Assessment of supervised learning techniques: case of bikesharing

Bc. Anastassiya Li

Master's thesis

Abstract:

The topic of forecasting is relevant, as obtaining the most accurate prediction of the future allows making the most advantageous decisions for achieving positive results in any sphere of human activity, including social. Research subject of this study deals with prediction of bikesharing hourly count using regression analysis and neural network. R is chosen as a programming language for realisation. The first section of this paper provides the literature review needed to understand the topic of the work and the methods used. The second section focuses on practical implementation on selected techniques. The third section describes results of predictions and their performance evaluation. The relevance of the thesis is due to forecasting demand in general and supervised learning techniques assessment.

Keywords

neural network, training, testing, regression, MSE

Language used: English

Date on which the thesis was submitted / produced: 3. 4. 2023

Thesis defence

• Date of defence: 25. 5. 2023
• Supervisor: doc. Ing. Vojtěch Merunka, Ph.D.
• Reader: Martin Molhanec, externi

Czech University of Life Sciences Prague

Faculty of Economics and Management

Master programme / field:
Systems Engineering and Informatics / Informatics