Convergence, Mining, and Application: A Data Collaboration Framework for Spatial-Gene Research and Practice

Abstract

In the digital era, data collaboration constitutes a critical trend in urban planning and design. It is of paramount importance in addressing contemporary issues related to the misinterpretation, misapplication, and misunderstanding of spatial genes, as well as facilitating data sharing and value creation associated with spatial genes. In this paper, targeting the complex problems of multiple entities and threads in spatial gene research and practice through, initially, a literature review, the correlation process between urban planning and data collaboration is examined, the spatial gene concept and the background of its proposal are expounded, and the challenges confronted in spatial-gene data collaboration are analyzed. Then, with an elaboration of the data value chain concept, a data collaboration framework for spatial-gene research and practice is constructed, specifically encompassing three main links: convergence, mining, and application. Finally, from the three aspects of data collection and storage, data analysis and processing, and data circulation and sharing, technical implementation paths and suggestions are put forward. We firmly contend that through the establishment of the framework, it is anticipated to promote data collaboration among multiple entities, enhance the efficiency and scientificity of urban design, and thereby facilitate the preservation of cultural diversity and the sustainable development of cities.

1. Introduction

Currently, we are in the digital era, and the quantity and types of data are increasing exponentially, presenting unprecedented opportunities and challenges to urban planning. On one hand, through the collection of the constantly emerging social, economic and spatial data, urban planners can gain a broader comprehension of urban development phenomena [1], analyze the shaping process of contemporary urban structures and urban environments more precisely [2], and undertake urban planning in a timelier fashion to frame the goals of sustainable development [3]. Urban research based on abundant data is anticipated to play a more significant role in urban planning and also offers the possibility of citizen-opinion-driven urban governance [4,5]. On the other hand, the continuous accumulation of high-frequency heterogeneous data has rendered the formulation of urban planning policies increasingly complex. Urban planners require a substantial amount of data-resource support at different stages, such as research, analysis, evaluation, and design, to elucidate various aspects of urban development. Government officials also need to conduct real-time tracking of data regarding the quality, progress, and effectiveness of urban planning projects. Some scholars critically assert that issues such as insufficient data reliability, information islands, and the absence of virtual collaboration platforms are posing challenges to the validity of urban planning decisions [6,7].

In view of the extensive scope and diverse contents of urban planning data, numerous urban planning institutions, administrative departments, and enterprises have regarded digitalization as a decisive factor and conducted numerous new technological explorations in digital urban planning. Database technology is utilized for the aggregation and integration of various types of spatial fundamental data, planning information data, and planning management data, providing a unified storage space for the querying and sharing of planning data [8]. GIS, which is oriented towards the management, analysis, and visualization of massive geographic information data, compensates for the deficiencies of traditional drawing software like CAD and SketchUp in the capabilities of large-scale spatial-data statistics, analysis, and processing [9]. Spatial-data service systems based on Information and Communication Technology (ICT), offer a novel Cyberinfrastructure (CI) for the sharing and exchange of urban planning data [10]. Additionally, Foth et al. proposed a sociable smart city framework and probed into the possibility of achieving cognitive urban development through more participatory technological innovations [11]. Nevertheless, efforts in a single direction can only bring about some local improvements [12] and have not systematically addressed the intelligent issues of data-driven urban planning.

Hence, how can a systematic solution be provided to effectively integrate the links, such as the collection, analysis, and application of multi-source planning data, facilitate the circulation and utilization of them among different departments and personnel, and enhance the efficacy of urban planning research, practice, and management in order to promote sustainable urban development?

Previously, rather significant information barriers were predominant among multiple entities involved in urban planning. Despite possessing a certain quantity of data resources, it was likely that territorial planners merely focused on land use, and transportation planners only concentrated on traffic organization. The data among these entities did not circulate, resulting in a waste of data resources and low planning efficiency. With the continuous advancement of urban science and its integration with databases, ICT, Workflow, and GIS [5], data collaboration that integrates up-down, down-up, and inside-out urban planning approaches is emerging as a trend in urban planning digitalization. It not only entails the aggregation of planning data resources but also encompasses the integration of workflows tracking the entire process of urban planning. This can facilitate the formation of an organic entity among multiple entities and better utilize data resources to meet the requirements of urban development. In the specific research domain of urban planning data collaboration, certain achievements have emerged successively in domains such as transportation planning, environmental protection planning, and community planning. Nevertheless, there is a relative lag in urban design, where the completeness of data and its architecture is inadequate. It is requisite to promote the systematic enhancement of the capabilities of urban design data storage, extraction, analysis, and application through data collaboration [13].

