2022-01-252022-01-252021https://hdl.handle.net/1920/12615Human activities in cities, such as eating foods in restaurants and attending special events, are traditionally measured using surveys, questionnaires, and interviews. While such methods have provided valuable insights, there are issues, such as cost when exploring large geographical areas during long periods of time. With the development of social media platforms and open data initiatives, a large amount of data that contains information on people’s interests and opinions, geolocation, and timestamps are becoming available to the general public. The question is, how can we utilize these datasets with three dimensions (i.e., textual, geographical, and temporal) to explore questions related to human activities in cities? How do we use the results discovered from the data to better study and plan our cities? This dissertation research used a multi-disciplinary Computational Social Science approach to explore these questions. The research specifically demonstrates how to explore food related discussions, special events and their impact on traffic flow using open sources data. It also shows the potential and methods to explore human activities in cities using social media data together with other data sources, and through agent-based simulation and analysis, the results can help planners and engineers to study, plan, and manage our cities.