[Objectives] The trajectory knowledge graph effectively captures the deep semantic relationships between trajectories and geospatial entities, offering significant advantages in revealing complex associated information. However, traditional methods for constructing knowledge graphs from domain-specific data sources rely heavily on expert knowledge, involve extensive data preprocessing and entity-relationship extraction, and require high levels of professional expertise. [Methods] To address these challenges, this paper proposes a trajectory knowledge graph construction method that supports natural language-driven task execution through prompt learning with large language models. First, a prompt strategy for the preprocessing task is designed to guide large language models in automatically generating data processing code for cleaning abnormal trajectories. Second, a two-level system prompt strategy is developed to enable tool invocation by matching and calling the trajectory knowledge extraction tool. This strategy allows non-expert users to complete the graph construction process using simple natural language instructions, significantly reducing reliance on programming skills and deep semantic understanding. [Results] To evaluate the feasibility and effectiveness of the proposed prompt strategies, a set of test sentences was created for trajectory preprocessing and entity-relation extraction tasks. Real-world ship and vehicle trajectory datasets were used to support knowledge graph construction. Experiments conducted on two representative large language models, Tongyi Qianwen and Baidu Qianfan, achieved average accuracy rates exceeding 75% and 80%, respectively, demonstrating strong generalization ability and practical value. [Conclusions] This study verifies the effectiveness of combining large language models with prompt learning in constructing trajectory knowledge graphs with low technical barriers, demonstrating the strong generalization and application value of the proposed prompt strategy.

[Objectives] The deep integration of Graph Neural Network (GNN) and Building Information Modeling (BIM) has significantly advanced the Architecture, Engineering, and Construction (AEC) industry. In this paper, we propose a self-supervised GNN model that enhances node features to address the spatial identification and classification problem in BIM, thereby closing the gap between high‑fidelity BIM data and graph‑based representation learning through a task‑oriented framework that captures both geometric and functional semantics. [Methods] The model represents each building space as a node and fuses edge features based on node characteristics. To maximize the descriptive power of the features, topologically meaningful relationships—shared‑wall, passageway, corner, and open‑space connections—are encoded as weighted edges and injected into the node embeddings through a learnable fusion layer before graph propagation. The model is trained on an autonomously constructed knowledge base that includes 12 space types and 4 relationship features across 3 major types of building-space layout graphs, enabling the automatic recognition of functional space types in buildings. [Results] The experimental results show that (1) on the BuildingGraph benchmark, which spans school, apartment, and office typologies and comprises 300 large‑scale architectural layout subgraphs, the enhanced GNN with fused edge features achieved accurate building‑space node classification, attaining 96.83% accuracy while maintaining stable convergence of the training loss on graphs containing thousands of nodes. (2) Compared with existing GNN models under identical hyper‑parameter settings in ablation experiments, BGFEF significantly improved performance in terms of accuracy (97.08%) and F₁ score (96.75%), surpassing Graph‑BERT and SAGE‑E by 3%~12% (3) In real‑world applications, nine BIM models across school, apartment, and office buildings were evaluated by extracting room connectivity topology graphs, and the method's spatial classification accuracy and interoperability were verified, with the best single school instance reaching 96.87% accuracy. [Conclusions] This study improves the efficiency and automation of spatial analysis and provides practical guidance for implementation in real AEC projects. By producing high‑confidence functional labels that map seamlessly to Industry Foundation Classes (IFC), the proposed model pipeline accelerates rule‑based design review, facility management, and evacuation‑route optimization, offering a scalable foundation for future ontology‑grounded graph reasoning across diverse building typologies.

