ELREFAEI, I. Time Series Forecasting Using Machine Learning [online]. Brno: Vysoké učení technické v Brně. Fakulta elektrotechniky a komunikačních technologií. 2024.
The thesis, titled "Time Series Forecasting Using Machine Learning," focuses on implementing and applying machine learning (ML) techniques to datasets to predict future trends. In this revised version of the diploma thesis, the student has addressed most of the major issues, leading to the conclusion that the overall goals were achieved. The updated structure is more logical, allowing the reader to better follow the methodology and key contributions. However, the thesis would still benefit from more careful proofreading, as it contains several grammatical errors and minor formatting issues. Although the list of references has been updated, it still includes a high number of internet sources, which are not considered reliable. Overall, the thesis presents good results and is suitable for defense in front of the committee.
The student presents a masters thesis dedicated to time series forecasting. He studied the performance of multiple algorithms and models and compared their performance given multiple datasets. In the end a student gave conclusions about the capabilities of the models, strengts, weeknesses, limitation, etc. The positive side of the thesis is the length in which a student described various algorithms, experiments, etc. The negative side lies in the non-optimal format look of it as there are quiet a few inacuracies when it comes to the presentation of tables, figures, spacing, etc. I would also like to see deeper discussion of the results and stronger conclusions as presented. Nevertheless, a student did what the thesis required and due to its limitations I suggest rating the thesis with 60 points and the grade D.
eVSKP id 161929