Zoning of St. Petersburg Through the Prism of Social Activity Networks

Landsman, David; Kats, Philipp; Nenko, Aleksandra; Sobolevsky, Stanislav

doi:10.1016/j.procs.2020.11.014

Cited by 6 publications

(4 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Another use of LBSNs includes zoning spaces, either on classification or popularity (Cranshaw et al 2012;Zhou and Zhang 2016;Gupta et al 2018;Landsman et al 2020). Zhou and Zhang (2016) mine the Twitter and Foursquare data for Boston and Chicago for 2013.…”

Section: Introductionmentioning

confidence: 99%

Forecasting commercial real estate indicators under COVID-19 by adopting human activity using social big data

Taşcılar

Arslanlı

2022

Asia-Pac J Reg Sci

View full text Add to dashboard Cite

Dependence of the real estate sector on human activity has been unveiled during the COVID-19 pandemic. In addition, it is assumed that trends emitted from the location-based social networks (LBSNs) successfully reflect human activities, hence commercial property trends. This study examined the use of social media to forecast commercial real estate figures during COVID-19 in Istanbul and determined the potential of social media data for forecasting the future rent/price levels of retail properties. Instagram and Twitter, two major LBSN platforms, were selected as social media data sources. First, 17 million geo-tagged Instagram posts and 230 thousand geo-referenced tweets were collected. Then, the data sets were superposed on COVID-19 key points in Turkey and the relationships observed. Finally, the data sets were combined with the commercial real estate data to monitor increases in the accuracy of rent and price predictions. Beşiktaş District of Istanbul was chosen as the pilot region to test the methodology. The results showed that the LBSN-supported models outperformed baseline models most of the time for price predictions and occasionally for rent predictions. Also, both Instagram and Twitter were found essential to the study and could not be omitted. This study demonstrates the significance and leveraging potential of applying human activities to the decision-making processes of the commercial real estate sector under COVID-19 conditions. This is the first study to adopt LBSN data to forecast commercial property prices.

show abstract

Section: Introductionmentioning

confidence: 99%

Forecasting commercial real estate indicators under COVID-19 by adopting human activity using social big data

Taşcılar

Arslanlı

2022

Asia-Pac J Reg Sci

View full text Add to dashboard Cite

show abstract

“…Thus understanding the underlying community structure of the networks saw a wide range of applications, including social science (Plantié and Crampes 2013), biology (Guimerà and Nunes Amaral 2005), and economics (Piccardi and Tajoli 2012). In particular, partitioning the networks of human mobility and interactions is broadly applied to regional delineation (Ratti et al 2010;Blondel et al 2010;Sobolevsky et al 2013;Amini et al 2014;Hawelka et al 2014;Kang et al 2013;Belyi et al 2017;Grauwin et al 2017;Xu et al 2021) as well as urban zoning (Sobolevsky et al 2018;Landsman et al 2020Landsman et al , 2021.…”

Section: Introductionmentioning

confidence: 99%

Graph neural network inspired algorithm for unsupervised network community detection

Sobolevsky

Belyi

2022

Appl Netw Sci

Self Cite

View full text Add to dashboard Cite

Network community detection often relies on optimizing partition quality functions, like modularity. This optimization appears to be a complex problem traditionally relying on discrete heuristics. And although the problem could be reformulated as continuous optimization, direct application of the standard optimization methods has limited efficiency in overcoming the numerous local extrema. However, the rise of deep learning and its applications to graphs offers new opportunities. And while graph neural networks have been used for supervised and unsupervised learning on networks, their application to modularity optimization has not been explored yet. This paper proposes a new variant of the recurrent graph neural network algorithm for unsupervised network community detection through modularity optimization. The new algorithm’s performance is compared against the state-of-the-art methods. The approach also serves as a proof-of-concept for the broader application of recurrent graph neural networks to unsupervised network optimization.

show abstract

“…Thus understanding the underlying community structure of the networks saw a wide range of applications, including social science [1], biology [2], economics [3]. In particular, partitioning the networks of human mobility and interactions is broadly applied to regional delineation [4][5][6][7][8][9][10][11][12] as well as urban zoning [13][14][15].…”

Section: Introductionmentioning

confidence: 99%

Recurrent Graph Neural Network Algorithm for Unsupervised Network Community Detection

Sobolevsky

2021

Preprint

Self Cite

View full text Add to dashboard Cite

Network community detection often relies on optimizing partition quality functions, like modularity. This optimization appears to be a complex problem traditionally relying on discrete heuristics.And although the problem could be reformulated as continuous optimization, direct application of the standard optimization methods has limited efficiency in overcoming the numerous local extrema.However, the rise of deep learning and its applications to graphs offers new opportunities. And while graph neural networks have been used for supervised and unsupervised learning on networks, their application to modularity optimization has not been explored yet. This paper proposes a new variant of the recurrent graph neural network algorithm for unsupervised network community detection through modularity optimization. The new algorithm's performance is compared against a popular and fast Louvain method and a more efficient but slower Combo algorithm recently proposed by the author. The approach also serves as a proof-of-concept for the broader application of recurrent graph neural networks to unsupervised network optimization.

show abstract

Zoning of St. Petersburg Through the Prism of Social Activity Networks

Cited by 6 publications

References 24 publications

Forecasting commercial real estate indicators under COVID-19 by adopting human activity using social big data

Forecasting commercial real estate indicators under COVID-19 by adopting human activity using social big data

Graph neural network inspired algorithm for unsupervised network community detection

Recurrent Graph Neural Network Algorithm for Unsupervised Network Community Detection

Contact Info

Product

Resources

About