Attributes Of Locations Research Articles

DL-SLICER: Deep Learning for Satellite-Based Identification of Cities with Enhanced Resemblance

This paper introduces a deep learning (DL) tool capable of classifying cities and revealing the features that characterize each city from a visual perspective. The study utilizes city view data captured from satellites and employs a methodology involving DL-based classification for city identification, along with an Explainable Artificial Intelligence (AI) tool to unveil definitive features of each city considered in this study. The city identification model implemented using the ResNet architecture yielded an overall accuracy of 84%, featuring 45 cities worldwide with varied geographic locations, Human Development Index (HDI), and population sizes. The portraying attributes of urban locations have been investigated using an explanatory visualization tool named Relevance Class Activation Maps (CAM). The methodology and findings presented by the current study enable decision makers, city managers, and policymakers to identify similar cities through satellite data, understand the salient features of the cities, and make decisions based on similarity patterns that can lead to effective solutions in a wide range of objectives such as urban planning, crisis management, and economic policies. Analyzing city similarities is crucial for urban development, transportation strategies, zoning, improvement of living conditions, fostering economic success, shaping social justice policies, and providing data for indices and concepts such as sustainability and smart cities for urban zones sharing similar patterns.

Spatial transcriptomics in breast cancer: providing insight into tumor heterogeneity and promoting individualized therapy.

A comprehensive understanding of tumor heterogeneity, tumor microenvironment and the mechanisms of drug resistance is fundamental to advancing breast cancer research. While single-cell RNA sequencing has resolved the issue of "temporal dynamic expression" of genes at the single-cell level, the lack of spatial information still prevents us from gaining a comprehensive understanding of breast cancer. The introduction and application of spatial transcriptomics addresses this limitation. As the annual technical method of 2020, spatial transcriptomics preserves the spatial location of tissues and resolves RNA-seq data to help localize and differentiate the active expression of functional genes within a specific tissue region, enabling the study of spatial location attributes of gene locations and cellular tissue environments. In the context of breast cancer, spatial transcriptomics can assist in the identification of novel breast cancer subtypes and spatially discriminative features that show promise for individualized precise treatment. This article summarized the key technical approaches, recent advances in spatial transcriptomics and its applications in breast cancer, and discusses the limitations of current spatial transcriptomics methods and the prospects for future development, with a view to advancing the application of this technology in clinical practice.

Modelling geographical and built-environment attributes as predictors of human vulnerability during tsunami evacuations: a multi-case-study and paths to improvement

Abstract. Evacuation is the most important and effective method to save human lives during a tsunami. In this respect, challenges exist in developing quantitative analyses of the relationships between the evacuation potential and the built-environment and geographical attributes of coastal locations. This paper proposes a computer-based modelling approach (including inundation, evacuation, and built-environment metrics), followed by multivariate regressive analysis, to estimate how those attributes might influence the expected tsunami death ratios of seven Chilean coastal cities. We obtained, for the examined variables, their average values to different thresholds of the death ratio. Also, our statistical analysis allowed us to compare the relative importance of each metric, showing that the maximum flood, the straightness of the street network, the total route length, and the travel time can have a significant impact on the expected death ratios. Moreover, we suggest that these results could lead to spatial planning guidelines for developing new urban areas into exposed territories (if this expansion cannot be restricted or discouraged) or retrofitting existing ones, with the final aim of enhancing evacuation and therefore increasing resilience.

Learning about Spatial and Temporal Proximity using Tree-Based Methods

AbstractLearning about the relationship between distance to landmarks and events and phenomena of interest is a multi-faceted problem, as it may require taking into account multiple dimensions, including: spatial position of landmarks, timing of events taking place over time, and attributes of occurrences and locations. Here I show that tree-based methods are well suited for the study of these questions as they allow exploring the relationship between proximity metrics and outcomes of interest in a non-parametric and data-driven manner. I illustrate the usefulness of tree-based methods vis-à-vis conventional regression methods by examining the association between: (i) distance to border crossings along the US-Mexico border and support for immigration reform, and (ii) distance to mass shootings and support for gun control.