As a novel concept in urban design that has garnered extensive attention, spatial-gene research is also confronted with severe issues regarding data collaboration. Spatial-gene research pertains to the relatively stable and distinctive spatial-combination pattern that has gradually emerged during the historical course of a city, functioning as the fundamental units of urban form inheritance. It conveys information on the interactive evolution of the “space-natural-cultural” system and encompasses knowledge and regulations for shaping the spatial environment within a specific region. This concept was put forward by Duan et al. from Southeast University in 2019 [14] and has emerged as a rapidly evolving research domain. Currently, more than 71 institutions have been involved in spatial-gene research, and their outcomes have been widely employed in urban design, urban renewal, and heritage protection projects [15,16,17]. There are three factors that have engendered an urgent demand for spatial-gene data collaboration: spatial gene involves a complex data-content system, encompassing multiple spatial scales such as the city, district, street, and building, as well as all facets of urban morphology, including mountain, water, city, people, and culture; studying the genetic systems of multiple cities on a global scale can contribute to protecting the local characteristics of cities and maintaining cultural diversity; and the research on biological genes [18] and materials genomes [19] has attained remarkable and commendable results in data collaboration. However, the research regarding spatial-gene data collaboration is still at the stage of theoretical exploration, and the majority of scholars have not yet been aware of this. Furthermore, rather common phenomena such as misinterpretation, misapplication, and misunderstanding data have emerged in spatial-gene research and practice [20]. This not only restricts the data-sharing and value creation [21] but also influences the development and dissemination of this concept. Consequently, within the context of urban planning digitalization, it is of dual urgency, both theoretical and practical, to undertake an analysis and propose systematic solutions for the collection, storage, analysis, mining, application, and sharing of spatial-gene data from the perspective of data collaboration.

This paper presents the challenges emerging in the data collaboration of spatial-gene research and practice and undertakes a study on the data architecture of spatial-gene research from an integrated viewpoint of urban planning, information science, and management science. Then, a data collaboration framework for spatial-gene research and practice, encompassing convergence, mining, and application, is established. The objective is to integrate multi-source planning data, rationalize the planning process, and achieve data collaboration among researchers, planners, government managers, and the public. This framework is designed to enhance the efficiency and scientificity of urban design research, formulation, and management, facilitate the maintenance of cultural diversity, and facilitate sustainable development.

This paper consists of seven main sections. Starting with the introduction, the outline of the research is presented. In the second section, the literature review is presented, concentrating on the latest thoughts and issues regarding data collaboration for spatial-gene research and practice. In the third section, by elaborating on the concept of the data value chain, a data collaboration framework for spatial-gene research and practice is constructed. The next three sections forming the main content of this paper depict, respectively, the three routes of data collaboration: data convergence, data mining, and data application. Finally, this study ends with conclusions and research prospects.

2. Literature Review

2.1. Urban Planning Data Collaboration

Recognized digital scenarios incorporate multiple modules, such as data collection, storage, extraction, and decision-making, to varying degrees within their respective data architectures [22]. In conventional working paradigms, greater emphasis was placed on data collection and extraction. Due to the constraints imposed by organizational structures, professional divisions of labor, and established workflows, data often remains confined within discrete modules and is unable to flow freely; this phenomenon has led to the emergence of “Isolated Data Islands”, resulting in disruptions in the flow of data values.

With the progress of digital technology, researchers have transitioned from traditional querying and reporting to data extraction, management, and analysis for obtaining insights. At present, data is defined as a crucial production factor that demonstrates significant value and increasingly participates as a strategic resource in inter-organizational flows. The concept of data collaboration specifically emerged under this background. It pertains to the process of integrating diverse data sources, exploiting them collaboratively, and, ultimately, sharing data products to achieve a unified business objective [23]. Specifically, to overcome barriers that impede the flow of data values, data engineers address the data relationship issue by integrating scattered data from various terminals to ensure the consistency and accuracy of the data. Subsequently, through multi-agent collaboration, data is processed and applied, facilitating data sharing and value creation. Moreover, the digital portraits, AI data models, and algorithms generated through multi-agent data collaboration are constantly accumulating and forming the social value foundation in the digital era.

