UCL Discovery
UCL home » Library Services » Electronic resources » UCL Discovery

Human versus machine - testing validity and insights of manual and automated data gathering methods in complex buildings

Sailer, K; Pachilova, R; Brown, C; (2013) Human versus machine - testing validity and insights of manual and automated data gathering methods in complex buildings. In: Ook Kim, Y and Tae Park, H and Wook Seo, K, (eds.) Proceedings of the 9th International Space Syntax Symposium. Sejong University Press: Seoul, South Korea. Green open access

[thumbnail of Sailer_Pachilova_Brown_2013_HumanMachine_SSS9.pdf]
Preview
PDF
Sailer_Pachilova_Brown_2013_HumanMachine_SSS9.pdf
Available under License : See the attached licence file.

Download (765kB)

Abstract

With the advancement of information technologies, automated methods of gathering data on space usage patterns in complex buildings using sensors are gaining popularity. At the same time, typical Space Syntax studies still rely on traditional social science methods and manual data gathering, for instance through direct observations and user surveys. How insights generated by each approach compare to each other is still poorly understood. Therefore this paper reports findings from an in-depth two week long study of space usage in a university building, where both manual methods (direct observations, user surveys) and automated data gathering methods (RFID sensors recording locations and interactions of users) were employed in parallel. The main hypotheses to be tested are that automated data captured by RFID sensors delivers comparable findings (1), complementary findings (2) or contradictory findings (3) to direct observations and self-reported surveys. The user behaviour under investigation includes movement flows, patterns of occupancy, interactivity and interaction networks. Results suggest that variable degrees of overlap can be established between the two approaches with rather few comparable findings. For certain space usage behaviours high levels of variance between the automated and manual datasets are found, pointing towards predominantly complementary and contradictory findings. It is shown that the goodness of the fit between automated and manual data depends on the way data is aggregated. This allows systematic reflections on the strengths and weaknesses of each of the approaches. In summary, evidence suggests that both human and machine based data gathering reveal crucial insights into behaviours of building users. Substituting manual methods with automated ones cannot be supported by the data of this study. Further suggestions for future studies of social life in complex buildings are made, thus contributing to the development of research methods in the field.

Type: Proceedings paper
Title: Human versus machine - testing validity and insights of manual and automated data gathering methods in complex buildings
Event: 9th International Space Syntax Symposium
Location: Seoul, South Korea
Dates: 2013-10-31 - 2013-11-03
Open access status: An open access version is available from UCL Discovery
Publisher version: http://www.sss9.or.kr/paperpdf/mmd/SSS9_2013_REF05...
Keywords: space syntax, social network analysis, observation, movement, interaction, RFID
UCL classification: UCL
UCL > Provost and Vice Provost Offices > UCL BEAMS
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of the Built Environment
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of the Built Environment > The Bartlett School of Architecture
URI: https://discovery.ucl.ac.uk/id/eprint/1425880
Downloads since deposit
323Downloads
Download activity - last month
Download activity - last 12 months
Downloads by country - last 12 months

Archive Staff Only

View Item View Item