A Five-Star Guide for Achieving Replicability and Reproducibility When Working with GIS Software and Algorithms

Reference: John P. Wilson, Kevin Butler, Song Gao, Yingje Hu, Wenwen Li & Dawn J. Wright (2020) A Five-Star Guide for Achieving Replicability and Reproducibility When Working with GIS Software and AlgorithmsAnnals of the American Association of Geographers, DOI: 10.1080/24694452.2020.1806026

Abstract: The availability and use of geographic information technologies and data for describing the patterns and processes operating on or near the Earth’s surface have grown substantially during the past fifty years. The number of geographic information systems software packages and algorithms has also grown quickly during this period, fueled by rapid advances in computing and the explosive growth in the availability of digital data describing specific phenomena. Geographic information scientists therefore increasingly find themselves choosing between multiple software suites and algorithms to execute specific analysis, modeling, and visualization tasks in environmental applications today. This is a major challenge because it is often difficult to assess the efficacy of the candidate software platforms and algorithms when used in specific applications and study areas, which often generate different results. The subtleties and issues that characterize the field of geomorphometry are used here to document the need for (1) theoretically based software and algorithms; (2) new methods for the collection of provenance information about the data and code along with application context knowledge; and (3) new protocols for distributing this information and knowledge along with the data and code. This article discusses the progress and enduring challenges connected with these outcomes.

New Protocols for Distributing the Data and Code of Geospatial Research

Here, we propose a five-star practical guide for sharing data and code in geospatial research, modeled after the five-star system offered by Berners-Lee (2009) for publishing linked open data on the Web. Instead of asking researchers to share all pieces of data and code, this five-star guide encourages a simple start of data and code sharing, and researchers can move to a higher level when time and other resources allow.

See more papers on the Forum on Reproducibility and Replicability in Geography.

New IJGIS Editorial on Movement Data Science

There has not been a time in the history of GIScience when movement analytics and mobility insights have played such an important role in policymaking as in today’s global responses to the COVID-19 crisis. This special section further builds on previous efforts by the editorial team and others from the GIScience community and beyond to advance the body of knowledge in Computational Movement Analysis (CMA). CMA generally refers to series methods and analytical approaches to process, structure, visualize and analyze tracking data and movement patterns to facilitate knowledge discovery and modeling of movement. Specifically, this special section was proposed as part of a pre-conference workshop on Analysis of Movement Data (AMD 2018) at the GIScience 2018 meeting, 28 August 2018, Melbourne, Australia. The focus of this special section is on three aspects of CMA: (1) representation and modeling of movement; (2) urban mobility analytics; and (3) movement analytics using social media data. With the papers presented in the special section, we highlight recent advancements in CMA with the development of methods and techniques for big movement data analytics and utilization of trajectories constructed using user-generated crowdsourced contents such as geo-tagged social media posts. Traditional CMA methods were often developed and evaluated using a smaller set of movement data involving smaller numbers of individuals and contextual variables.

As the momentum to generate more geo-enriched movement data at large volumes, high frequencies and for longer durations continues, this is a timely and significant achievement towards movement data science. As the papers of this special section illustrate, movement data science leverages the advancements in big data analytics, cyberinfrastructure, parallel computing and data fusion to enhance the analysis of large, multi-faceted and multi-sourced movement data. Below are the editorial and the six original papers presented in this special section on the International Journal of Geographical Information Science (IJGIS). 

Dodge, S., Gao, S., Tomko, M., & Weibel, R. (2020). Progress in computational movement analysis – towards movement data science. International Journal of Geographical Information Science, 1-6.

Buchin, M., Kilgus, B., & Kölzsch, A. (2019). Group diagrams for representing trajectories. International Journal of Geographical Information Science, 1-33.

Graser, A., Widhalm, P., & Dragaschnig, M. (2020). The M³ massive movement model: a distributed incrementally updatable solution for big movement data exploration. International Journal of Geographical Information Science, 1-24.

