Publication
Predicting the Movement of People in Lisbon’s Nightlife: Leveraging Machine Learning and Terminal Roaming Data to Support Urban Planning
| Summary: | This master’s thesis investigates the prediction of mobility patterns within Lisbon’s nightlife areas, focusing on temporal elements. The primary objective is to predict these movements using historical data and machine learning models. Various regression models were employed and evaluated based on their accuracy in forecasting nightlife activities, with the Multi-Layer Perceptron Regression model emerging as the most accurate. Findings revealed that temporal features, particularly previous hour data, were the most significant predictors with year and month features also playing important roles. This research supports the potential of leveraging machine learning and terminal data in urban planning, providing insights that can enhance decision-making and improve the nightlife experience through strategic service scheduling and overcrowding management. Ultimately, the understanding of mobility patterns and key determinants of nightlife activity in Lisbon enhances the city's advance towards the development of a smarter and more sustainable urban environment. |
|---|---|
| Main Authors: | Santos, Maria Carolina Bentes Pimenta Paisana |
| Subject: | Smart Cities Lisbon Urban Planning Mobility Patterns Nightlife Mobility Modelling SDG 9 - Industry, innovation and infrastructure SDG 11 - Sustainable cities and communities SDG 17 - Partnerships for the goals |
| Year: | 2024 |
| Country: | Portugal |
| Document type: | master thesis |
| Access type: | open access |
| Associated institution: | Universidade Nova de Lisboa |
| Language: | English |
| Origin: | Repositório Institucional da UNL |
| Summary: | This master’s thesis investigates the prediction of mobility patterns within Lisbon’s nightlife areas, focusing on temporal elements. The primary objective is to predict these movements using historical data and machine learning models. Various regression models were employed and evaluated based on their accuracy in forecasting nightlife activities, with the Multi-Layer Perceptron Regression model emerging as the most accurate. Findings revealed that temporal features, particularly previous hour data, were the most significant predictors with year and month features also playing important roles. This research supports the potential of leveraging machine learning and terminal data in urban planning, providing insights that can enhance decision-making and improve the nightlife experience through strategic service scheduling and overcrowding management. Ultimately, the understanding of mobility patterns and key determinants of nightlife activity in Lisbon enhances the city's advance towards the development of a smarter and more sustainable urban environment. |
|---|
Funded activities
No funded projects found.