Předpovídání časových řad v městské dopravě pomocí RNN LSTM

Abstract

This diploma thesis focuses on the development of a comprehensive forecasting system for urban mobility, with a specific emphasis on parking occupancy and public transport delays. Robust data collection and preprocessing pipelines are established to incorporate data from diverse sources. State-of-the-art machine learning techniques, particularly Long Short-Term Memory (LSTM) models, are implemented to forecast urban mobility variables. External factors such as weather data are integrated to enhance forecasting accuracy. A user-friendly web application is developed for data exploration and analysis. The research highlights the potential of data-driven forecasting systems in optimizing transportation efficiency and contributes to the field of intelligent transportation systems.

This diploma thesis focuses on the development of a comprehensive forecasting system for urban mobility, with a specific emphasis on parking occupancy and public transport delays. Robust data collection and preprocessing pipelines are established to incorporate data from diverse sources. State-of-the-art machine learning techniques, particularly Long Short-Term Memory (LSTM) models, are implemented to forecast urban mobility variables. External factors such as weather data are integrated to enhance forecasting accuracy. A user-friendly web application is developed for data exploration and analysis. The research highlights the potential of data-driven forecasting systems in optimizing transportation efficiency and contributes to the field of intelligent transportation systems.