Description
Research proposal.Transport planning relies heavily on traditional data collection methods such as manual surveys and traffic counts, yet these methods are expensive, time-consuming, and prone to errors. Emerging technologies like big data developments can tackle some of the biggest challenges for mobility planning while complementing traditional research methods. More specifically, large language models, such as GPT provide an opportunity to gain an understanding of citizens' perceptions and feelings regarding changes in active mobility. These natural language models can be used to analyse user-generated content (UGC), such as social media, to help policymakers better understand mobility behaviour. Yet this recent methodology and its scope have been under researched in an urban mobility context. Therefore, this project aims to explore the usability of natural language models for analysing UGC (such as tweets) and press articles to understand the public's perceptions regarding active and shared mobility as a whole and their related infrastructure. Special attention will be paid to the differences in perceptions according to gender to understand how to develop a gender-inclusive active and shared mobility policy. Despite the differences in mobility patterns and habits between men and women, very little is known about the role of gender perceptions in the choice of transport modes.
We do this in the context of Brussels, Belgium, for UGC and press articles between 2018-2023. Brussels has a fast-changing mobility context that has been quite controversial, aiming to increase the share of active modes in the city. This research project will be of value to policymakers by allowing them to understand the general public's perceptions in a broader way than they usually have access to by obtaining up-to-date qualitative data on their feelings and needs. Additionally, we will also demonstrate whether methodologies based on large language models can be a complementary source of information during decision-making processes.
| Period | 2023 |
|---|---|
| Degree of Recognition | International |