Article
Computer Science, Information Systems
Jiannan Cai, Mei-Po Kwan
Summary: A novel detection method based on frequent-pattern mining and spatial statistics is proposed for discovering spatial flow co-location patterns, establishing statistical significance, and verifying the advantages of the method through synthetic experiments.
INTERNATIONAL JOURNAL OF GEOGRAPHICAL INFORMATION SCIENCE
(2022)
Article
Geography
Jacob Kruse, Song Gao, Yuhan Ji, Daniel P. Szabo, Kenneth R. Mayer
Summary: Redistricting is an important process in political elections. This paper proposes using spatial interaction to quantify the degree to which districts capture coherent communities of interest (COIs), and presents a regionalization algorithm based on Markov chain for evaluating redistricting plans. The experiments demonstrate the effectiveness of the proposed methods in redistricting.
CARTOGRAPHY AND GEOGRAPHIC INFORMATION SCIENCE
(2023)
Article
Computer Science, Interdisciplinary Applications
Jiannan Cai, Mei-Po Kwan
Summary: Spatial flow outlier (SFO) detection aims to discover spatial flows with significantly different non-spatial attribute values from their neighborhoods. This study proposes a spatial-autocorrelation-aware detection method that tests the local difference to detect SFOs and constructs a distribution-free model as the null hypothesis.
COMPUTERS ENVIRONMENT AND URBAN SYSTEMS
(2022)
Article
Health Care Sciences & Services
Iris Hendrickx, Tim Voets, Pieter van Dyk, Rudolf B. Kool
Summary: This study explored the use of text mining techniques to analyze patient complaint databases in order to identify potential patient safety problems at health care providers and automatically predict the severity of complaints. The research found that a simple text classification approach using bag-of-words feature representation worked best for severity prediction of complaints, achieving high accuracy rates on the test set.
JOURNAL OF MEDICAL INTERNET RESEARCH
(2021)
Article
Geography
Luc Anselin, Xun Li, Julia Koschinsky
Summary: The GeoDa software has evolved from a closed-source Windows-only solution to an open-source cross-platform product as a library (libgeoda) which can be integrated into other software environments. Two empirical examples were used to investigate local clustering and socioeconomic health determinants, while a timing experiment demonstrated competitive performance compared to established solutions in R and Python.
GEOGRAPHICAL ANALYSIS
(2022)
Article
Green & Sustainable Science & Technology
Trisalyn A. Nelson, Colin Ferster, Avipsa Roy, Meghan Winters
Summary: Cities are investing in infrastructure to improve the safety and attractiveness of cycling. However, using data to support decision making and ensure representativeness is a challenge. This study applies ecological classification methods to diverse spatial data on the built environment, communities, and bicycling, in order to classify street and path segments and map streetscape categories. The approach is piloted in Ottawa, Canada and demonstrates how streetscape categories can be used for monitoring, safety, and infrastructure interventions.
INTERNATIONAL JOURNAL OF SUSTAINABLE TRANSPORTATION
(2023)
Article
Computer Science, Information Systems
Jakub Nowosad, Tomasz F. Stepinski
Summary: This paper introduces a method called Integrated Co-occurrence Matrix (INCOMA) for unsupervised identification and mapping of multi-thematic categorical patterns in landscape types (LTs). By numerical representation of local landscapes using INCOMA signatures and calculating dissimilarities between them, LTs can be identified and mapped using standard clustering or segmentation techniques. The resulting LTs are typically heterogeneous with respect to categories of contributing themes, reflecting human perception of the landscape.
INTERNATIONAL JOURNAL OF GEOGRAPHICAL INFORMATION SCIENCE
(2021)
Article
Geosciences, Multidisciplinary
Manh Xuan Trinh, Frank Molkenthin
Summary: This study demonstrates the successful simulation of flooding in the Tra Bong River basin in Vietnam by integrating multiple models and data sources. It shows that applying advanced hydrodynamic models to analyze flood inundation and hazard in data-scarce river basins is feasible, providing a valuable approach for understanding flooding and assessing flood risk in ungauged areas.
Article
Computer Science, Information Systems
Gebeyehu Belay Gebremeskel, Birhanu Hailu, Belete Biazen
Summary: This paper proposes a new approach to optimize data mining modeling for visualization of knowledge extraction, which can handle and analyze large-scale data and deep understand data mining modeling techniques. This approach is of great importance in the field of healthcare.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2022)
Article
Surgery
Zhi Ven Fong, Daniel A. Hashimoto, Ginger Jin, Alex B. Haynes, Numa Perez, Motaz Qadan, Cristina R. Ferrone, Carlos Fernandez-del Castillo, Andrew L. Warshaw, Keith D. Lillemoe, Lara N. Traeger, David C. Chang
Summary: This study found that limiting access to low-volume pancreatectomy hospitals would increase patients' round-trip driving time by 24 minutes, but up to 54 minutes for 25% of patients. Population mortality rates may improve by 1.5%.