Estimating Socio-Cultural Services of Coastal Wetlands: A Case of Negombo Lagoon in Sri Lanka

Knowledge on social and cultural systems is essential for comprehensive understanding of ecosystem changes, monitoring biodiversity, development and implementation of resource management strategies. Social values of coastal landscapes reveal the attributes of locations and their relative importance for people. Aim of the study was to evaluate the cultural services and derive the monetary value of social landscape through valuing socio-cultural services of Negombo lagoon. A discrete choice experiment approach was used to estimate the individual’s preference towards social landscape by incorporating featured attributes. The individual preference was expressed through the marginal willingness to pay (MWTP) by integrating a monetary attribute into choice sets. Thus, the MWTP or the implicit price of social landscape attributes included in the choice task was estimated and then the total social landscape value was derived as 13,055 LKR per monthly per household. The social landscape values of Negombo lagoon demonstrated the important ecological, and socio-cultural knowledge, which should be integrated as guiding policies in ecosystem conservation and management strategies. Further, the challenging management paradigm for coastal and marine landscape of Negombo lagoon, and associated services should contain not only the biological and physical benefits but also socio- cultural values too.

Winter Roosting by Eastern Red Bats in Ozark Mountain Forests of Missouri

The eastern red bat (Lasiurus borealis Müller, 1776) is a widespread species that roosts in evergreen or dead foliage suspended in trees during winter but retreats to leaf litter during colder periods. Roosting in leaf litter by eastern red bats makes them vulnerable to prescribed fires in winter. Using radio telemetry, we tracked 33 male eastern red bats to 101 winter (November–February) roosts and quantified roost locations, habitat surrounding roosts, and landscape attributes of roost locations. When roosting in trees, bats preferred oaks but generally avoided other tree species; they used pines in proportion to their availability. During colder periods, bats retreated to roosts in leaf litter where 21% suffered mortality either from predation/scavenging or unknown causes while roosting on the ground. Models of roost selection indicated that southerly aspect was the most important factor determining roost selection, and both tree- and leaf-litter roosts were predominately (≥94%) on upper south-facing slopes. Prescribed burning in late morning/early afternoon on clear days when temperatures under leaf litter are warmest in winter could reduce potential mortality by allowing faster arousal time for hibernating bats.

Multi-view group representation learning for location-aware group recommendation

With the development of location-based services (LBS), many location-based social sites like Foursquare and Plancast have emerged. People can organize and participate in group activities on those sites. Therefore, recommending venues for group activities is of practical value. However, the group decision making process is complicated, requiring trade-offs among group members. And the data sparsity and cold-start problems make it difficult to make effective group recommendation. In this manuscript, we propose a Multi-view Group Representation Learning (MGPL) framework for location-aware group recommendation. The proposed multi-view group representation learning framework can leverage multiple types of information for deep representation learning of group preferences and incorporate the spatial attributes of locations to further capture the group mobility preferences. Experiments on two real datasets Foursqaure and Plancast show that our method significantly outperforms the-state-of-art approaches.

The Importance of Importance Sampling: Exploring Methods of Sampling from Alternatives in Discrete Choice Models of Crime Location Choice

ObjectivesThe burgeoning field of individual level crime location choice research has required increasingly large datasets to model complex relationships between the attributes of potential crime locations and offenders’ choices. This study tests methods of sampling aiming to overcome computational challenges involved in the use of such large datasets.MethodsUsing police data on 38,120 residential and non-residential burglary, commercial and personal robbery and extra-familial sex offense locations and the offenders’ pre-offense activity locations (e.g., home, family members’ homes and prior crime locations), and in the context of the conditional logit formulation of the discrete spatial choice model, we tested a novel method for importance sampling of alternatives. The method over-samples potential crime locations near to offenders’ activity locations that are more likely to be chosen for crime. We compared variants of this method with simple random sampling.ResultsImportance sampling produced results more consistent with those produced without sampling compared with simple random sampling, and provided considerable computational savings. There were strong relationships between the locations of offenders’ prior criminal and non-criminal activities and their crime locations.ConclusionsImportance sampling from alternatives is a relatively simple and effective method that enables future studies to use larger datasets (e.g., with more variables, wider study areas, or more granular spatial or spatio-temporal units) to yield greater insights into crime location choice. By examining non-residential burglary and sexual offenses, in New Zealand, the substantive results represent a novel contribution to the growing literature on offenders’ spatial decision making.