Qiang, Y., & Xu, J. (2019). Empirical assessment of road network resilience in natural hazards using crowdsourced traffic data. International Journal of Geographical Information Science, 1-17.

Li, W., Wang, S., Zhang, X., Jia, Q., & Tian, Y. (2020). Understanding intra-urban human mobility through an exploratory spatiotemporal analysis of bike-sharing trajectories. International Journal of Geographical Information Science, 1-24.

Ma, D., Osaragi, T., Oki, T., & Jiang, B. (2020). Exploring the heterogeneity of human urban movements using geo-tagged tweets. International Journal of Geographical Information Science, 1-22.

Xin, Y., & MacEachren, A. M. (2020). Characterizing traveling fans: a workflow for event-oriented travel pattern analysis using Twitter data. International Journal of Geographical Information Science, 1-20.

Moving forward, we see a clear need for more reproducible research in CMA, following a growing mega-trend in data-driven sciences. Data quality and privacy challenges as well as uncertainty in data, analytics, and modeling have been largely overlooked in the CMA literature so far. For a more responsible movement data science, careful considerations should be given to the quality, uncertainty and representativeness of ‘large’ mobility data that are being used for generating important mobility insights for policymaking. Lastly, with the recent exciting developments in data access, as a community, we should think about leveraging this advantage to make movement data science more relevant to real-world problems for the mitigation of societal and environmental challenges such as disease outbreaks, population mobility, natural hazards and human-wildlife conflicts.

Link: GIScience 2020 Workshop on Advancing Movement Data Science (AMD’20) turns to 2021

Digital Contact Tracing and Surveillance: Geospatial opportunities, limitations, and research directions

Reference: Trisalyn Nelson, Peter Kedron,  Michael F. Goodchild,  Stewart Fotheringham,  Amy Frazier,  Wenwen Li, Song Gao, Yingjie Hu,  Ming-Hsiang Tsou, May Yuan, Bo Zhao (2020). Digital Contact Tracing and Surveillance: Geospatial opportunities, limitations, and research directions. ASU Spatial Analysis Research Center (SPARC) White Paper. pp 1-13.

Executive Summary

As efforts to mitigate and suppress COVID-19 continue, many decision makers are asking if digital contact tracing—a method for determining contact between an infected individual and others using tracking systems commonly based on mobile devices—can help us safely transition from population-wide social distancing to targeted case-based interventions such as individualized self-quarantine. In response, the Spatial Analysis Research Center (SPARC) at Arizona State University organized a panel of national experts to discuss the use of geospatial technologies in digital contact tracing and identify the practical challenges researchers can address to make digital contact tracing as effective as possible.

The major themes of the discussion included (i) the capabilities and limitations of geospatial technology, (ii) privacy, and (iii) future research directions. Key takeaways from each of these areas include:

Capabilities and limitations of geospatial technology: There are many geospatial technologies (e.g., GPS, Bluetooth, Cellular, WiFi) embedded in mobile devices that can be leveraged for digital contact tracing. However, GPS technology in smartphones lacks accuracy to map interactions in the detailed way one might expect. For instance, the horizontal accuracy of GPS is 15m, and the vertical accuracy is insufficient to pick up which floor of a building a person is on. Indoor accuracy is particularly poor, which is problematic given people spend 87% of their time indoors. However, information about the absolute location of an individual may not be as important to digitally tracing epidemiologically meaningful contacts as identifying the types of interactions most likely to result in the spread of the virus. The importance of tracing interactions creates an opportunity to use Bluetooth-based exchange of encrypted keys to record person-to-person contacts that can then be analyzed within the space-time prism framework. This approach will not require storing of all individuals’ movement data, which will reduce computation complexity. Geotargeted and geotagged social media are useful for tracking transmission between cities or within cities, detecting large gatherings, and helping individuals recall location and contact history during contact tracing interviews. Social media can also provide useful context, such as check-in locations and textual content, to reduce false positives in interactions identified through other forms of digital contact tracing.