In the domain of urban planning, when confronted with issues such as information islands, high heterogeneity, poor readability, and difficult collaboration that exist in traditional data architectures [24,25], data collaboration fully leverages the professional strengths of multiple entities in planning. Through the systematic construction, integration, and utilization of data resources in a top-down and bottom-up manner, it promotes business collaboration and entities’ collaboration in urban planning. Particularly in the trend of the breakthrough development of regions and the increasing complexity of time and space in the digital era, data collaboration has gained increasing significance due to its notable advantages in terms of efficiency, accuracy, and coordination.

In recent years, the research regarding urban planning data collaboration has mainly focused on technological applications such as Geospatial Cyberinfrastructure (GCI) [10], Big Data [26], and the Large Language Model (LLM) [27]. GCI provides a software platform that integrates computer hardware and software, facilitating the integration of complex system modeling and distributed scientific data [10]. This constitutes a transformation from independent desktop paradigms to a framework based on Web services [28]. Jing et al. proposed a lightweight data architecture integrating Web services and component technologies, featuring a unified data-access interface and strong editing capabilities, ensuring high feedback speed and unrestricted editing [29]. Despite the issue of data incoordination, systems based on big data can be utilized to capture and monitor the relationships between human beings and ecosystems in urban spaces [30], offering new possibilities for continuously describing urban phenomena, collecting the preferences and perceptions of human beings, and conducting urban regeneration dynamically [31]. Additionally, LLM, based on a multi-agent collaboration framework, has shown considerable potential in urban planning computations, achieving a performance leap in terms of planning inclusiveness, fairness, and diversity [32,33,34,35].

Overall, the positive effects of urban planning data collaboration are mainly demonstrated in the following aspects:

(1)

Information Integration: Merging multi-source heterogeneous data to assist planners in grasping the urban development status in the form of “information sets”.

(2)

Workflow Optimization: Streamlining the planning practice process, reducing repetitive work and unnecessary connections among different technical links.

(3)

Decision Support: Integrating unified analytical viewpoints to support precise policy-making by planners and managers.

(4)

Collaborative Sharing: Establishing a mechanism to facilitate information exchange and collaborative work among different institutions.

Nonetheless, the subsequent issues are how to share these data among geographically dispersed institutions and how to ensure data flow in complex urban planning procedures.

2.2. Spatial Gene

Since the second half of the 20th century, in the course of globalization, an academic self-examination on urban cultural diversity has arisen worldwide. Scholars have carried out research on the regionality of cityscapes from different perspectives, such as anthropology [36], genetics [37], and ecology [38], and relevant disciplinary theories have also been extensively applied in the domain of urban planning.

In urban planning research from a genetic perspective, Langton initially explored the possibility of the existence of molecular logic in artificial life and proposed the utilization of cellular automata (CA) to study it [37]. Batty and Longley discussed that the growth process of cities encodes rules that determine how the organization and repetition of basic social space entities can achieve certain urban forms and functions at different scales [39]. Subsequently, Silva put forward the concept of Regional DNA and constructed a CA model that could be employed to define it, which was the first attempt to combine gene theory with urban research directly [40]. Based on Silva’s study, Wilson defined urban DNA as the different combinations of multiple structural variables, and this concept has recently been applied in normative urban design [41]. Subsequently, the hypothesized urban DNA was classified into two aspects: physical gene and non-physical gene [42].

In terms of the physical gene, there are three significant research branches in history:

(1)

Urban Morphology. Conzen adhered to the morphogenetic research tradition, formulated the research methodology of morphological factors, and specified the research framework and terminology of Urban Morphology [43].

(2)

Architectural Pattern Language. Alexander focused on the interaction between form and context, proposed 253 patterns ranging from the macro to the micro level of the city to describe the consistency between place form and activities, and indicated that the relationship pattern in space is in accordance with a certain event pattern. The creation of an architectural form requires following a rule system composed of patterns [44].

(3)

Building Typology. Caniggia observed the impact of the leading type on the evolution process of spatial entities, encompassing two main characteristics, namely, synchronic variations and diachronic variations, and firmly believed that by perpetuating this leading type, a harmonious coordination between new and old buildings and urban tissue could be achieved, thereby inheriting history [45].

