As a component of the process of rationalization, lowcost housing Institution, as well as major investiment companies, commonly reuse a typical design as a standard. The goal of this procedure is to obtain cheaper cos...As a component of the process of rationalization, lowcost housing Institution, as well as major investiment companies, commonly reuse a typical design as a standard. The goal of this procedure is to obtain cheaper costs both in the construction and in the maintenance of an homogeneous lot of buildings. The paper shows that, nevertheless an identical design being proposed in different towns, the final results are buildings with a different aspect. This is mainly due to the influence of building codes of the urban plan of the specific town, and to the site conditions, such as street orientation, ground inclination and so on. As a result the paper offers the ability to evaluate the role of the external factors on a standard design （with the afore said scale vantages） under the criminological point of view as well. This may be considered a further vantage that aims to obtain a lower crime risk level in town design.展开更多
Crime risk prediction is helpful for urban safety and citizens’life quality.However,existing crime studies focused on coarse-grained prediction,and usually failed to capture the dynamics of urban crimes.The key chall...Crime risk prediction is helpful for urban safety and citizens’life quality.However,existing crime studies focused on coarse-grained prediction,and usually failed to capture the dynamics of urban crimes.The key challenge is data sparsity,since that 1)not all crimes have been recorded,and 2)crimes usually occur with low frequency.In this paper,we propose an effective framework to predict fine-grained and dynamic crime risks in each road using heterogeneous urban data.First,to address the issue of unreported crimes,we propose a cross-aggregation soft-impute(CASI)method to deal with possible unreported crimes.Then,we use a novel crime risk measurement to capture the crime dynamics from the perspective of influence propagation,taking into consideration of both time-varying and location-varying risk propagation.Based on the dynamically calculated crime risks,we design contextual features(i.e.,POI distributions,taxi mobility,demographic features)from various urban data sources,and propose a zero-inflated negative binomial regression(ZINBR)model to predict future crime risks in roads.The experiments using the real-world data from New York City show that our framework can accurately predict road crime risks,and outperform other baseline methods.展开更多
文摘As a component of the process of rationalization, lowcost housing Institution, as well as major investiment companies, commonly reuse a typical design as a standard. The goal of this procedure is to obtain cheaper costs both in the construction and in the maintenance of an homogeneous lot of buildings. The paper shows that, nevertheless an identical design being proposed in different towns, the final results are buildings with a different aspect. This is mainly due to the influence of building codes of the urban plan of the specific town, and to the site conditions, such as street orientation, ground inclination and so on. As a result the paper offers the ability to evaluate the role of the external factors on a standard design （with the afore said scale vantages） under the criminological point of view as well. This may be considered a further vantage that aims to obtain a lower crime risk level in town design.
基金This work was partly supported by the National Natural Science Foundation of China(Grant No.61772460)Ten Thousand Talent Program of Zhejiang Province(2018R52039).
文摘Crime risk prediction is helpful for urban safety and citizens’life quality.However,existing crime studies focused on coarse-grained prediction,and usually failed to capture the dynamics of urban crimes.The key challenge is data sparsity,since that 1)not all crimes have been recorded,and 2)crimes usually occur with low frequency.In this paper,we propose an effective framework to predict fine-grained and dynamic crime risks in each road using heterogeneous urban data.First,to address the issue of unreported crimes,we propose a cross-aggregation soft-impute(CASI)method to deal with possible unreported crimes.Then,we use a novel crime risk measurement to capture the crime dynamics from the perspective of influence propagation,taking into consideration of both time-varying and location-varying risk propagation.Based on the dynamically calculated crime risks,we design contextual features(i.e.,POI distributions,taxi mobility,demographic features)from various urban data sources,and propose a zero-inflated negative binomial regression(ZINBR)model to predict future crime risks in roads.The experiments using the real-world data from New York City show that our framework can accurately predict road crime risks,and outperform other baseline methods.
