AUM2020 Session 5 - Modelling from micro to meso and macro scales

Duration: 2 hours 3 mins

Share this media item:

Embed this media item:

<iframe width="_width_" height="_height_" src="https://sms.cam.ac.uk/media/3324209/embed" frameborder="0" scrolling="no" allowfullscreen></iframe>

Choose size:

About this item

Available Formats

About this item

AUM2020 Session 5 - Modelling from micro to meso and macro scales's image

Description:	1. Enriching Survey Datasets Using Big Data and Machine Learning,with an Application to Transferring Attitudinal Variables across Transport Surveys Dr Patricia Mokhtarian, Georgia Institute of Technology 2. The emergence of an Urban Science: from microscopic individuals' behaviour to macroscopic universal urban scale laws and system of cities Prof. Francisco Javier Martinez, University of Chile Discussant: Prof. Ram Pendyala, Arizona State University Session chair: Dr Kaveh Jahanshahi, University of Cambridge

Description:

1. Enriching Survey Datasets Using Big Data and Machine Learning,with an Application to Transferring Attitudinal Variables across Transport Surveys
Dr Patricia Mokhtarian, Georgia Institute of Technology

2. The emergence of an Urban Science: from microscopic individuals' behaviour to macroscopic universal urban scale laws and system of cities
Prof. Francisco Javier Martinez, University of Chile

Discussant: Prof. Ram Pendyala, Arizona State University
Session chair: Dr Kaveh Jahanshahi, University of Cambridge


Created:	2020-11-06 16:03
Collection:	Martin Centre AUM2020: Modelling the New Urban World
Publisher:	University of Cambridge
Copyright:	The Martin Centre
Language:	eng (English)
Keywords:	AUM2020; Modelling the New Urban World; Martin Centre; Modelling from micro to meso and macro scales; Architecture;


Abstract:	1. Declining survey response rates make it increasingly critical for survey designers across disciplines to utilize mechanisms that facilitate timesaving on the part of respondents. In practice, this often means that questionnaires are shortened, yielding increased response rates but reduced information/variables available for modeling and forecasting purposes. We address this challenge using data-driven approaches such as machine learning within the context of the rapidly growing big data landscape, to develop and apply a transfer learning-based framework for integrating and enriching surveys, thereby expanding the amount of information available for use. 2. A new science of cities is emerging using the microeconomic model of individuals’ behaviour in a geographical context. Heterogeneous agents -individuals, households and firms- interact between then with social and economic objectives on the time and spatial dimensions. At a meso level such interaction requires time, spatial and prices matching, which means that a large number of interdependent markets takes place in the city. This configures a complex system of economic markets and social networks connected by transport of people and goods and communication systems. At a macro scale, cities a universal pattern described by scale laws, where their inputs and outputs increase non-linearly with the population size. This presentation describes a model of the complex urban system from the micro to the macro levels, with interesting properties such as the convergence of the system to attractors that explains universality of the scale laws. Its fractal structure allows the extension of the model from the single city to a structure of cities and regions. Building a coherent multi-scales model of cities is the foundation rock of an urban science, which in this case is built from the contribution of multiple disciplines.

Abstract:

1. Declining survey response rates make it increasingly critical for survey designers across disciplines to utilize mechanisms that facilitate timesaving on the part of respondents. In practice, this often means that questionnaires are shortened, yielding increased response rates but reduced information/variables available for modeling and forecasting purposes. We address this challenge using data-driven approaches such as machine learning within the context of the rapidly growing big data landscape, to develop and apply a transfer learning-based framework for integrating and enriching surveys, thereby expanding the amount of information available for use.

2. A new science of cities is emerging using the microeconomic model of individuals’ behaviour in a geographical context. Heterogeneous agents -individuals, households and firms- interact between then with social and economic objectives on the time and spatial dimensions. At a meso level such interaction requires time, spatial and prices matching, which means that a large number of interdependent markets takes place in the city. This configures a complex system of economic markets and social networks connected by transport of people and goods and communication systems. At a macro scale, cities a universal pattern described by scale laws, where their inputs and outputs increase non-linearly with the population size. This presentation describes a model of the complex urban system from the micro to the macro levels, with interesting properties such as the convergence of the system to attractors that explains universality of the scale laws. Its fractal structure allows the extension of the model from the single city to a structure of cities and regions. Building a coherent multi-scales model of cities is the foundation rock of an urban science, which in this case is built from the contribution of multiple disciplines.

Available Formats

Format	Quality	Bitrate	Size
MPEG-4 Video	640x360	988.03 kbits/sec	890.10 MB	View
WebM	640x360	343.61 kbits/sec	309.56 MB	View
iPod Video	480x360	489.95 kbits/sec	441.39 MB	View
MP3	44100 Hz	251.24 kbits/sec	226.34 MB	Listen
Auto *	(Allows browser to choose a format it supports)