Privacy: Digital contact tracing raises numerous privacy concerns. By creating some record of the location history or contacts of an individual, digital contact tracing creates an opportunity to identify an individual without their consent. At present, the privacy implications of digital contact tracing are unclear because these systems have yet to be fully developed or deployed in the US. An evaluation of pros and cons in the existing digital contact tracing plans operating in other countries can inform policy makers on privacy mediation during and after contact tracing. While companies and officials working on this issue have made statements that preserving privacy is an important goal, the details of how privacy will be preserved and the safeguards that will be put in place are not yet available. If any privacy protections are lifted to enable contact tracing, a plan should be put in place to restore protections once the pandemic subsides.

Future Research: To support digital contact tracing and surveillance, several research areas must be advanced. Key technical areas include increasing the accuracy of indoor positioning, developing approaches for reducing false positive of potential exposure (not to be confused with false negatives which are more common in COVID-19 diagnostic test) ensuring a focus on high accuracy in relative positioning, addressing computational complexities, developing group or bubble based approaches to surveillance, and developing a system for the creation and distribution of high resolution risk data and to enable self-determination of the need of quarantine and testing based on possible exposure. Research into how digital contact tracing systems link with existing contact tracing infrastructure and with other digital contact tracing systems also needs to be conducted. The implications of digital contact tracing for society and privacy will emerge along with these systems. Researchers need to study these issues as they emerge to ensure that we have the ability to hold an informed public debate about the effectiveness and costs of digital contact tracing.

Mobile location big data can help predict the potential infected areas as coronavirus spreads

The travels and close contact-tracing from/to infected communities is useful for identifying potential hotspots and assessing the potential risk across different places. A recent research published in Science showed that “substantial undocumented infection facilitates the rapid dissemination of novel coronavirus (SARS-CoV2) “. Understanding the human physical movement patterns and social contacts is a key for saving more lives as one may be surrounded by latent exposed people who don’t show SARS-CoV2 symptoms. Therefore, human mobility patterns and changes could be one indicator for understanding the status of physical social distancing. Here are the neighborhood mobility pattern and the Spring 2019 and March 2020 travel patterns for US cities and counties using the anonymized and aggregated mobile phone location big data in collaboration with SafeGraph, which covers over 3.6 million points of interest (POI) and business venues with visit patterns. Meanwhile, we are working on the whole US 2020 census block data and monitoring new infected areas from the CDC and from a list of Coronavirus dashboards in response to COVID-19.

Reference: Gao, S., Rao, J., Kang, Y., Liang, Y., & Kruse, J. (2020). Mapping county-level mobility pattern changes in the United States in response to COVID-19. SIGSPATIAL Special. 12(1), 16-26.

The spatial density distribution of over 3.6 million SafeGraph POIs with visit patterns.

The U.S. government, tech industry are discussing ways to harness smartphone location data to combat the novel coronavirus (COVID-19).

You can find out where people from those POIs / neighborhoods / a county connecting with other neighborhoods and counties across the US. By comparing the POI visits between last March and March 2020, we can summarize the changes and visualize the patterns on the maps to understand whether people in each County/State has reacted to (Physical) Social Distancing.

Interactive Web: Mapping Human Mobility Changes at the County level since March 1, 2020

Mapping Human Mobility Changes at the County level in March, 2020 (Data Source: Descartes Labs )
See also: https://www.nytimes.com/interactive/2020/03/23/opinion/coronavirus-economy-recession.html

Interactive Web on COVID-19 Physical Distancing and the relation with the infectious cases in Wisconsin (Using the latest SafeGraph weekly movement patterns in March 2020): https://geods.geography.wisc.edu/covid19/WI/

In addition, the maps below show the origin-destination (OD) flows larger than a travel frequency threshold at different spatial scales. The one at the urban scale can help understand the potential spread and hotspots in a city/metropolitan area.