Locational analysis of child streetism in urban centers

BACKGROUND AND OBJECTIVES: Against the background of the growing awareness of the need to provide empirical answers to the fundamental questions of the location of human activities, the construction of social space and the relationship between social space and physical environment, this study examines the locational implications of child streetism in selected urban centres in South-western Nigeria. METHODS:Using simple random sampling, 45 locations were drawn from the three urban centres for the purpose of data collection. The required data was collected through the instrumentation of head count and analysed using Relative Incidence of Street Children (RISC) and Hierarchical Cluster Analysis (HCA). FINDINGS: The results of RISC show that Oja-oba Central Mosque, Ibadan (247), Gbagi-Dugbe Market, Ibadan and Gbagi-Dugbe Motor Park, Ibadan generated highest incidence of street children. However, the results of HCA show that the problem of child streetism is most severe in Oja-oba Market (Akure); Igbona Market (Osogbo); Oja-oba Central Mosque (Ibadan). CONCLUSION: The study concludes that incidence of street children is greatly influenced by social, economic and physical attributes of locations. The study recommends physical planning measures to combat the menace of child streetism.

Application of Spatiotemporal Analysis and Knowledge Discovery for Databases in the Bureau of Fire Protection as Incident Report System: Tool for Improving Fire Services

Purpose–This study aims to contribute to the fire research by developing a fire report management system for the BFP that can analyze spatiotemporal attributes of fire and 520apply Knowledge Discovery in Databases (KDD) methods to identify patterns of fire incidents in the city of Manila.Method–The proponents applied the Knowledge Discovery in Databases (KDD) methods for the processing of identifying fire patterns as well as the application of SMOTE, One-Hot Encoding, and Agile Method as Software developmental model. Result–The records obtained from the BFP headquarters in Manila had a total of 3,506 cases during the six years from 2011 to 2016. The accuracy of the Decision Tree classifier model was 95.92%.Using KDD approach, it generated decision rules fire pattern in Manila. Most fire causes fall under the 'Under Investigation' category while Residential-Commercial types of establishments in Intramuros were affected. Lastly, the fire occurred in the mornings, during Sundays when most people are in their homes and the majority of which took place in the Pandacan district.Conclusion–The application of KDD in building a predictive model to be integrated into the system was the major part of this project. The outputs generated by the system can provide material for use in more accurate fire risk assessments, more efficient allocation of fire resources and personnel, and more targeted fire awareness and prevention programs.Recommendation–Future research in this area may include other factors contributing to a higher likelihood of fire incidences such as weather conditions and other geographical attributes of fire-prone locations. Analysis of these and other relevant factors may allow the BFP to gainfurther insights into the causes of fire incidents, which will enable the agency to make the necessary adjustments and changes in their current fire prevention and risk reduction programsPractical Implication–This study provides direct implication for the Bureau of Fire Protection and community through the given insights of the fire activities and the created model of the system that determine Manila's fire patterns that help identify appropriate information about fire activities and preventive measures of fire incidents.

