Standard

A Convolutional Neural Network-Based Model for Predicting The Perceived Attractiveness of Urban Places. / Psyllidis, Achilleas; Choiri, Hendra Hadhil.

2018. 1-1 Abstract from American Association of Geographers Annual Meeting 2018, New Orleans, United States.

Research output: Contribution to conferenceAbstractScientific

Harvard

Psyllidis, A & Choiri, HH 2018, 'A Convolutional Neural Network-Based Model for Predicting The Perceived Attractiveness of Urban Places' American Association of Geographers Annual Meeting 2018, New Orleans, United States, 10/04/18 - 14/04/18, pp. 1-1.

APA

Psyllidis, A., & Choiri, H. H. (2018). A Convolutional Neural Network-Based Model for Predicting The Perceived Attractiveness of Urban Places. 1-1. Abstract from American Association of Geographers Annual Meeting 2018, New Orleans, United States.

Vancouver

Psyllidis A, Choiri HH. A Convolutional Neural Network-Based Model for Predicting The Perceived Attractiveness of Urban Places. 2018. Abstract from American Association of Geographers Annual Meeting 2018, New Orleans, United States.

Author

Psyllidis, Achilleas ; Choiri, Hendra Hadhil. / A Convolutional Neural Network-Based Model for Predicting The Perceived Attractiveness of Urban Places. Abstract from American Association of Geographers Annual Meeting 2018, New Orleans, United States.1 p.

BibTeX

@conference{44da4f2fd7f34b908086bfbbf5952824,
title = "A Convolutional Neural Network-Based Model for Predicting The Perceived Attractiveness of Urban Places",
abstract = "An understanding of how people perceive attractive or unattractive places in cities is vitally important to urban planning and policy making. Given the subjective nature of human perception and the ambiguous character of attractiveness as an attribute of urban places, it is challenging to quantify and reliably assess the extent to which a place is perceived as attractive or not. It is even more difficult to do this at scale. This research proposes a novel machine learning approach to quantifying and predicting the perceived attractiveness of urban places. It introduces a predictive model, employing a Convolutional Neural Network (CNN), to automatically estimate the attractiveness of places in cities, based on their Google Street View representation. A set of street-level images (four consecutive images make up the panoramic overview of each place) with similar daylight conditions and level of complexity (e.g. amount of people present in a scene) is assessed by means of crowdsourcing, drawing on attractiveness-related factors identified in environmental psychology studies. Using these judgments as ground truth, in combination with a new CNN architecture, the model automatically assesses the perceived attractiveness of any place in a city, by rating them on the basis of a five-point Likert scale score. Moreover, it identifies features of the urban environment that could influence positively or negatively the overall attractiveness of a place. The resulting accuracy of 55.9{\%} and root-mean-square error of 0.70 illustrate that the model holds promise as a scalable and reliable tool for estimating the perceived attractiveness of urban places.",
keywords = "predictive urban analytics, machine learning, convolutional neural networks, urban attractiveness, spatial data science",
author = "Achilleas Psyllidis and Choiri, {Hendra Hadhil}",
year = "2018",
language = "English",
pages = "1--1",
note = "American Association of Geographers Annual Meeting 2018, AAG 2018 ; Conference date: 10-04-2018 Through 14-04-2018",

}

RIS

TY - CONF

T1 - A Convolutional Neural Network-Based Model for Predicting The Perceived Attractiveness of Urban Places

AU - Psyllidis, Achilleas

AU - Choiri, Hendra Hadhil

PY - 2018

Y1 - 2018

N2 - An understanding of how people perceive attractive or unattractive places in cities is vitally important to urban planning and policy making. Given the subjective nature of human perception and the ambiguous character of attractiveness as an attribute of urban places, it is challenging to quantify and reliably assess the extent to which a place is perceived as attractive or not. It is even more difficult to do this at scale. This research proposes a novel machine learning approach to quantifying and predicting the perceived attractiveness of urban places. It introduces a predictive model, employing a Convolutional Neural Network (CNN), to automatically estimate the attractiveness of places in cities, based on their Google Street View representation. A set of street-level images (four consecutive images make up the panoramic overview of each place) with similar daylight conditions and level of complexity (e.g. amount of people present in a scene) is assessed by means of crowdsourcing, drawing on attractiveness-related factors identified in environmental psychology studies. Using these judgments as ground truth, in combination with a new CNN architecture, the model automatically assesses the perceived attractiveness of any place in a city, by rating them on the basis of a five-point Likert scale score. Moreover, it identifies features of the urban environment that could influence positively or negatively the overall attractiveness of a place. The resulting accuracy of 55.9% and root-mean-square error of 0.70 illustrate that the model holds promise as a scalable and reliable tool for estimating the perceived attractiveness of urban places.

AB - An understanding of how people perceive attractive or unattractive places in cities is vitally important to urban planning and policy making. Given the subjective nature of human perception and the ambiguous character of attractiveness as an attribute of urban places, it is challenging to quantify and reliably assess the extent to which a place is perceived as attractive or not. It is even more difficult to do this at scale. This research proposes a novel machine learning approach to quantifying and predicting the perceived attractiveness of urban places. It introduces a predictive model, employing a Convolutional Neural Network (CNN), to automatically estimate the attractiveness of places in cities, based on their Google Street View representation. A set of street-level images (four consecutive images make up the panoramic overview of each place) with similar daylight conditions and level of complexity (e.g. amount of people present in a scene) is assessed by means of crowdsourcing, drawing on attractiveness-related factors identified in environmental psychology studies. Using these judgments as ground truth, in combination with a new CNN architecture, the model automatically assesses the perceived attractiveness of any place in a city, by rating them on the basis of a five-point Likert scale score. Moreover, it identifies features of the urban environment that could influence positively or negatively the overall attractiveness of a place. The resulting accuracy of 55.9% and root-mean-square error of 0.70 illustrate that the model holds promise as a scalable and reliable tool for estimating the perceived attractiveness of urban places.

KW - predictive urban analytics

KW - machine learning

KW - convolutional neural networks

KW - urban attractiveness

KW - spatial data science

M3 - Abstract

SP - 1

EP - 1

ER -

ID: 53255745