Interactive Web: geods.geography.wisc.edu/covid19/King_WA.html (Notice that some paired neighborhood-to-neighborhood flows are not shown after the data filtering based on OD flow frequency)
Interactive Web: https://geods.geography.wisc.edu/covid19/King_US.html (neighborhood mobility to San Francisco, Alaska, Boston, and New York, etc. stand out. )

Spring Travel Risk

By using the county-level Spring travel data in March, we can see thousands of trips generated from the U.S. counties in the Spring season and widely across the U.S., which may help explain the rapid growth of infection cases across the whole U.S.

Interactive Web demo: Dane County, WI

The spring travels from the Dane County, WI with current U.S. confirmed cases: http://geods.geography.wisc.edu/covid19/WI_DaneCounty.html

Interactive Web: The King County, WA https://geods.geography.wisc.edu/covid19/KingCounty_Spring.html

Spring travels patterns aggregated at the Country-level in March 2019 from the people who reside in the King County, WA
Spring travels patterns aggregated at the County-level in March 2019 from the people who reside in the King County, WA
(Zoom in to the Pacific Coast Map) Spring travels patterns aggregated at the County-level in March 2019 from the people who reside in the King County, WA

The following table shows the top 20 counties which the people reside in the King County traveled to in March 2019.

And using the Country-to-US Counties flow data from last March, we can assess how the global travels from other countries outside of US will influence the potential coronavirus outbreak and spread in the US.

The spring international travels to US in March 2019. (filtered by at least 100 people)
The spring travels from China and Japan in March 2019.

Credit: The data sources were from SafeGraph Inc., Descartes Labs, and the Web geovisualization was created using the Kepler.gl tool.

Acknowledgment: We would like to thank all individuals and organizations for collecting and updating the COVID-19 observation data and reports. Dr. Song Gao acknowledges the funding support provided by the National Science Foundation (Award No. 2027375). Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.

New research article about Regional Economy and Transportation Network Analytics published in Scientific Reports

Bin Li, Song Gao, Yunlei Liang, Yuhao Kang, Timothy Prestby, Yuqi Gao, and Runmou Xiao. (2020) “Estimation of Regional Economic Development Indicator from Transportation Network Analytics.” Scientific Reports, 10(2647), 1-15. DOI: https://doi.org/10.1038/s41598-020-59505-2

Abstract: With the booming economy in China, many researches have pointed out that the improvement of regional transportation infrastructure among other factors had an important effect on economic growth. Utilizing a large-scale dataset which includes 3.5 billion entry and exit records of vehicles along highways generated from toll collection systems, we attempt to establish the relevance of mid-distance land transport patterns to regional economic status through transportation network analyses. We apply standard measurements of complex networks to analyze the highway transportation networks. A set of traffic flow features are computed and correlated to the regional economic development indicator. The multi-linear regression models explain about 89% to 96% of the variation of cities’ GDP across three provinces in China. We then fit gravity models using annual traffic volumes of cars, buses, and freight trucks between pairs of cities for each province separately as well as for the whole dataset. We find the temporal changes of distance-decay effects on spatial interactions between cities in transportation networks, which link to the economic development patterns of each province. We conclude that transportation big data reveal the status of regional economic development and contain valuable information of human mobility, production linkages, and logistics for regional management and planning. Our research offers insights into the investigation of regional economic development status using highway transportation big data.

Fig. Mapping the annual traffic volumes of cars and buses among cities in Shaanxi province from 2014 to 2017.
Fig. The Pearson’s correlation coefficients between city GDP value, betweenness, closeness centrality measures and the PageRank index in transport flow networks of cars & buses (C) and trucks (K) in three provinces. The Betw (D) and Closeness (D) measures are calculated using the spatial interaction networks of cities with the inter-city distances as edge weights.

Fig. Mapping the annual traffic volumes of trucks among cities in Jiangsu province in 2017.
Fig. Mapping the annual traffic volumes of trucks among cities in Liaoning province in 2017.

Geomasking techniques for protecting the location privacy of social media users

Figure 1: The spatial distribution of geotagged tweets around a Twitter user’s home.