Den structure and selection of denning habitat by brown bears in the Romanian Carpathians

The Romanian Carpathian Mountains provide one of the largest areas suitable for brown bear (Ursus arctos) in Europe, but the long history of logging has reduced old-growth forest to fragments. Continuous timber extraction, along with new recreation opportunities from motorized vehicles, may affect brown bear denning habitat through disturbance. As such, understanding den site selection at the landscape and local levels is important for the conservation and sustainable management of the Romanian brown bear population. We used data on 86 den sites collected between 2010 and 2013 in Southeastern Carpathians and developed Resource Selection Functions for second-order (landscape-level) and third-order (local level) den habitat selection, using habitat structure and topographic attributes of den locations. The altitude of dens ranged between 440 and 1,320 m, with a mean slope of 19.7 ± 0.8%. Aspect was evenly distributed between southwest (22.9% of the dens), east (20.5%), south (18.1%), and southeast (15.7%). Dens stabilized by boulders were dominant (68%), and had maximum mean length = 149.2 ± 5.6 cm, width = 109.1 ± 4.8 cm, and height = 113.8 ± 10.5 cm. At both local and the landscape scales, bears selected for steeper slope and higher percent coverage of mixed (beech–fir–spruce [Fagus sylvatica–Abies alba–Picea abies]) forest. At the landscape scale, bears also selected dens at higher altitude, and with greater coverage of old forest, and away from urban areas and recent clear-cuts. Our spatial predictions have the potential to inform forest management by identifying areas where disturbance of brown bear denning habitat should be avoided or limited, thus contributing to brown bear management and conservation planning in the Romanian Carpathians.

Personalized location recommendation by fusing sentimental and spatial context

Abstract Internet users would like to obtain interesting location information for a travel. With the rapid development of social media, many kinds of location recommender systems are proposed in recent years. Existing methods mostly focus on mining user check-in information that could be leveraged to understand their trajectories. However, the characteristics and attributes of geographical locations also play an important role in recommender systems. In this paper, sentimental attributes of locations are explored and we propose a Point of Interest (POI) mining method and a personalized recommendation model by fusing sentimental spatial context. First, a Sentimental–Spatial POI Mining (SPM) method is utilized to mine the POIs by fusing the sentimental and geographical attributes of locations. Second, we recommend the POIs to users by a Sentimental–Spatial POI Recommendation (SPR) model incorporating the factors of sentiment similarity and geographical distance. Last, the advantages and superior performance of our methods are demonstrated by extensive experiments on a real-world dataset.

지원정보 속성과 시각적 디스플레이 공간에 따른 상호작용 특성

본 연구는 시선처리의 우선순위에 기초하여 시각적 정보제시 인터페이스의 위치를 구분하고, 해당 위치별로 분류된 내비게이션의 다양한 지원정보 제시를 위한 방법과 기준을 제시하고자 하였다. 연구 1에서 전방 영역은 대부분의 운전자 지원정보가 높이 평가되었고, 클러스터 영역은 현재주행속도정보가, 그리고 센터페시아 영역은 목적지 도착시간, 남은 거리를 알려주는 경로계획 정보가 높은 적합성과 선호도를 보였다. 연구 2에서는 정보조합의 적합성을 기준하여 구성한 운전자 지원정보와 정보제시 위치조합 세트(상/중/하 구분)를 참가자들에게 운전시뮬레이션과 연동하여 경험하게하고, 이에 대한 적합성 및 선호도를 평가하였다. 그 결과는 중 세트에서 높은 적합성과 선호도를 나타냈고, 하 세트는 낮은 적합성과 선호도 나타냈다. 운전자들은 내비게이션의 운전자 지원정보 중 자주 확인하는 운전자 지원정보나 유사한 정보는 동일한 위치 혹은 운전자의 시야가 상하로 움직이는 전방과 클러스터 영역에 위치할 때 운전 시 편리성과 지원정보의 적합성이 높은 것으로 분석되었다.