Article
Health Care Sciences & Services
Ravi Aggarwal, Soma Farag, Guy Martin, Hutan Ashrafian, Ara Darzi
Summary: The survey highlighted low levels of AI knowledge among participants, with most being comfortable sharing health data with the NHS or universities but less so with commercial organizations. The majority supported AI research on health care data and imaging in a university setting, provided that concerns about privacy and consent processes were addressed.
JOURNAL OF MEDICAL INTERNET RESEARCH
(2021)
Article
Health Care Sciences & Services
Sara C. Handley, Molly Passarella, Sindhu K. Srinivas, Scott A. Lorch
Summary: The study developed an empiric approach to determine a hospital's level of maternal care using administrative data and described the levels of maternal care available over time. High-risk patients were more likely to deliver in hospitals with higher level maternal care, and the proportion of hospitals with high-level maternal care increased over time.
BMC HEALTH SERVICES RESEARCH
(2021)
Article
Environmental Sciences
Indishe P. Senanayake, Anthony S. Kiem, Gregory R. Hancock, Vaclav Metelka, Chris B. Folkes, Phillip L. Blevin, Anthony R. Budd
Summary: Mineral prospectivity mapping is crucial for discovering new economic mineral deposits, but it often requires significant costs, time, and human resources. This study utilized an ensemble machine learning approach with geoscience datasets to map Cu-Au and Pb-Zn mineral prospectivity in the Cobar Basin, NSW, Australia. The results showed improved classification accuracy compared to existing methods and demonstrated the potential for this approach to serve as a preliminary evaluation technique, providing guidance for more detailed geological investigations in other regions.
Article
Multidisciplinary Sciences
Masaki Kotsubo, Tomoki Nakaya
Summary: Understanding spatial interactions such as human mobility has long been a major focus in geography, spatial economics, and traffic engineering. A new formulation of intervening opportunities with a kernel function has been proposed to improve the radiation model, leading to better predictions of inter-regional flows. The modified radiation model, incorporating kernel-based intervening opportunities, outperformed the original model when fitted to four datasets.
SCIENTIFIC REPORTS
(2021)
Article
Public, Environmental & Occupational Health
Laura I. L. Poulin, Mark W. Skinner, Mary T. Fox
Summary: This study examines the influence of patient flow prioritization on older adult care and highlights the undervaluation of older patients' needs and local contexts. Considering the spatial and temporal dimensions of older adult care has significant implications for future research, policy, and practice.
SOCIAL SCIENCE & MEDICINE
(2023)
Article
Computer Science, Information Systems
Caglar Koylu, Diansheng Guo, Yuan Huang, Alice Kasakoff, Jack Grieve
Summary: The study collected and cleaned 92,832 user-contributed family trees with 250 million individuals, creating a population-scale and longitudinal dataset that showed biases in data and high mobility among individuals.
INTERNATIONAL JOURNAL OF GEOGRAPHICAL INFORMATION SCIENCE
(2021)
Article
Computer Science, Information Systems
Fengli Xu, Zhen Tu, Hongjia Huang, Shuhao Chang, Funing Sun, Diansheng Guo, Yong Li
Summary: The study suggests that check-in records can be vulnerable to privacy attacks and proposes a new privacy criterion to protect user privacy.
IEEE TRANSACTIONS ON MOBILE COMPUTING
(2021)
Article
Green & Sustainable Science & Technology
Seda Ertan, Rahmi Nurhan Celik
Summary: The study introduced different sustainable drainage solutions, such as green infrastructure, to mitigate flood hazards and found that implementing greener infrastructure can reduce runoff coefficient, peak flowrate, flood inundation area, and the number of structures affected by flood risk.
Article
Computer Science, Information Systems
Tong Xia, Junjie Lin, Yong Li, Jie Feng, Pan Hui, Funing Sun, Diansheng Guo, Depeng Jin
Summary: The article introduces the 3-Dimensional Graph Convolution Network (3DGCN) framework for predicting citywide crowd flow, achieving superior performance compared to state-of-the-art baselines. By modeling dynamic spatio-temporal graph prediction problems and learning urban structures, the accuracy of predictions is enhanced.