Image super-resolution technology enhances image clarity and enriches image detail by improving image spatial resolution, enabling quality enhancement without changing hardware conditions. Given the large size, complex target features, and abundant details of remote sensing images, along with the need for efficient information acquisition, we propose a Diffusion Super-Resolution (DSR) algorithm based on a conditional diffusion model. This approach uses low-resolution remote sensing images from the same region as conditioning inputs to the diffusion model, while high-resolution images with added noise are concatenated as inputs. A deep noise training network was constructed with U-Net as the backbone, incorporating residual connections and self-attention mechanisms. The loss function was also improved for better super-resolution results. The DSR method was tested using high-resolution remote sensing images from multiple periods of the domestic Gaofen and SuperView satellite series. The super-resolution results demonstrated pixel dimension expansion from 32 to 128. Comparative experiments with Bicubic, SRGAN, Real-ESRGAN, and SwinIR super-resolution algorithms showed that the DSR method outperforms these algorithms in both PSNR and SSIM metrics. Additionally, the DSR method significantly improves the quality of multispectral remote sensing images. By leveraging the conditional diffusion model, it successfully preserves rich detail and enhances spatial resolution without compromising image clarity. This method offers an efficient solution for super-resolution reconstruction, ensuring effective information acquisition in remote sensing applications and fulfilling the requirements of various domains such as land use classification, environmental monitoring, and urban planning. Moreover, the DSR method also opens new avenues for future research by demonstrating the potential of diffusion models in remote sensing image processing. It overcomes the limitations of simple convolutional networks, which extract only shallow features, and avoids the convergence issues commonly seen in adversarial neural networks during training, ultimately improving the restoration of rich details in remote sensing images.

Compared with urban rainstorm hazardous events (such as waterlogging, flood, debris flow, etc.), existing studies pay less attention to the feature composition and objective assessment of risks associated with small-scale and diversified rainstorm cascading events (such as house damage, subway inundation, etc.), making is difficult to meet the goals of refined city management. At the same time, constructing risk assessment models for rainstorm cascading events faces constraints due to incomplete risk features in sample data. To address these issues, this paper proposes a risk assessment model for urban rainstorm cascading events that integrates risk features and spatial features, considering the spatial correlation between them. Firstly, for the risk scenarios of different rainstorm cascading events, the risk features are extracted from data sources such as grassroots officials' inspection, citizen reporting, and social media posts. Secondly, using the spatial localization of the original risk samples as a connection, an improved marginal Fisher method is employed to mine spatial features from multi-source spatial data to supplement the missing risk features. Finally, using a machine learning approach, the relationship between risk features and risk categories is established, leading to the construction of a risk assessment model for multi-category rainstorm cascading events. Experimental results from Wuhan, Hubei Province, China, show that the proposed method effectively addresses the problem of incomplete features in the construction of risk feature models through multi-source spatial feature mining, enabling the construction of diversified rainstorm cascading event risk assessment models. The overall accuracy, F₁-score and AUC increased by 23%, 24%, and 25%, respectively. Additionally, the complexity and diversity of spatial features highlighted the risks of subjective and arbitrary feature fusion, which can negatively affect the performance of machine learning model construction by adding irrelevant features. The proposed method mitigates this issue with an adaptive feature selection approach. Furthermore, grassroots officials’ inspection records contributed the most to the construction of urban rainstorm cascading event risk assessment models, followed by citizen-reported texts, and finally, social media data. Compared to traditional disaster event risk assessment methods, urban rainstorm cascading event risks have smaller risk granularity and involve more complex and diverse risk types and features. Traditional comprehensive evaluation models face challenges of subjectivity in manual evaluation, while traditional disaster loss curve methods encounter high experimental costs and data scarcity. The method proposed in this paper utilizes objective data to generate multidimensional risk features and establishes relationships between diverse risk levels, resulting in a machine learning-based risk prediction model that is more suitable for small-scale risk assessment scenarios.

Community jurisdictional boundaries are pivotal in the context of smart communities, where their prompt and accurate delineation is critical for providing high-quality grassroots services. Currently, the delineation of these boundaries relies heavily on manual labeling by grassroots workers, which poses considerable challenges, including substantial data collection barriers and delays in updating information. Utilizing spatially associated points within community jurisdictions offers a promising approach to address the complexities involved in generating accurate community boundaries. In this paper, we propose CB-GCN, a novel community boundary generation algorithm that integrates volunteered geographic information with graph neural network. This approach ensures the generation of high-quality, cost-effective, and timely community boundaries, even in the presence of complex point distributions and intricate regional divisions. CB-GCN consists of three fundamental components: semantic spatial feature extraction, spatial relationship graph construction, and jurisdictional area affiliation inference. In the initial phase of semantic feature extraction, the city area is partitioned into blocks using a multi-level road network. These blocks are then divided into spatial units based on the spatial coordinates of individual buildings. The use of semantic Areas of Interest (AOIs) and the multi-level road network blocks facilitates the extraction of containment and adjacency relationships between spatial units, which are crucial for constructing the spatial relation graph. During the phase of spatial relationship graph construction, edges are established between spatial units according to the extracted spatial semantic relationships, resulting in a comprehensive spatial relationship graph. The affiliation inference phase involves inferring proximity relationships between spatial units using graph convolutional networks. Spatially related point features of neighboring nodes are then aggregated with weighted adjustments based on these proximity relationships to accurately determine community affiliations. Based on the aggregation results, community affiliations of spatial units are classified, leading to the identification of definitive community jurisdictional boundaries. Experimental results demonstrate that CB-GCN substantially outperforms baseline methods in generating community jurisdictional boundaries, achieving notable improvements of 9.4% in F1-score and 14.4% in Intersection over Union (IoU). CB-GCN has also proven effective in complex scenarios involving fragmented jurisdictional areas, such as when a single AOI is intersected by multiple jurisdictional regions. Furthermore, CB-GCN can effectively generating regional AOIs despite variations in building distribution patterns and densities. By automating the generation of community jurisdictional boundaries, CB-GCN significantly enhances the efficiency of producing community boundary interest areas, representing a substantial advancement in boundary delineation methodologies.

