CENG 399
Concepts of Artificial Intelligence
This course is a service course offered by the Department of Computer Engineering at İzmir Institute of Technology for students from different disciplines. It is designed as an application-oriented introductory course that aims to introduce artificial intelligence through its basic concepts, methods, and application areas rather than through technical details.
The course content begins with the basic concepts of artificial intelligence and machine learning, and continues with data representation, data preprocessing, visualization, classification, regression, clustering, and model validation. In the following weeks, artificial neural networks, an introduction to deep learning, generative artificial intelligence, large language models, discipline-specific applications, case studies, and responsible AI are covered.
Practical exercises will make use of tools such as Python, Jupyter Notebook, scikit-learn, WEKA/Orange. Prior programming knowledge is not required; however, programming experience may provide an advantage.
Course Objectives
The main objective of the course is to help students develop artificial intelligence literacy. In this context, students are expected to recognize the types of data specific to their own fields, understand data preprocessing and analysis processes, and learn the basic working principles of commonly used machine learning and deep learning approaches. They are also expected to understand how these methods can be applied to real-world problems in their own disciplines, how to select appropriate models, how to determine suitable evaluation metrics, and how to interpret model outputs correctly.
Assessment
Assessment consists of a midterm exam (30%), a final exam (40%), and a semester-long project (30%). As part of the project, students solve a problem related to their own field by using artificial intelligence models, prepare a final report, and present their work orally at the end of the semester.
Recommended or Required Reading
Hands-On Machine Learning with Scikit-Learn and PyTorch: Concepts, Tools, and Techniques to Build Intelligent Systems, Aurélien Géron, 2025.
| Week | Topics |
| 1 | Introduction to Artificial Intelligence and Machine Learning |
| 2 | Data and information; data representation; Colab/Python working environment |
| 3 | Supervised and unsupervised ML models; hands on scikit-learn, WEKA/Orange |
| 4 | Data preprocessing and visualization; data quality; feature scaling and normalization |
| 5 | Classification: kNN and Logistic Regression; intuition for F1 and ROC-AUC |
| 6 | Decision Trees and Random Forests; interpretability (feature importance) |
| 7 | Regression: linear and polynomial models; MAE and RMSE |
| 8 | Unsupervised learning: k-means and hierarchical clustering; PCA; visualization |
| 9 | Model validation: cross-validation; overfitting vs underfitting; proper metric selection |
| 10 | Introduction to Neural Networks and Deep Learning; perceptron; intuition of backpropagation |
| 11 | GenAI and Large Language Models (LLMs): core concepts, use cases, and risks |
| 12 | Disciplinary applications and Case Studies; domain adaptation; domain-specific models |
| 13 | Ethics and responsible AI; bias, fairness, privacy, explainable AI, human-in-the-loop AI |
| 14 | Project Presentations |
