Akademik Dersler - Tanıtım

Course Introduction Information


Code - Course Title	YZM4002 - Graduation Project II
Course Type	Required Courses
Language of Instruction	İngilizce
Laboratory + Practice	0+4
ECTS	5.0
Course Instructor(s)	DOKTOR ÖĞRETİM ÜYESİ NİHAL İNCE
Mode of Delivery	Face-to-Face / Project-Based Learning, Weekly Advisor Meetings, and Independent Study
Prerequisites	No formal prerequisites. (However, students are strongly recommended to have successfully completed foundational programming, mathematics, and artificial intelligence courses up to the end of the 4th semester.)
Courses Recomended	Algorithm Design and Analysis, Principles of Artificial Intelligence, Statistics, Data Acquisition and Processing, Introduction to Optimization Theory, Fundamentals of Supervised Learning, Artificial Neural Networks, Deep Learning, Applied Machine Learning, Introduction to Image Processing, Natural Language Processing.
Required or Recommended Resources	Current international academic papers (IEEE, ACM, Springer) depending on the nature of the selected project topic (Computer Vision, NLP, Data Analytics, etc.), technical documentations, library manuals, and relevant literature.
Recommended Reading List	Departmental Graduation Project Principles and Writing Manual.Current literature from Google Scholar, IEEE Xplore, and ScienceDirect databases.Official documentations of Python Data Science and AI libraries (NumPy, SciPy, PyTorch, TensorFlow, Hugging Face).
Assessment methods and criteria	In-Term Studies / Advisor EvaluationInterim Report / Project Proposal FileEnd-of-Term Project Report & Jury Presentation
Work Placement	There is no mandatory external work placement within the scope of this course. However, if the project is carried out jointly with an industrial partner (R&D Center, Technopark company, or industry-oriented projects like TUBITAK 2209-B), industrial application/fieldwork can be conducted.
Sustainability Development Goals

Content

Weeks	Topics
Week - 1	Term-start planning with the advisor; reviewing the model architecture and timeline designed in the first semester.
Week - 2	Setting up the development environment; coding data preprocessing, cleaning, and data loader blocks.
Week - 3	Commencing the coding of the theoretically designed AI/deep learning model architecture in Python (PyTorch/TensorFlow, etc.).
Week - 4	Completing the modular code structure of the model; establishing the training and validation loops.
Week - 5	Implementation and preparation of baseline models identified in the literature for comparative analysis.
Week - 6	Initiating the initial training processes of the proposed model and conducting hyperparameter tuning studies.
Week - 7	Monitoring model training processes; overfitting checks and application of regularization techniques.
Week - 8	Preparation of the Interim Report: Documentation of the code architecture, training processes, and preliminary results.
Week - 9	Running the model on the test dataset; computing performance metrics (Accuracy, F1-Score, ROC-AUC, MSE, etc.).
Week - 10	Comparing the experimental results of the proposed model with baseline models and conducting graphical/tabular analysis.
Week - 11	Interpreting model results and potential errors from a statistical and algorithmic perspective; conducting ablation studies.
Week - 12	Starting the writing of the final graduation project thesis/report in accordance with academic writing principles.
Week - 13	Finalizing the report; preparing the running project codes (GitHub repo) and the jury presentation/poster.
Week - 14	End-of-Term Defense: Presenting the developed AI system, experimental results, and the final thesis report before the jury.

Learning Activities and Teaching Methods

Teaching Methods
Discussion
Question & Answer
Problem Solving
Report Preparation and/or Presentation
Proje Design/Management
Competences
Productive
Rational
Questoning
Follow ethical and moral rules
Effective use of a foreign language
Use time effectively
Information Management
To work autonomously
Organization and planning
To work in interdisciplinary projects
Project Design and Management

Assessment Methods

Assessment Method and Passing Requirements
	Quamtity	Percentage (%)
Toplam (%)		0