Realizing convenient transfers between subway and regular bus systems is fundamental to advancing the integration and development of these two transportation networks, which is crucial for constructing a multi-modal and accessible public transportation system. This paper takes Shenzhen as a case study and innovatively combines mobile phone signal data with IC card data to identify the transfer characteristics between subway and regular bus systems. These characteristics include temporal and distance aspects, which effectively illustrate the daily travel patterns of transfer passengers. Through a detailed analysis of the overall transfer characteristics, this study establishes a distance threshold to estimate potential transfer demand and the gap in transfer demand at each subway station. Furthermore, this paper uses the Entropy Weight-TOPSIS Model to conduct a preliminary evaluation of the transfer supply conditions at various subway stations. Based on the evaluation results of the matching between transfer supply and demand, as well as the size of the transfer demand gap, this study proposes corresponding optimization strategies for subway stations, providing an effective method for identifying inefficient stations. The research findings indicate that, in Shenzhen, the subway stations with high potential demand for transfers to regular buses are mainly located near densely populated residential areas. The central urban area exhibits a high degree of matching between transfer supply and demand, with some old urban areas experiencing an oversupply due to the well-developed public transportation infrastructure. However, peripheral stations commonly face a situation where demand exceeds supply, necessitating focused attention on improving transfer supply conditions at these sites. Regarding the transfer demand gap, even among subway stations with the same level of transfer supply, variations in the size of the demand gap exist. Stations with insufficient transfer supply but efficient operations offer valuable lessons, while stations with large demand gaps and inefficient operations should be targeted for specific improvements based on their individual supply and demand matching situations. The results demonstrate that evaluating the alignment between potential subway-bus transfer demand and the level of transfer supply using multi-source data, and formulating optimization strategies in conjunction with the transfer demand gap, is of significant importance for enhancing the refined management level of subway and bus transfer services. Overall, the theory and calculation methods of transfer potential demand and transfer demand gap proposed in this study provide a new perspective and reference for transfer research, public transport planning, and urban planning in the field of public transportation.

The assurance of a consistent supply of daily necessities in megacities is pivotal in fortifying community supply resilience. It is axiomatic that a community system is not an insular entity; rather, it intricately intertwines with various elements of urban systems. As a foundational unit of urban governance, the urban community is instrumental in facilitating a congruent nexus between supply and demand, thereby augmenting urban resilience. This study proposes an exploratory evaluation method for the urban community supply support and resilience based on complex network theory, attempting to achieve a breakthrough in the underlying theoretical framework of resilience assessment from "single system assessment" to "multi-system correlation assessment". Taking the six districts in the central city of Guangzhou as an example, we build a supply-demand network based on citizens' spatio-temporal behaviors using multi-source data such as mobile phone signaling data and other data. The attacking strategies of network are based on five community resilience indicators. Besides, the cascade failure mechanism is introduced to evaluate the network resilience, and the entropy-weighted method is employed to obtain resilience evaluation results. The influence mechanism of community resilience on the supply system is further analyzed by studying the factors affecting community node failure at different stages of supply network. The findings are as follows: (1) The proposed evaluation model of the community supply support and resilience can effectively simulate urban community supply-demand networks and evaluate the resilience of communities. Low-resilience communities are mainly categorized into three spatial types: old blocks, urban villages, and suburban blocks; (2) Through the analysis of network resilience under five different attack strategies, it is found that the dominant influencing factors are different, with the population density being the primary factor; (3) There exists a complex bidirectional relationship between community resilience and supply security, including the obvious vulnerability of low-resilience communities. And the community self-organization ability, the supply facility layout, and the linkage scheduling between supply points all affect the overall community resilience.

