Zhand, S;
Karimi, K;
(2022)
Using space syntax method to train a model for unsupervised detection of socio-economic conditions - the case of metropolitan area of Tehran.
In:
Proceedings 13th International Space Syntax Symposium, SSS 2022.
(In press).
Preview |
Text
527zhand.pdf - Published Version Download (11MB) | Preview |
Abstract
The availability of open-source data, coupled with recent advances in technology has made it easier to create large scale urban and regional models used in the field of environmental studies and specifically space syntax. With the use of open data, large scale regional road-centre line models (Turner, 2005) can now be created and processed to explain the spatial configuration as well as the structure of the built environment. While the study of the socio-economic condition of the built environment correlated with the configuration of the space has been the general use of these models, there has been less focus on multi-layered analysis and metrics across a large model. On another hand, with restrictions on datasets from formal resources, the conventional use of space syntax theories and methods are limited. However, incorporating more advanced methods of quantitative analysis, space syntax can compensate for the lack of available formal dataset in reading and/or predicting environmental phenomena. Given that with the available data sources such as Open Street Map, consistent spatial network models are available, RCL segment models can be trained to predict the socio-economic condition of areas where the formal data is not obtainable. This research puts forward a workflow through which, the spatial network model can be used to train a model that predicts mentioned phenomena. This workflow uses a large segment model of the metropolitan area of Tehran and uses the centrality measures from space syntax analysis to train an unsupervised model which can predict possible missing information. It also assesses the efficacy of the model and shows to what extent the model is to be trusted and what the shortcomings of the model are. It is shown that although the models are very efficient in predicting the required conditions there should be a supervised assessment on the parameters of the algorithms to optimize the outcome.
| Type: | Proceedings paper |
|---|---|
| Title: | Using space syntax method to train a model for unsupervised detection of socio-economic conditions - the case of metropolitan area of Tehran |
| Open access status: | An open access version is available from UCL Discovery |
| Publisher version: | https://www.spacesyntax.net/symposia/ |
| Language: | English |
| Additional information: | This version is the version of record. For information on re-use, please refer to the publisher’s terms and conditions. |
| 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-pp.ucl.ac.uk/id/eprint/10163400 |
Archive Staff Only
 |
View Item |
