Grading rules 2026/2027
Assessment
The assessment for this course consists of:
- Semestral Project (Maximum 70 points): The semestral project is an individual undertaking. Detailed project guidelines are provided below.
Qualification for the Final Exam: to be eligible to take the final exam, students are required to attain a minimum total score of 35 (Maximum 70) points across both the tutorial homework and the semestral project components.
Exam
As part of the assessment, there will be an oral exam with the following parameters:
- Maximum Score: 30 points (To pass, a minimum of 15 points on the final exam is required).
- You will receive two topics for discussion: 1) a question related to your semestral project; and 2), a topic based on the content covered in lectures and tutorials.
- You will have 15 minutes for preparation before the oral exam begins.
- The oral exam itself will follow the preparation period.
Points breakdown:
- You can earn up to 15 points for your responses to questions related to your semestral project.
- You have the opportunity to earn a maximum of 15 points for answering questions related to lecture topics. Please note that lecture topics for the final exam will be announced at a later date.
Exam Topics (2025/2026):
TBA
Evaluation scale
Your grade is determined by sum of points from tutorials and exam (at most 100 points).
Evaluation scale is according to The Study and Examination Code for Students of CTU in Prague.
| Grade | Points | Wording |
|---|---|---|
| A | 90 - 100 | excellent |
| B | 80 - 89 | very good |
| C | 70 - 79 | good |
| D | 60 - 69 | satisfactory |
| E | 50 - 59 | sufficient |
| F | less than 50 | failed |
Semester Project (WS 2025)
This description is from the previous semester. The description for the current semester will be announced soon.
In the semester project, your task involves conducting scientific or applied research to enhance existing PML methods, culminating in the creation of a concise scientific paper. The primary focus lies within the domain of Recommender Systems; however, potential applications of PML extend to various other fields. This paper should build upon the latest developments presented at the following conferences: RecSys, KDD, CIKM, or WWW in the years 2023, 2024, or 2025.
Instead of a report, the scientific paper must follow the Association for Computing Machinery (ACM) template. The paper should range from 4 to 8 pages in length, excluding references. It is expected to extend and improve existing PML methods. This may involve enhancing the accuracy, efficiency, or applicability of the method. The extension should demonstrate both originality and innovation.
Evaluation Criteria:
Students will be assessed based on the quality of their scientific paper, the novelty and significance of their research extension, and their ability to communicate their findings effectively.
Important Dates:
- 16.10.2025 - Presentation of idea / based-paper / dataset (5 points)
- 30.10.2025 - Related works and methodology (5 points) - Plase, send a version by 29.10.2025 (in PDF) via email with subject "[NIE-PML2526] - your KOS login - Data/Method". For example, my login in KOS is dasilrod, therefore the subject would be: "[NIE-PML2526] - dasilrod - Data/Method".
- 13.11.2025 - First experimental results (5 points)
- 27.11.2025 - Project tutorial
- 18.12.2025 - Poster presentation
- 05.01.2026 - Paper submission (together with the poster, 55 points)
Please add (by 23.10.2025) the topic of your project here. We will have different topics for each student, so it’s first come, first served. The poster must be printed and presented during our lecture on 18.12.2025, as in academic conferences. For the paper submission, you must send a PDF file to my email with the subject line: "[NIE-PML2526] - your KOS login - Paper". For example, my login in KOS is dasilrod, therefore the subject would be: "[NIE-PML2526] - dasilrod - Paper".
Topic Suggestions
- Boost serendipity by using semantic similarity to introduce meaningful but surprising recommendations.
- Create curated recommendation paths that use semantic meaning and user context for point-of-interest suggestions.
- Model evolving user preferences over time with temporal embeddings or decay-based representations.
- Produce short, human-readable explanations that clarify why each item was recommended.
- Protect user privacy during recommender model training.
- Detect and counteract fake reviews or injected user profiles that attempt to manipulate recommendations.
- Automatically generate coherent baskets or bundles of complementary items for users with LLMs.
- Visualize the behavior and dynamics of temporal recommender models through interactive data plots.
- Analyze cognitive aspects (with semantic) of how users perceive and interact with recommendations.
- Use sparse autoencoders to learn compact and interpretable representations for personalized recommendations.
- Apply personalized machine learning techniques to improve sports prediction.