Car-sharing services can meet the diverse travel needs of users while helping to alleviate traffic congestion and reduce pollution. In many scenarios, car-sharing is more economical than taxis. One-way car-sharing allows users to rent and return cars at any station within the system, which leads to low operating costs and flexible services. However, the spatiotemporal skewness of user travel demand gives rise to imbalances between vehicle supply and demand among stations, which limits the profitability of car-sharing companies. Relocating vehicles can alleviate the above problems to some extent. Most existing studies construct optimization models with the goal of maximizing expected revenue or reducing system imbalance. The former is limited by the insufficient accuracy of travel demand prediction, and the mode of discarding definite orders and pursuing higher possible expected benefits instead cannot guarantee actual profits. To improve system balance, the latter pays more relocation costs such that reduces the profitability. To this end, we propose a revenue-driven one-way car-sharing user relocation model RUG that is suitable for real-time scenarios. The model is based on the deterministic effect of prospect theory, which ensures the current definite gains. For orders that cannot be fulfilled due to imbalanced resources, RUG provides users with alternative travel routes, which not only attempts for promising gains but also effectively balances the system. Users are incorporated into the system as relocation subjects by designing rational user incentive and acceptance models. Public transportation is utilized to break through the distance limitations of user relocation. Relocation plans are evaluated with a greedy heuristic. Experimental results on real-world New York datasets show that the RUG model has significant advantages over existing user-based relocation methods. Under the same parameters, compared to the representative user-based relocation method, RUG increases service order volume and profit by 14% and 60%, respectively. Notably, RUG can effectively raise unit profits during traffic rush hours. By incorporating travel demand forecasting, the model further increases revenue by 5.4% while also improving user service level and system balance.

The converting evolution of cascading disaster scenario refers to that in the process of disaster scenario evolution, the disaster bearing bodies transform into new disaster hazards, forming a disaster chain. Rainstorm can easily cause serious secondary disasters such as waterlogging, debris flow and flood, and the combination of these secondary disasters will make the city more vulnerable. However, existing research on rainstorm cascading scenario deduction lacks the analysis of specific scenario evolution situations such as multi disaster combination, scenario element converting, and human-induced emergencies. Meanwhile, traditional research often relies on the probability inference based on existing scenario evolution networks, without providing a construction method for scenario evolution networks, making it difficult to adapt to the knowledge requirements of actual scenario situation converting deduction. To address the scenario converting evolution problems of urban rainstorm cascading disasters, this paper proposes a scenario converting deduction method for rainstorm cascading disaster response based on multi-source spatial data and probability analysis tools. First, based on local and non-local historical emergency cases, the scenario elements involved in the rainstorm cascading disaster scenarios and their potential converting paths are identified. Next, with the support of Baidu Encyclopedia and Wikipedia network knowledge resources, relevant scenario element features and their associations are extracted, and a Group Lasso machine learning method is adopted to achieve feature selection of involved scenario elements. Then, considering the multi-stage and complex scenario correlation in the process of cascading scenario evolution, a dynamic Bayesian network model for scenario converting deduction is constructed. Finally, a Markov chain Monte Carlo method is used to solve the Bayesian network and generate the converting probabilities. The proposed method is applied to the rainstorm response practice of Wuhan High-tech Zone. The use case results show that the proposed method can combine historical cases and network data to achieve rapid and effective generation of key scenario elements and their features, helping to improve the reliability of scenario converting deduction. At the same time, the proposed method supports the scenario converting deduction of small-scale disaster-bearing bodies such as geographic grids, which helps to provide more accurate rainstorm emergency decision-making support and provide good performance in visual analysis. The uncertainty analysis of the proposed method shows that the precision of original probabilities of key scenario element features and the size of generated geographic grids significantly affect the scenario converting deduction results. These findings provide important information for the local area and are expected to help the rainstorm disaster management of other jurisdictions.