Assessing the Macro-Economic Benefit of Installing a Farm of Oscillating Water Columns in Scotland and Portugal

The nascent wave energy sector has the potential to contribute significantly to global renewables targets, yet at present there are no proven commercially viable technologies. Macro-economic assessment is seldom used to assess wave energy projects, yet can provide insightful information on the wider economic benefits and can be used in conjunction with techno-economic analysis to inform policy makers, investors and funding bodies. Herein, we present a coupled techno–macro-economic model, which is used to assess the macro-economic benefit of installing a 5.25 MW farm of oscillating water column wave energy devices at two locations: Orkney in Scotland and Leixoes in Portugal. Through an input-output analysis, the wide-reaching macro-economic benefit of the prospective projects is highlighted; evidenced by the finding that all 29 industry sectors considered are either directly or indirectly stimulated by the project for both locations. Peak annual employment is expected to be 420 and 190 jobs in Portugal and Scotland respectively during the combined installation and manufacturing stage, with an associated peak annual GVA of over €16.6 m and €12.8 m. The discrepancies between the two locations is concluded to largely be a result of the site-specific attributes of the farm locations: specifically, increased water depth and distance to shore for the Portuguese site, resulting in higher costs associated with mooring and electrical cables and vessels. The insights gained through the presented results demonstrate the merit of macro-economic analysis for understanding the wider economic benefit of wave energy projects, while providing an understanding over key physical factors which will dominate estimated effects.

소셜 빅데이터를 활용한 영화촬영지의 관광목적지로서의 매력성 지속요인 분석

This study aims to investigate how long the attractiveness of film locations continues. For this purpose, attractive attributes of film locations were extracted and categorized with the text mining method of social big data associated with an increase in tourist numbers. The film location selection was based on domestic film-goer numbers from the Korean Film Commission and tourist statistics of local governments. Research findings are as follows. First, native attractions, accommodations, convenient facilities, accessibility, and other additive factors in tourist destinations changed similarly with the change of average numbers of tourists per month. Second, tourism products, souvenirs, and human service provision changed differently. These facts indicate that important attributes of film locations as tourist destinations are native attractions, accommodations, convenient facilities, accessibility, and other additive factors. Tourists perceive these attributes as important factors for tourist destination choice.

An Exploration of Tourists’ Perceptions of a Rural Mountain Bike Event

The natural attributes of rural locations are well suited to sport event tourism and can play a significant role in strategic tourism development in new markets. The purpose of the study was to explore consumer perceptions of sport event tourism in rural locations, which is important for event managers and for those involved in rural tourism development. Data collected over two days at a mountain bike event in the rural highlands of Scotland explored the visitor awareness and perception of events in a rural location. Results indicates that attendee background does influence the perception of the impact a sport event has on a rural location. Subscribe to JASM

Spawning lake sturgeon (Acipenser fulvescens Rafinesque, 1817) and their habitat characteristics in Rainy River, Ontario and Minnesota

Lake Sturgeon (Acipenser fulvescens) spawning habitat in the Rainy River was studied in light of the ongoing review of the rules governing water levels upstream that affect discharge rates through the International Falls dam. The objectives of this study were to assess the current status of spawning Lake Sturgeon below the dam by: (i) evaluating weight–length relationships, condition, age and growth; (ii) identifying spawning locations and characterizing its physical attributes; and, (iii) evaluating the effects of water surface elevation on the availability of spawning habitat. Spawning was confirmed with use of egg mats, and targeted sampling of spawning individuals was completed using gillnets and electrofishing. Physical attributes of spawning locations, including temperature, depth and water velocity were collected at a range of water levels using an Acoustic Doppler Current Profiler and representative sediment sampling. Biological characteristics and timing of spawning events were consistent with those previously reported for Lake Sturgeon. However, a wider range of depths (0.3–10.5 m) and velocities (0.01–1.9 m3/s) than reported in most single site studies was found here, with the ranges encompassing what is reported in the literature as a whole. Lake Sturgeon were found to prefer coarse spawning substrates dominated by bedrock, boulder or cobble. The availability of preferred spawning habitat varied significantly with the changing water levels observed during the study and was eventually limited at higher water levels by river channel form. Because of the implications for dewatering of nest sites, keeping flows constant during spawning is critical to Rainy River Lake Sturgeon spawning success.

