3 check_project_files.py proposal using the optional structural checker. Save the permanent GitHub commit URL or complete 40-character SHA after pushing.
  6. Register through one MAT1206E topic-board issue. Search for your exact group-repository URL and the founder’s GitHub username. Reuse an existing group issue; repair an incomplete issue rather than opening another, and ask staff if the canonical issue is unclear. If no group issue exists, the founder opens the Project topic proposal form, completes every field, and links the exact proposal commit. Follow the completed issue example.
  7. Continue in the same canonical issue. Keep proposal revisions, exact-duplicate corrections, scheduling communication, and final submission in that issue. Push each revised proposal and post its new commit URL or SHA there. Changing the representative does not create a new issue; record the agreed handover in the existing issue.

Important Notices

Acceptance is not topic registration. Each group uses one private Classroom50 repository and one canonical topic-board issue. The founder handles administrative actions, but cannot unilaterally change membership, the selected problem, scope, proposal, or submitted commit; every member must agree.

Protect private identities and identify the exact version. Full names and student IDs stay in the private repository; identify members in the class-visible issue by GitHub username only. Use an immutable commit URL or full SHA, not a branch link, a screenshot, or a link to the latest file.

Registration is not academic approval. status: submitted means awaiting the exact-duplicate check. status: recorded means no earlier exact duplicate was found when staff reviewed the issue; it is not approval or a guarantee of feasibility, correctness, quality, or a passing grade. status: duplicate-problem requires a revised problem and a new proposal commit in the same issue. The proposal is required but ungraded; the optional checker gives no score and does not evaluate project quality.

Deadlines

Required submission Deadline (ICT, UTC+7)
Topic registration, initial proposal and required corrections September 27, 2026 at 23:59
Exact-duplicate correction, when required September 27, 2026 at 23:59
Final commit and FINAL SUBMISSION comment in the same issue November 4, 2026 at 23:59

The final deadline is the same for all groups; proposal revisions and presentation order do not extend it. See the submission guide for the complete calendar and comment formats.

Final Report, Presentation, and Evaluation

Follow the report instructions and slide instructions. Every group member participates in the presentation. Complete the private contribution record, AI-use declaration, and external-resource declaration. Evaluation follows the manual grading rubric.

Lecturers Involved in Evaluation

  • Hoàng Anh Đức (Đại học KHTN, ĐHQG Hà Nội)
    • Email: hoanganhduc[at]hus.edu.vn (replace [at] with @)
    • GitHub Username: hoanganhduc
  • Lê Huy Hùng (Đại học KHTN, ĐHQG Hà Nội)
    • Email: lehuyhung94[at]gmail.com (replace [at] with @)
    • GitHub Username: HuyHung0

Proposed Topics

The private topic board remains the official record for registration and updates. Only reviewed registrations currently labeled status: recorded are published below. Group numbers follow the most recent assignment of that label, from earliest to latest; removing and reapplying the label changes the ordering timestamp. Numbers may therefore change. Recording is not academic approval or a guarantee of quality. Presentation times will be announced separately. Repository links remain private and require the appropriate access.

  1. Group 1: Traffic Sign Recognition for Autonomous Driving Using Deep Learning
  2. Group 2: Hanoi Explorer Chatbot
  3. Group 3: Phân loại chữ số viết tay bằng học máy
  4. Group 4: Developing an Intelligent Agent for Gomoku using Minimax with Alpha-Beta Pruning and Heuristic Evaluation
  5. Group 5: ChessMind: A Reinforcement Learning-Based Chess AI
  6. Group 6: Autonomous Driving in a Simulation Environment Using Object Detection and Reinforcement Learning
  7. Group 7: Hệ thống AI hỗ trợ phát hiện sớm và chuẩn đoán tổn thương ung thư vòm họng & khoang miệng
  8. Group 8: Xây dựng Agent AI chơi Cờ Vây kích thước nhỏ (9x9) sử dụng Học có nhãn (Supervised Learning)
  9. Group 9: Movie Recommendation System using Collaborative Filtering and Content-Based Filtering