About 60% of all known causes of cancer are related to environmental pollution. Identifying the spatial co-location pattern of prevalent neighbor spatial feature sets in geographical space is important to explore the potential relationship between industrial outdoor air pollutants and cancer risk. The traditional spatial co-location pattern mining algorithms usually calculate the prevalence of co-locations based on the frequency of cancer instances when measuring pattern interest. However, the influence of pollution source on cancer instances is also dependent on their proximity. In addition, pollution sources are also influenced by factors such as meteorological conditions, concentration levels, and the degree of harm. So, the pattern interest cannot be measured by relying solely on the number of instance occurences. To address this issue, a new spatial co-location pattern (called spatial ordered-pair pattern) is defined, and a novel mining algorithm is proposed based on the Gaussian kernel density estimation model. The Gaussian kernel function can well capture the decay of the influence of pollution sources on cancer cases with distance. To better represent the real-world diffusion of pollution sources, a spatial neighbor relationship between pollution source and cancer is defined, which considers urban wind direction, wind speed, and pollution emission concentration. Furthermore, pollution sources are categorized into different carcinogenic groups, and a weighted differentiation method is employed to distinguish pollutants based on their carcinogenic categories. The influence of various pollutants on cancer is calculated by weighting their contributions by the "carcinogenic coefficient." Therefore, a novel metric of the influence of pollution sources on cancer along with corresponding mining algorithm is presented. It not only effectively measures the impact of distance between pollution sources and cancer instances on the prevalence patterns but also models the mechanism of the influence of pollution sources on cancer by incorporating real-world conditions, overcoming the limitations of the traditional methods. Furthermore, this study improves the robustness of the method by using a smoothing factor to mitigate mining anomalies caused by uneven distributions of cancer instances. Finally, the effectiveness and efficiency of the metric and the mining algorithm proposed in this study are tested through experiments on real and synthetic datasets, and insights are also provided for cancer prevention and urban planning for Yunnan Province. The experimental results indicate that both the influence degree and participation index can accurately reflect the pattern interest from both macroscopic and microscopic perspectives. Furthermore, the mining efficiency increases by an average of 60% compared to other algorithms. The proposed influence degree measurement can more effectively capture spatial co-location patterns and can better reflect the impact of pollution sources on the incidence of cancer.

The discrete global grid system refers to the discrete partitioning of the earth's surface into grid cells with multi-resolution hierarchical structure according to certain rules, which is widely used in organization, management, and analysis of massive multi-source spatial data. The hexagonal global discrete grid has excellent geometric properties and is well suited for spatial data processing. However, how to further improve the efficiency of the hexagonal global discrete grid coding operation is still the focus of current research. In this paper, we adopt the model of icosahedral snyder equal-area projection aperture 4 hexagonal discrete global grid system and construct the base coding structure of aperture 4 hexagon based on the correspondence between the hexagonal triaxial coordinates and the coded binary numbers, consisting of 7 base digits in the first layer and 4 base digits int the other layer. We divide the icosahedron into 3 base hexagonal subdivision tiles according to the different subdivision structures and adopt the base coding structure for coding scheme in each hexagonal subdivision tile to establish the aperture 4 hexagonal discrete global grid coding scheme. Besides, this paper designs and implements a fast conversion between aperture 4 hexagonal code and hexagonal triaxial coordinates, based on which an efficient aperture 4 hexagonal discrete global grid encoding operation scheme is constructed, including arithmetic operation of encoding, spatial topology operation, and neighbourhood retrieval operation and cross-plane operation of encoding. Compared with the existing hexagonal discrete global grid coding scheme, the coding scheme proposed in this paper has fewer base code digits, is more concise, and facilitates faster conversion to the hexagonal triaxial coordinates of the grid. Compared with the existing coding operation scheme, the proposed scheme further improves the efficiency of coding arithmetic operation, spatial topology operation, and neighbourhood retrieval operation. The coding addition operation is 2~3 times more efficient than HLQT. The neighbourhood retrieval operation is 3~5 times and 2~3 times more efficient than HLQT and H3, respectively, and is less affected by the coding level of the grid coding. The proposed coding scheme in this paper has the same efficiency of additive operation and subtractive operation, and the efficiency of spatial topology operation is 2 times that of arithmetic operation. The coding cross-plane neighbourhood retrieval operation time is slightly longer than that of the in-plane operation, and the impact on the overall operation time is not significant. This study provides support for the research application of discrete global grid system.