Regarding the non-physical gene, Marshall argued that, contrary to the role of biological genes in biological evolution and variation, the synthesis of urban DNA has a feature of intervention [46]. D’Acci focused on how to balance the cooperative and competitive relationships among different individuals in urban agglomerations through the synthesis of urban DNA in spatial planning and urban governance [47]. Votsis and Haavisto proposed that urban DNA could control the evolutionary mode of a city and its morphological manifestation. Through the collection of urban DNA, i.e., livability and sustainability data of global city samples, they distinguished six types of cities with distinct behaviors and performances [48].

Duan et al. synthesized the notions of physical genes and non-physical genes and put forward the concept of a spatial gene [14]. Physically, spatial genes are the distinctive and relatively stable spatial-combination patterns formed in the historical process, specifically encompassing three core aspects: spatial elements, spatial relationships, and their attributes at different spatial scales. They constitute the fundamental framework for the formation of regional cityscape characteristics [49]. Non-physically, spatial genes carry the information of the interactions among the natural environment, historical culture, and urban space within a certain region, which is of paramount importance for maintaining the harmonious relationship among the three and achieving sustainable urban development [14]. Take the urban axis genes of China and Western countries as an example; different natural environments and historical cultures have created distinct urban axis genes for China and Western countries, which are manifested as significant differences in aspects such as the location, form, and function of the constituent elements of the urban axis, like landmarks, squares, roads, and building clusters. Furthermore, this specific spatial-combination pattern shapes different cultural characteristics of China and Western countries. For instance, the urban axis gene of China encompasses the ancient ritual thought [14]. Subsequently, Duan et al. expounded on the generation and continuation mechanisms of spatial genes. The generation mechanism of spatial genes was generalized into two aspects: variations in built forms and selections by urban systems, while the continuation mechanism of spatial genes was disclosed through the three processes of encoding, replication, and expression [20].

2.3. Spatial-Gene Data Collaboration

As a rapidly evolving research domain, the data architecture of spatial genes has undergone continuous alterations and updates, posing significant challenges to the establishment of its data collaboration framework. To achieve this research breakthrough, we conducted a comprehensive review of the literature, standards, and projects to analyze the primary issues and identify key areas in need of improvement.

(1)

The basis for the collaboration of spatial-gene data is rooted in a shared comprehension of its conceptual ontology and characteristics. While Duan et al. have articulated the theoretical foundations, conceptual implications, function mechanisms, and technical frameworks of spatial genes in two articles titled “Space Gene” [14] and “Space Gene: Connotation and Functional Mechanism” [20], there remains widespread confusion in information accumulation. Some researchers have not fully comprehended the concept of the spatial gene and have mistakenly used it interchangeably with related concepts, such as the cultural gene, landscape gene, morphological gene, and institutional gene. Furthermore, within the traditional paradigm of morphological typology, some have equated spatial genes with architectural elements like windows, roofs, and totems while overlooking the “spatial-combination pattern” as the ontological essence of spatial genes along with their characteristics of self-organization, hierarchical structures, and openness. This phenomenon created a “logical data island”, which has brought about disorder in the input and output of spatial-gene data, raising doubts regarding data reliability and compatibility. As a consequence, multiple rounds of checks were necessary in data application, consuming a considerable number of human resources.

(2)

Spatial genes, as a theoretical framework for exploring the principles of spatial development and knowledge pertaining to spatial intervention, focus on the past and present status of the city and the deep structure underlying the harmonious interactions among natural environments, historical culture, and urban spaces. How can these principles be effectively utilized to conduct urban planning and design? To address this issue, we have systematically reviewed the literature on spatial genes available in the China National Knowledge Infrastructure (CNKI), Science Citation Index Expanded (SCIE), Elsevier ScienceDirect, and MDPI to discern the primary objectives pursued by scholars in this domain. According to statistical data, current research related to spatial genes can be broadly classified into three domains: fundamental research, technical research, and engineering research. Notably, engineering research papers constitute approximately 54% of the total, underscoring the practical significance of spatial genes at an application level. However, from a “Data Life Cycle Management (DLM)” perspective, while Duan et al. have proposed an application system for spatial genes that encompasses identification and extraction, analysis and evaluation, and guidance and inheritance [14], there remains a deficiency of targeted and detailed investigation into the interrelationships among these stages. This gap contributes to insufficient continuity in numerous planning research and practices as well as leading to “physical data islands”.

