Practical prediction systems 1000-2D20PRD
The topic of the seminar will be an overview of various aspects of creating practical prediction systems. We will tackle all the challenges facing such systems in applications in real-world systems. We will discuss all elements of the process of creating systems such as data preparation, feature generation, model building and explainability.
At the seminar, students will present papers from journals and conference reports on this field of computer science.
Master's theses can be both review and experimental.
In the presence of foreigners, classes will be conducted in English.
Type of course
Mode
Course coordinators
Term 2024: | Term 2023: |
Learning outcomes
Knowledge (KW_01, KW_02):
- The student knows the basic techniques used in prediction systems and understands the current "trends" in this field.
Skills (K_U11 - KU_15):
- The student can specify prediction problems, is able to analyze the problem and propose an efficient solution.
- The student is able to write scientific publication in the field of algorithms, also in English.
- The student is able to prepare a presentation in the field of algorithms, also in English.
Social competence (K_K01-K_K09):
- The student is able to find the information on the given research problem in available sources and evalutes its reliability and usefulness.
- The student can present specialized results in a manner understandable to non-specialists.
- The student understands the importance of intellectual property rights in the use of other people's results.
Assessment criteria
Active participation in the seminar and giving a talk together with obligatory requirements of the studies.
Bibliography
Modern scientific literature of the subject, including scientific journals and data from Internet. Details are provided by the lecturers at the first meeting.
Additional information
Additional information (registration calendar, class conductors, localization and schedules of classes), might be available in the USOSweb system: