Institutional Repository of Vilnius University
Keleivių srauto stotelėse prognozavimas taikant mašininio mokymosi metodus /
Abstract
dc:descriptionThis work investigates the potential of machine learning methods to predict the number of passengers at the stop after an hour. This study aims to improve the prediction accuracy achieved according to MAE and RMSE metrics at the workplace. The dataset used for this study contains passenger flow data at stops in the South Korean capital and weather variables. Work includes a literature review, data analysis, and dataset preparation, which involves removing outliers, assessing multicollinearity, and performing normalization for independent variables. The k-nearest neighbors, support vector, GBDT, XGBoost, LightGBM, and Catboost methods were applied and reviewed. Catboost with 15000 maximum trees and maximum tree depth 10 showed the best results, with 0.981 coefficient of determination and 50.605 and 82.663 values of MAE and RMSE.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2023
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Šepikas, Matas,
Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- lit
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD192845722&prefLang=en_US