(3)

Under the urban planning institutional system of many countries, spatial genes predominantly achieve their data value creation through urban design and the construction of human settlements and places. In this context, the Ministry of Natural Resources of China (MNRC) issued the industry standard “Urban Design Guidelines for Territorial Planning (TD/T 1065-2021) [50]” in 2021. For the first time, it integrated spatial genes into national-level urban design standards, establishing them as essential components for planning formulation, management, and implementation. However, this document fails to specify the requisite design depth, output formats, review criteria, or other relevant aspects. The absence of established data conversion rules complicates the translation of spatial genes into planning management terminologies such as quantities, locations, and proportions. Consequently, planners and managers are obliged to conduct related work based on their individual interpretations and methods, resulting in considerable ambiguity. This deficiency in data products and application scenarios impedes spatial-gene data from effectively entering subsequent stages of construction engineering and also limits its ability to play more specific roles within urban development.

(4)

Although the basic data of spatial genes is constantly on the rise, due to the absence of convenient sharing channels, researchers are incapable of directly querying and comprehensively collecting spatial gene information from other regions. This is highly detrimental to the protection of urban cultural diversity and sustainable development based on spatial genes globally. Consequently, establishing an open and collaborative working environment for this expanding cohort of researchers has emerged as a critical prerequisite for facilitating academic exchanges and cooperation in spatial-gene studies, as well as for disseminating its theoretical advancements. We have noted that a series of standardized outcomes related to spatial genes have gradually emerged, demonstrating significant effectiveness in unifying the understanding and definition of spatial-gene data while reducing communication costs associated with data collaboration. For instance, the group standard issued by the China Urban Planning Society, titled “Inheritance and Control of Spatial Genes in Characteristic Villages and Towns”, supports both the preservation and transformation of characteristic villages and towns across different regions through standardization and systematization [51]. However, standards alone are insufficient to support effective spatial-gene data collaboration. This type of external data collaboration, rooted in expert allocation and specific organizational management, continues to encounter challenges due to strict constraints related to time and resources, among other factors.

3. A Spatial-Gene Data Collaboration Framework Based on the Data Value Chain

In 2012, data scientist Kasim introduced the concept of the data value chain, positing that the entire process ranging from data collection to decision-making constitutes this value chain [52]. Firstly, researchers are required to comprehensively gather various forms of data from multiple sources and store it reliably. Given that raw data exists in a multi-source and heterogeneous state, the processes of extraction-transformation-loading (ETL) can transform and aggregate this data while deriving unified new attributes to meet subsequent analytical requirements [53,54]. Subsequently, researchers will employ a range of data analysis algorithms to process the collected information for effective knowledge discovery. Traditional research methodologies primarily focus on efficient access to and the exploration of datasets to identify patterns or relationships; however, a significant challenge persists in constructing analytical algorithms capable of addressing the complexities inherent in both data and processes [55]. Finally, the results obtained from analysis will be visualized to assist researchers in comprehending large volumes of complex information for informed decision-making [52]. To present these analytical outcomes visually and intuitively, sophisticated applications involving extensive computations will be utilized to enhance researchers’ insights.

We have referred to Kasim’s study and maintain the opinion that the fundamental cause for the issues hindering data collaboration, such as inaccuracy, misalignment, ineffectiveness, and lack of sharing of spatial-gene data, resides in the deficiency of effective guidance and control over the data value chain. In an ideal situation, during the development of the data value chain, spatial-gene data mainly traverses three stages: input, process, and output, manifesting in four forms: “raw data”, “information”, “knowledge”, and “wisdom” [56,57]. Only through the continuous application of this data chain can collaboration among multiple entities be enhanced, thereby maximizing the transformation of data forms and facilitating both the creation and augmentation of the spatial-gene data value (Figure 1). The attainment of these objectives requires a comprehensive and systematic planning process that encompasses multiple phases, including data collection, storage, analysis, processing, application, and sharing, with the scope of data transitioning from “flexible” to “rigorous” and then back to “flexible”. It is only through this approach that effective collaboration on spatial-gene data can be actualized.

Therefore, aiming to generate and enhance data value through multi-agent collaboration, we have structured the spatial-gene data collaboration framework into three essential steps: data convergence, data mining, and data application (Figure 2). This framework is based on the technical chain that facilitates the sequential utilization of spatial-gene data within urban planning and design practices.

(1)

The data convergence stage is intended to establish standards for spatial-gene raw data, data submission, and verification mechanisms, ensuring the reliability and compatibility of spatial-gene data sources and facilitating the dynamic collection and storage of multi-modal and high-quality data through a shared database.