Watermarks play an important role in the copyright protection of electronic map tiles. Research on visible watermark removal can help to evaluate the effectiveness of watermarks and improve anti-attack capabilities in an adversarial manner. The existing deep learning-based methods of visible watermark removal have problems such as requiring a large number of training samples, low efficiency, and watermark residual or unrealistic maps generated in results after watermark removal. To address these problems, inspired by the idea of image inpainting, this paper proposes a method based on conditional Generative Adversarial Networks (CGAN) to remove the visible watermarks on electronic maps. The network model mainly consists of a generator and a discriminator. Specifically, the generator adopts a U-Net structure, which includes an encoding stage and a decoding stage. In the encoding stage, multiple convolutional layers are used to obtain multi-scale features of the input watermarked map tile, and jump connections are used in the decoding stage to stitch features and up sample them to generate the watermark-free map tiles. The discriminator uses patch GAN, a full convolutional network model based on region discrimination, to evaluate the authenticity of the generated map tiles. In order to enrich the details and improve the verisimilitude of the generated maps, this paper further adds perceptual loss and L1 loss with the adversarial loss of CGAN. By optimizing the loss function of the real watermark-free map and the generated watermark-free map, an extremely real-like generated watermark-free map can be obtained. The proposed model has been extensively tested on a road map dataset, which was retrieved from domestic and foreign electronic map manufacturers such as Google, Gaode, Baidu, etc. It includes over 3000 watermarked map tiles with different watermark patterns like texts, colorful logos, and both. The results demonstrate that the proposed model in this paper can realize batch removal of visual watermarks on various map tiles such as navigation electronic maps and remote sensing images, and the training speed of the model is 4 times faster than the FCN-based approach. The comparison results by using different combination of loss functions show that the proposed perceptual loss and L1 loss can significantly improve the values of the evaluation metrics of MSE, PSNR, and DSSIM, which explain the similarity of the generated map tiles with the real map tiles. In a word, the approach proposed in this paper is efficient and simple. It can effectively protect the geometric and geographic information in the map tiles after watermark removal and realize batch removal without manual interferences.

Urban hotspots prediction is a basic but significant task for future urban management. Accurate urban hotspots prediction can improve the efficiency of urban planning and security construction. Existing deep learning methods mainly adopt geographic grid maps, provided urban network, or external data to capture spatiotemporal dynamics. However, we observe that mining some latent self-semantics from raw data and fusing them with geospatial based grid images can also improve the performance of spatiotemporal predictions. In this paper, we propose Geographic-Semantic Ensemble Neural Network (GSEN), a novel deep learning approach, to stack geographical prediction neural network and semantical prediction neutral network. GSEN model integrates the structures of Predictive Recurrent Neural Network (PredRNN), Graph Convolutional Predictive Recurrent Neural Network (GC-PredRNN), and Ensemble Layer to capture spatiotemporal dynamics from different views. Furthermore, this model can also be correlated with some latent high-level dynamics in the real-world without any external data. We evaluate our proposed model on three different real-world datasets. The experimental results demonstrate the generalization and effectiveness of GSEN in different urban hotspots spatiotemporal prediction tasks.

Night-time economy refers to the related economic activities mainly in the services taking place in urban space and at night, which is an important representation of a city's economic development and consumption level. Currently, researchers at home and abroad mostly rest on the theoretical level, or small-scale refined research based on market research and questionnaire survey, lacking in-depth mining using models and mathematical statistics methods, and rarely intuitively show the specific temporal and spatial distribution of large-scale night-time economy. With the development of information technology, night lighting data and perception big data provide new data sources for quantitative research of night-time economy. This paper provides a new perspective for night-time economy by fusing multi-source data. Compared with the traditional survey data, it is more rapid, efficient, and extensive, which is suitable for large-scale research of night-time economy. Based on taxi OD flow, this paper uses spatial clustering algorithms such as DBSCAN and K-Means ++ to identify hot areas of night-time activities in Xiamen City from the perspective of consumers. Based on the night-time lighting image and POI, this paper analyzes the supply and demand relationship by the method of profit and loss and identifies the distribution area of night service facilities from the perspective of merchants. Then we analyze the temporal and spatial distribution pattern of night-time economy in Xiamen City. The results show that: ① The spatial distribution of night activities in Xiamen City is multi ring and decreases to the surrounding areas. The distribution of hot spots of night activities varies from place to place; ② The existing service facilities in some areas of Xiamen City fail to serve the night economy well, and the existing lighting infrastructure, such as lighting and nightscape, is insufficient; ③ There is a moderate positive correlation between residential population density and night activity density, and the results are valid. The profit and loss value and quantity of night service facilities, night lighting, and night activity density are moderately and weakly correlated, and catering facilities are more dependent on night lighting. Finally, we put forward some suggestions for Xiamen's future night-time economic construction, such as providing different night-time services according to different consumer groups and psychology, strengthening the construction of night light infrastructure and market support. The research conclusions are of positive significance to promote social employment, and enhance the utilization rate of infrastructure. At the same time, they can also provide reference for urban economic development and policy formulation.