Analyzing tour chaining patterns of urban commercial vehicles

This paper presents an innovative methodology to comprehensively model two dimensions of commercial vehicle travel activity using disaggregate level commercial vehicle daily travel data. Two dimensions of freight activity considered in this model include the daily tour-chaining pattern(s), and the number of trips made in the tour-chains. The results of this study indicate that a host of factors including cargo characteristics, trip characteristics, activity features, base location of commercial vehicle trips, and the attributes of intermediate stop locations affect the choice of tour chain strategy and the associated number of daily trips.

Toward advanced representations of the urban microclimate in building performance simulation

Abstract The present contribution is concerned with the potential of empirically-based methods to capture the microclimate variance across a city and its implications for the performance of buildings. We explore the possibility to explain microclimatic variance across an urban area based on geometric and semantic attributes of specific locations. We use high-resolution and dynamic weather data streams across numerous urban locations in the city of Vienna, Austria. Using advanced data extraction methods, the values of a number of urban attributes that are hypothesized to contribute to the urban microclimate variance (e.g. morphological factors, semantic properties of urban surfaces) are derived for these locations. The results point to possible correlations between location-based climatic conditions and distinct urban attributes that could be harnessed to formulate empirically-based algorithms for generating customized microclimatic boundary conditions.

Predicting Influenza Dynamics using a Deep Learning Approach

GIScience 2016 Short Paper Proceedings Predicting Influenza Dynamics using a Deep Learning Approach Shiran Zhong and Ling Bian University at Buffalo, the State University of New York, 105 Wilkson Quad, Buffalo, NY 14261 Email: shiranzh@buffalo.edu, lbian@buffalo.edu Abstract Disease transmission is a complex spatio-temporal process. A great number of approaches have been developed to predict influenza epidemics. Few of them have focused on the temporal dynamics of individual infected locations. Location networks, where locations are nodes and disease flows between them are links, provide a promising approach for such dynamic analyses, but also present challenges. In this study, we employ a deep learning approach to capture the dynamics of disease flows in location networks. We also analyze how the attributes of locations have an impact on the prediction accuracy via a sensitivity analysis. Introduction Disease transmission is a complex spatio-temporal process (Ferguson et al., 2005; Charaudeau et al. 2014). Among the prevailing approaches to predicting the dynamics of influenza epidemics, few have focused on the transmission at a location-specific scale. Location networks, where locations are nodes and disease flows between them are links, provide a promising basis for such dynamic analyses (Zhong and Bian 2016), but also present challenges as each location might have peaks and troughs of different magnitudes throughout the epidemic (Bian et al. 2012). Conventional approaches are not adequate to capture such complex, dynamic patterns. The objectives of this study are two-fold. We explore the use of Deep Convolutional Networks (DCN), a deep learning approach, to capture and predict disease flow dynamics represented by the presence of links between locations. We also analyze how the attributes of locations impact the prediction accuracy via a sensitivity analysis. Data and Study Area The dataset consists of 73 daily disease flow networks of an urban area. In these networks, all 1,026 location nodes remain the same across 73 days. Links can be present or absent depending on the occurrence of disease flow between locations on any particular day. The dataset was obtained from the China Information System for Diseases Control and Prevention. Methodology To achieve the objectives stated above, the methodology is divided into three parts: the first part describes the principle of DCN; the second part introduces the training and testing processes involved in the DCN; and the last part focuses on the sensitivity analysis. 3.1. Principle of DCN In contrast to classic neural networks which require good features for supervised training, DCN is a training process where good features could be automatically learned from input data (LeCun et al. 2015). The workflow of DCN in this study (Figure 1) contains three processes: 1) the convolution process, where a convolution filter is applied on the input data (in a format of matrix) in order to amplify the feature signal and suppress the noise; 2) the pooling process, which extracts features that represent location characteristics in the past few days; and 3) the training and testing process, where the learned features are used as input to predict the presence