ACM TRANSACTIONS ON KNOWLEDGE DISCOVERY FROM DATA
(2021)
Article
Geography
Caglar Koylu, Geng Tian, Mary Windsor
Summary: FlowMapper.org is a web-based framework that automates the production and design of origin-destination flow maps. It allows users to upload and process their own data, customize flow maps, and supports supplementary layers.
Article
Computer Science, Information Systems
Mete Ercan Pakdil, Rahmi Nurhan celik
Summary: This paper explores the utilization of serverless technologies for geospatial data processes and proposes a system design and architecture for handling complex geospatial data processing tasks. It also introduces new models for workflow and task definitions, as well as web services based on the Open Geospatial Consortium (OGC) API Processes specification to enhance interoperability with other geospatial applications.
ISPRS INTERNATIONAL JOURNAL OF GEO-INFORMATION
(2022)
Article
Geography
Caglar Koylu, Alice Kasakoff
Summary: Studying migration over a long period presents challenges, but crowd-sourced family tree data can provide valuable information for analyzing population dynamics and migration. This article introduces a methodology to measure and map long-term changes in migration flows using population-scale family-tree data, and applies it to study internal migration in the continental United States between 1789 and 1924.
CARTOGRAPHY AND GEOGRAPHIC INFORMATION SCIENCE
(2022)
Article
Computer Science, Information Systems
Jie Feng, Yong Li, Ziqian Lin, Can Rong, Funing Sun, Diansheng Guo, Depeng Jin
Summary: This article proposes a deep learning-based convolutional model, DeepSTN+, for predicting crowd flows in different regions of a city. The model utilizes spatial dependence, time factor, and prior knowledge to improve performance through a stable training process.
ACM TRANSACTIONS ON KNOWLEDGE DISCOVERY FROM DATA
(2022)
Article
Criminology & Penology
James C. Wo, Ethan M. Rogers, Mark T. Berg, Caglar Koylu
Summary: The present study investigates the relationship between Twitter-derived measures and crime rates across block groups, finding that density and proportion of Twitter users and tweets are associated with crime counts.
CRIME & DELINQUENCY
(2022)
Article
Public, Environmental & Occupational Health
Zhuo Tang, Margaret Carrel, Caglar Koylu, Andrew Kitchen
Summary: Studies have shown that human ecological factors greatly influence the circulation of H5N1 avian influenza virus. Human cultural landscapes and human-induced factors can facilitate the spread of the virus between geographically distant areas, while physical and cultural barriers may impede its movement between adjacent regions.
Article
Geography
Hoeyun Kwon, Caglar Koylu, Bryce J. Dietrich
Summary: The article introduces a workflow that combines natural language processing, spatial time series analysis, and geovisualization techniques to identify and visualize the variations in sentiment trends on Twitter across different geographic regions and topics. By examining the 2016 presidential debates as a case study, distinct temporal patterns in sentiment distributions across various topics and states were uncovered.
CARTOGRAPHY AND GEOGRAPHIC INFORMATION SCIENCE
(2023)
Article
Computer Science, Interdisciplinary Applications
Yingjing Huang, Fan Zhang, Yong Gao, Wei Tu, Fabio Duarte, Carlo Ratti, Diansheng Guo, Yu Liu
Summary: This study proposes a deep learning-based module called Vision-LSTM, which can obtain vector representation from varying numbers of street-level images. The module is validated to effectively recognize urban villages by combining street-level imagery with remote sensing imagery and social sensing data. Compared to existing image fusion methods, Vision-LSTM demonstrates significant effectiveness in capturing associations between street-level images.
COMPUTERS ENVIRONMENT AND URBAN SYSTEMS
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Mingyang Zhang, Yong Li, Funing Sun, Diansheng Guo, Pan Hui
Summary: The paper introduces an Adaptive Spatio-Temporal Convolutional Network (ASTCN) to address the spatial and temporal dependencies in traffic prediction. By utilizing a spatial graph learning module and an adaptive temporal convolution module, the ASTCN effectively captures dynamic relationships and complex dependencies in traffic data.
2021 21ST IEEE INTERNATIONAL CONFERENCE ON DATA MINING (ICDM 2021)
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Tong Xia, Yunhan Qi, Jie Feng, Fengli Xu, Funing Sun, Diansheng Guo, Yong Li
Summary: The study proposes a novel attentional neural network-based model called AttnMove to densify individual trajectories in sparse mobility data. By designing intra and inter-trajectory attention mechanisms, the model can better capture user mobility regularity and fully utilize periodic patterns from long-term history.
THIRTY-FIFTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, THIRTY-THIRD CONFERENCE ON INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE AND THE ELEVENTH SYMPOSIUM ON EDUCATIONAL ADVANCES IN ARTIFICIAL INTELLIGENCE
(2021)