Reference: Song Gao, Jinmeng Rao, Xinyi Liu, Yuhao Kang, Qunying Huang, Joseph App. (2019) Exploring the effectiveness of geomasking techniques for protecting the geoprivacy of Twitter users. Journal of Spatial Information Science. 19, 105-129. DOI: 10.5311/JOSIS.2019.19.510 [PDF]

Abstract: With the ubiquitous use of location-based services, large-scale individual-level location data has been widely collected through location-awareness devices. Geoprivacy concerns arise on the issues of user identity de-anonymization and location exposure. In this work, we investigate the effectiveness of geomasking techniques for protecting the geoprivacy of active Twitter users who frequently share geotagged tweets in their home and work locations. By analyzing over 38,000 geotagged tweets of 93 active Twitter users in three U.S. cities (Los Angeles, Madison, and Washington D.C.), the two-dimensional Gaussian masking technique with proper standard deviation settings is found to be more effective to protect user’s location privacy while sacrificing geospatial analytical resolution than the random perturbation masking method and the aggregation on traffic analysis zones. Furthermore, a three-dimensional theoretical framework considering privacy, spatial analytics, and uncertainty factors simultaneously is proposed to assess geomasking techniques. Our research offers insights into geoprivacy concerns of social media users’ georeferenced data sharing for future development of location-based applications and services.

Figure 2: The Gaussian geomasking with different standard deviations (SD) and the random perturbation with 1km and 2km threshold of a user’s geotagged tweets.
Figure 10: The violin plot of distance shifts of tweet locations after geomasking.
Figure 11: A 3D-cube framework for assessing different geomasking techniques; the position of each method is estimated from the results of our case study.

Broader Impacts: In fact, Twitter removes support for precise geotagging since June, 2019. However, the metadata of historical tweets prior to the policy change may still reveal precise GPS coordinates. In addition, when a user deletes a geotagged tweet , Twitter does not guarantee the information will be completely removed from all copies of the data on third-party applications or in external search results. Even if the precise GPS location is not available anymore, Twitter users are still able to add place tags (e.g., a city, office building, apartment, landmark, and many other types of places) to their geotagged tweets, which can be converted to the GPS coordinates (often using the centroid as a representation location). This is similar to the aforementioned aggregation-based masking approach, thus we may still be able to get users’ sensitive locations based on fine-scale place tags. People should be aware that sharing or publishing such kind of location data involve geoprivacy issues and the geomasking technique provides a way to help mitigate the problem not only for Twitter users but also for other telematics and social media platforms such as Facebook, Flickr, Weibo, and Instagram where geotagging or place-tagging is accessible, as well as for mobile applications that track individual locations.

GeoAI at ACM SIGSPATIAL: progress, challenges, and future directions

Geospatial artificial intelligence (GeoAI) is an interdisciplinary field that has received tremendous attention from both academia and industry in recent years. We recently published an article that reviews the series of GeoAI workshops held at the Association for Computing Machinery (ACM) International Conference on Advances in Geographic Information Systems (SIGSPATIAL) since 2017. These workshops have provided researchers a forum to present GeoAI advances covering a wide range of topics, such as geospatial image processing, transportation modeling, public health, and digital humanities. We provide a summary of these topics and the research articles presented at the 2017, 2018, and 2019 GeoAI workshops. We conclude with a list of open research directions for this rapidly advancing field.

Reference: Yingjie Hu, Song Gao, Dalton Lunga, Wenwen Li, Shawn Newsam, and Budhendra Bhaduri (2019): GeoAI at ACM SIGSPATIAL: progress, challenges, and future directions, ACM SIGSPATIAL Special, 11(2), 5-15. [PDF]

ACM SIGSPATIAL GeoAI Workshop Proceedings:

1st ACM SIGSPATIAL International Workshop AI and Deep Learning for Geographic Knowledge Discovery (GeoAI’17). Redondo Beach, CA, USA – November 7, 2017. DOI: 10.1145/3178392.3178408 [PDF]

2nd ACM SIGSPATIAL International Workshop AI and Deep Learning for Geographic Knowledge Discovery (GeoAI’18). Seattle, WA, USA – November 6, 2018. DOI: 10.1145/3307599.3307609[PDF]

3rd ACM SIGSPATIAL International Workshop AI and Deep Learning for Geographic Knowledge Discovery (GeoAI’19). Chicago, IL, USA – November 5, 2019. DOI: 10.1145/3356471 [PDF]

New paper about investigating urban metro stations as cognitive places

New paper published in the journal of Cities.