As the most typical public facility, public toilets reflect the civilized level and management service level of the city and are an important window for building a civilized image of the city. Current research focuses on the accessibility and coverage of public toilets, treats public toilets as spatial points without discrimination, and ignores the heterogeneity of public toilets in different urban functional areas. How to establish a comprehensive and accurate public toilet space evaluation system and analyze the comprehensive service capabilities of public toilets in different regions is obviously insufficient in the current research, which is not conducive to the deployment of public toilets and the advancement of the equalization of basic public services. The emergence of multi-source data provides a new perspective for the research of urban public facilities. Therefore, this paper proposes a rationality evaluation method of public toilet spatial layout based on POI big data from the perspective of urban functional area. We use Term Frequency-inverse Document Frequency (TF-IDF) information weighting technology combined with Point of Interest (POI) frequency density to identify urban functional areas, and integrate OpenStreetMap (OSM) road network density and WorldPop population data to construct a population travel vitality index and evaluate public toilet services in urban functional areas. Finally, the population and spatial coverage rate and the spatial imbalance index are calculated to determine the difference between streets and towns and the rationality of the layout of public toilets in streets and towns. Based on multi-source data, this method quantitatively analyzes the rationality of the allocation of public toilets in different functional areas and explores the differentiating factors of the space allocation of public toilets. This article takes Shanghai, one of the most urbanized cities in China, as an example for calculation. The study finds that: ① The number of toilets in different urban functional areas is different. The number of planned commercial service functional areas is the largest, but the public toilets in the commercial service functional area have the highest qualifications. "Industry-Commercial service" and "Green Space-Commercial service" and other commercial service-related joint functional areas are also at a high level of qualification. This is because a lot of commercial service organizations provide public toilet services to the outside, which has improved the service capacity of public toilets in the area; ② The public functional area has the lowest qualification rate, only 10.27%, which is related to the openness of this type of attached public toilet facilities; ③ The eligibility of public toilets in Shanghai’s streets and towns is generally reasonable, with an average space coverage rate of 67.31% and an average population coverage rate of 70.72%. The imbalance index between public toilet service and travel vitality distribution in streets and towns ranges significantly from 0 to 0.76, of which 147 have an imbalance index less than 0.4, accounting for 69.34%, indicating the comparison of the space allocation of public toilets between these streets and towns is reasonable. The rationality of the allocation of public toilets decreases significantly from the downtown area of Shanghai to the southwest, southeast, and Chongming Island, while there is no obvious attenuation to the northwest, showing a good contiguous service capacity. The results show that this method fully takes into account the issue of the heterogeneity of the functional areas of public toilets and the travel vitality of the population, and the spatial analysis is more accurate and has more practical value.

It is of profound significance in the fields of multi-scale map expression and digital mapping to recognize the architectural pattern intelligently by integrating spatial relations, geometry, semantics, and other features. Applying graph convolution neural network to intelligent recognition of building pattern can overcome the shortcomings of traditional methods, such as relying on artificial experience to set parameters and making rules too strict. However, this method often has the problem of sample proportion imbalance, which easily leads to the category with a small number of samples can not be properly identified. In this paper, the building centriod is used as the sample to cluster to obtain the building groups. Secondly, the building centriod in the same group is used as the node to construct the graph structure of the building group of Delaunay triangulation. The feature selection of the graph node can describe the area, size, direction, and other related indicators of the geometric characteristics of the building group. Thirdly, the building groups graph structure with small number of samples is enhanced by oversampling the graph structure. Then the data before and after the enhancement of the architectural group graph structure with small number of samples are input into the graph convolution neural network model for training, and the performance of the model is evaluated with several evaluation indexes such as ROC curve. The experimental results show that the recognition accuracy of the model for buildings with lesser number of samples is significantly improved after the structure of buildings with small number of samples is enhanced.