FOLLOWING USER TRACES IN URBAN CONTEXT: STIGMERGIC APPROACH AND A MACHINE LEARNING IMPLEMENTATION

Date

2023-4-13

Author

Akyürek, Handan

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

181
views

32
downloads

Cite This

Designers are responsible for responding to the user’s needs and expectations regarding the design output. Hence, the design feedback mechanism between the user and the designer is crucial to avoid any mismatch between the designer’s decisions and the user’s needs. On the other hand, displaying the user’s engagement with the design output provides a reliable communication ground for both designer and the user. Integrating state-of-art visualization mediums with the recent participatory design methods creates highly sophisticated communication platforms where the users experience the design ideas more immersively, which gives potential data regarding their engagement with the design ideas. However, these data are not authentic enough because they are retrieved from the user’s engagement with the representation of the design outcome, not the real one. In this context, this thesis focuses on the user's physical traces during the post-occupancy phase as the representation of their engagement with the design output. It formulates a feedback mechanism to understand, learn and provide from the user’s preferences to enhance future projects. vi Desired paths in urban contexts were selected to examine the user’s physical traces to represent the user's preferences regarding the designer's design decisions. A stigmergic approach was created to investigate their self-emerging and spatiotemporal characteristics. Finally, the desired paths were represented as environmentally mediated signs and formalized as a design feedback mechanism to utilize a machine learning algorithm to create a prediction tool in the urban design context.

Subject Keywords

User experience, design feedback, desired paths, stigmergy, machine learning

URI

https://hdl.handle.net/11511/103270

Collections

Graduate School of Natural and Applied Sciences, Thesis