Kang Liu, Peiyuan Qiu, Song Gao, Feng Lu, Jincheng Jiang, Ling Yin. (2020) Investigating urban metro stations as cognitive places in cities using points of interest. Cities. 97, 102561, 1-13. DOI: 10.1016/j.cities.2019.102561

Link: Map story presentation (open in a Web browser on PC or laptop)

Fig. 1. The proposed framework for extracting and understanding the cognitive regions of urban metro stations.

Abstract: The significance of urban metro stations extends beyond their roles as transport nodes in a city. Their surroundings are usually well developed and attract a lot of human activities, which make the metro station areas important cognitive places characterized by vague boundaries and rich semantics. Current studies mainly define metro station areas based on an estimation of walking distance to the stations (e.g., 700 m) and investigate these areas from the perspectives of transportation and land use instead of as cognitive places perceived by the crowd. To fill this gap, this study proposes a novel framework for extracting and understanding the cognitive regions of urban metro stations based on points of interest (POIs). First, we extract the cognitive regions of metro stations based on co-occurrence patterns of the stations and their surrounding POIs on web pages by proposing a cohesive approach combined of spatial clustering, web page extraction, knee-point detection, and polygon generation techniques. Second, we identify the semantics of metro stations based on POI types inside the regions using the term frequency-inverse document frequency (TF-IDF) method. In total 166 metro stations along with more than one million POIs in Shenzhen, China are utilized as data sources of the case study. The results indicate that our proposed framework can well detect the place characteristics of urban metro stations, which enriches the place-based GIS research and provides a human-centric perspective for urban planning and location-based-service (LBS) applications.

Implications for urban planning

As Kevin Lynch stated in The Image of the City (Lynch, 1960), the skeleton of individuals’ mental images is formed by five types of elements in the city: paths, edges, nodes, districts and landmarks, which mediates in the interaction between humans and their environment. The first thing we want to emphasize in this study is that urban metro stations are also one type of such cognitive elements (i.e., landmarks) in cities; their properties as cognitive places should be considered in urban planning and design so as to match people’s cognition. In addition, our extracted cognitive regions of urban metro stations show diverse and irregular shapes, which indicates that unified physical distances frequently used in existing studies and planning practices cannot precisely define TOD precincts perceived by humans. To this end, what we suggest in this study is that urban planning practices should attach importance to “cognitive place” and “cognitive distance”, which load human experiences and perceptions toward the environments (Briggs, 1973; Montello, 1991). This is also coincident with the ultimate goal of urban planning, urban design, and smart-city construction, i.e., making better human societies and improving human lives (Shaw & Sui, 2019).

New IJGIS Editorial on GeoAI

Abstract: What is the current state-of-the-art in integrating results from artificial intelligence research into geographic information science and the earth sciences more broadly? Does GeoAI research contribute to the broader field of AI, or does it merely apply existing results? What are the historical roots of GeoAI? Are there core topics and maybe even moonshots that jointly drive this emerging community forward? In this editorial, we answer these questions by providing an overview of past and present work, explain how a change in data culture is fueling the rapid growth of GeoAI work, and point to future research directions that may serve as common measures of success.

Moonshot (Editorial): Can we develop an artificial GIS analyst that passes a domain-specific Turing Test by 2030?

Keywords: Spatial Data Science, GeoAI, Machine Learning, Knowledge Graphs, Geo-Semantics, Data Infrastructure

Acknowledgement: we sincerely thank all the reviewers who contribute their time to the peer-review process and ensure the quality of the accepted papers.

