| COMPUTER ENGINEERING (ENGLISH, INTEGRATED PHD) | |||||
| PhD | TR-NQF-HE: Level 7 | QF-EHEA: Second Cycle | EQF-LLL: Level 7 | ||
| Course Code | Course Name | Semester | Theoretical | Practical | Credit | ECTS |
| CMP5130 | Machine Learning and Pattern Recognition | Fall | 3 | 0 | 3 | 12 |
| The course opens with the approval of the Department at the beginning of each semester |
| Language of instruction: | En |
| Type of course: | Departmental Elective |
| Course Level: | |
| Mode of Delivery: | Face to face |
| Course Coordinator : | Assist. Prof. CEMAL OKAN ŞAKAR |
| Course Lecturer(s): |
Assist. Prof. CEMAL OKAN ŞAKAR |
| Course Objectives: | Pattern recognition systems and components; decision theories and classification; discriminant functions; supervised and unsupervised training; clustering; feature extraction and dimensional reduction; sequential and hierarchical classification; applications of training, feature extraction, and decision rules to engineering problems. |
|
The students who have succeeded in this course; I. Understand the nature and inherent difficulties of the pattern recognition problems II. Understand concepts, trade-offs, and appropriateness of the different feature types and classification techniques such as Bayesian, maximum-likelihood, etc. III. Select a suitable classification process, features, and proper classifier to address a desired pattern recognition problem. IV. Demonstrate algorithm implementation skills using available resources and be able to properly interpret and communicate the results clearly and concisely using pattern recognition terminology V. Understand the mathematical statistics foundations of the pattern recognition algorithms VI. Evaluate current research and advanced topics in pattern recognition |
| 1.Density Based Clustering 2.Agglomerative Clustering 3.Cluster Evaluation 4.Cohesion, Separation, Cluster Tendency 5.Prototoype-Based Clustering 6.Fuzzy Clustering 7.Sparsification 8.Optimal Partitioning of Sparse Similarities Using Metis 9.Chamelon 10.Jarvis-Patris Clustering Algorithm 11.BIRCH 12.CURE 13.Combining Multiple Clusterings The teaching methods of the course include lectures, group work, technology-assisted learning, project preparation, and practice. |
| Week | Subject | Related Preparation | |
| 1) | Overview and Decision Trees | None | |
| 2) | Probability Review | None | |
| 3) | Instance-based Learning | None | |
| 4) | Naive Bayes | None | |
| 5) | Logistic Regression | None | |
| 6) | Linear Regression | None | |
| 8) | Neural Networks | None | |
| 9) | Midterm 1 | Review all the topics | |
| 10) | Model Selection | None | |
| 11) | K-means and Hierarchical Clustering | None | |
| 12) | Probabilistic Models for Clustering | None | |
| 13) | Semi-Supervised Learning | None | |
| 14) | Reinforcement Learning | None | |
| Course Notes: | Pattern Recognition and Machine Learning (Information Science and Statistics) by Christopher M. Bishop |
| References: |
| Semester Requirements | Number of Activities | Level of Contribution |
| Attendance | % 0 | |
| Laboratory | % 0 | |
| Application | % 0 | |
| Field Work | % 0 | |
| Special Course Internship (Work Placement) | % 0 | |
| Quizzes | % 0 | |
| Homework Assignments | % 0 | |
| Presentation | % 0 | |
| Project | 5 | % 10 |
| Seminar | % 0 | |
| Midterms | 1 | % 40 |
| Preliminary Jury | % 0 | |
| Final | 1 | % 50 |
| Paper Submission | % 0 | |
| Jury | % 0 | |
| Bütünleme | % 0 | |
| Total | % 100 | |
| PERCENTAGE OF SEMESTER WORK | % 40 | |
| PERCENTAGE OF FINAL WORK | % 60 | |
| Total | % 100 | |
| Activities | Number of Activities | Duration (Hours) | Workload |
| Course Hours | 14 | 3 | 42 |
| Laboratory | 0 | 0 | 0 |
| Application | 0 | 0 | 0 |
| Special Course Internship (Work Placement) | 0 | 0 | 0 |
| Field Work | 0 | 0 | 0 |
| Study Hours Out of Class | 14 | 6 | 84 |
| Presentations / Seminar | 0 | 0 | 0 |
| Project | 5 | 5 | 25 |
| Homework Assignments | 0 | 0 | 0 |
| Quizzes | 0 | 0 | 0 |
| Preliminary Jury | 0 | ||
| Midterms | 1 | 20 | 20 |
| Paper Submission | 0 | ||
| Jury | 0 | ||
| Final | 1 | 20 | 20 |
| Total Workload | 191 | ||
| No Effect | 1 Lowest | 2 Low | 3 Average | 4 High | 5 Highest |
| Program Outcomes | Level of Contribution |