Special Issue Papers (up to date):

Janowicz, K., Gao, S., McKenzie, G., Hu, Y., and Bhaduri, B. (2020, Editorial). GeoAI: Spatially Explicit Artificial Intelligence Techniques for Geographic Knowledge Discovery and Beyond. International Journal of Geographical Information Science, 34(4), 625-636.

Acheson, E., Volpi, M., & Purves, R. S. (2020). Machine learning for cross-gazetteer matching of natural features. International Journal of Geographical Information Science, 1-27.

Duan, W., Chiang, Y., Leyk, S., Uhl, J. and Knoblock, C. (2020). Automatic alignment of contemporary vector data and georeferenced historical maps using reinforcement learning. International Journal of Geographical Information Science, forthcoming. 1-27; DOI: 10.1080/13658816.2019.1698742.

Guo, Z., & Feng, C. C. (2020). Using multi-scale and hierarchical deep convolutional features for 3D semantic classification of TLS point clouds. International Journal of Geographical Information Science, 1-20.

Law, S., Seresinhe, C. I., Shen, Y., & Gutierrez-Roig, M. (2020). Street-Frontage-Net: urban image classification using deep convolutional neural networks. International Journal of Geographical Information Science, 1-27.

Li, W., & Hsu, C. Y. (2020). Automated terrain feature identification from remote sensing imagery: a deep learning approach. International Journal of Geographical Information Science, 1-24.

Ren, Y., Chen, H., Han, Y., Cheng, T., Zhang, Y., & Chen, G. (2020). A hybrid integrated deep learning model for the prediction of citywide spatio-temporal flow volumes. International Journal of Geographical Information Science, 1-22.

Sparks, K., Thakur, G., Pasarkar, A., & Urban, M. (2020). A global analysis of cities’ geosocial temporal signatures for points of interest hours of operation. International Journal of Geographical Information Science, 1-18.

Xie, Y., Cai, J., Bhojwani, R., Shekhar, S., & Knight, J. (2020). A locally-constrained YOLO framework for detecting small and densely-distributed building footprints. International Journal of Geographical Information Science, 1-25.

Zhu, D., Cheng, X., Zhang, F., Yao, X., Gao, Y., & Liu, Y. (2020). Spatial interpolation using conditional generative adversarial neural networks. International Journal of Geographical Information Science, 1-24.

The fusion of knowledge-driven and data-driven approaches to discovering urban functional regions

Papadakis, E., Gao, S., & Baryannis, G. (2019). Combining Design Patterns and Topic Modeling to Discover Regions Supporting Particular Functionality. ISPRS International Journal of Geo-Information. 8(9), 385; https://doi.org/10.3390/ijgi8090385.


The problem of discovering regions that support particular functionalities in an urban setting has been approached in literature using two general methodologies: top-down, encoding expert knowledge on urban planning and design and discovering regions that conform to that knowledge; and bottom-up, using data to train machine learning models, which can discover similar regions. Both methodologies face limitations, with knowledge-based approaches being criticized for scalability and transferability issues and data-driven approaches for lacking interpretability and depending heavily on data quality.

To mitigate these disadvantages, we propose a novel framework that fuses a knowledge-based approach using design patterns and a data-driven approach using latent Dirichlet allocation (LDA) topic modeling in three different ways: Functional regions discovered using either approach are evaluated against each other to identify cases of significant agreement or disagreement; knowledge from patterns is used to adjust topic probabilities in the learning model; and topic probabilities are used to adjust pattern-based results. The proposed methodologies are demonstrated through the use case of identifying shopping-related regions in the Los Angeles metropolitan area. Results show that the combination of pattern-based discovery and topic modeling extraction helps uncover discrepancies between the two approaches and smooth inaccuracies caused by the limitations of each approach.

Figure. The proposed framework of fusing knowledge-based and data-driven approaches
Figure. Extracted shopping regions by combining data-to-knowledge and knowledge-to-data approaches.