(2)

The data mining stage serves as a bridge between the application layer and the data resource layer, with the aim of establishing a workflow-based toolset for spatial-gene data analysis. Through the analysis and processing of data, data products of spatial genes and knowledge of urban genetic laws are obtained, making the data readable and available and meeting the constantly changing business requirements of spatial-gene applications.

(3)

The data application stage is designed to establish a management and sharing system for spatial-gene data products, integrating data as a production factor and strategic resource into all phases of planning and design, allowing it to be fully applied and circulated within the business chain and facilitating timely interaction among researchers, planners, the public, and government departments in order to achieve the transformation of data values into social, economic, and cultural values.

In this manner, through the establishment of an abstract and loosely coupled spatial-gene data collaborative framework, research and practice are integrated into standardized and operable services. This not only enhances data quality and business agility and efficiency but also enables data interoperability and sharing among multiple entities, ultimately facilitating the ecological governance of the spatial-gene data value chain.

4. Convergence: Collection and Storage of Spatial-Gene Data

Before the introduction of the methodology of data collaboration, it is necessary to identify and classify the raw data of spatial genes. What specific data should be collected, and in what structural format should they be stored? The standard of the database is the primary issue that needs to be addressed in the collaboration of spatial-gene data. Based on a literature review, the current accumulation of spatial-gene data mainly comprises remote sensing satellite images, photographs, multi-scale topographic maps, land-use conditions, engineering projects, and other datasets provided by management authorities. The collection methods mainly involve inter-departmental agreements and exchanges, and the data content, format, and storage methods are diverse. The challenges related to social data sharing are significant, exacerbated by a lack of comprehensive planning and sustainable institutional frameworks. While some scholars have proposed various compositions of spatial-gene data sources from their individual perspectives, satisfying the basic requirements for planning research and formulation, the overall dataset still shows “sparsity” and lacks the necessary conditions to support the comprehensive exploration of spatial-gene data.

In the face of the complexity of the raw data of spatial genes and the lack of collaboration among them, the crucial point for solving the problem lies in establishing unified norms of data scope, quality and structure, enabling scholars to conduct the same calculations and analyses within this framework. Grounded in the ontological concept that spatial genes represent “a stable and distinctive spatial-combination pattern formed throughout the historical trajectory of a city”, [14] we assert that the raw data sources for spatial genes should focus on the core principles of “diachrony”, “stability”, and “uniqueness”, rather than relying on fragmented snapshots of current urban forms as emphasized in existing research.

Specifically, we advocate for the integration of diverse cultural–historical data, spatio-temporal basic data, and behavioral-perception data to create a system of spatial-gene data sources that facilitate interaction between past and present while linking human beings with their environment (Table 1):

(1)

Systematically organize multi-source cultural–historical materials such as maps, scenic spots, poetry, photographs, and folk art; analyze implicit spatial elements, relationships, and connotative information within these materials to trace urban development’s evolution and historical context;

(2)

Collect contemporary three-dimensional morphological data through digital orthophotos, oblique imagery, and laser point clouds to gain insights into trends in urban spatial morphology;

(3)

Leverage internet-based big data, such as WeChat interactions, mobile phone signals, floating car datasets, and points of interest, to efficiently gather micro-scale individual behavior-perception information while interpreting superficial characteristics of urban spatial morphology from a social perspective.

Table 1. The system of spatial-gene input data.

Research Dimension	Data Sources	Specific Data Resources	Data Type	Data Accuracy *	Access Channels
Diachrony	Cultural–historical data	Map	Image	LOD0, LOD1	All kinds of cultural–historical data mainly originate from libraries, museums, and archives in major cities and universities. For instance, the website of the Library of the National University of Singapore openly offers the collation results of ancient books, with those collated by Zhonghua Book Company as the core and covering those of multiple professional publishing houses.
		Eight Sights	Text, image	LOD0, LOD1, LOD2
		Poetry	Text	LOD0, LOD1, LOD2, LOD3
		Photo	Image	LOD0, LOD1, LOD2, LOD3
		Genre Painting	Image	LOD0, LOD1, LOD2, LOD3
Stability	Spatio-temporal basic data	Digital elevation model	grid	LOD0, LOD1, LOD2, LOD3	Digital Elevation Models (DEMs), Digital Orthophoto Images (DOIs), oblique images, and laser point-cloud data can all be downloaded from the Geospatial Data Cloud Platform. The 3D model data of cities, buildings, roads, sites, and vegetation are often constituted by the 3D model databases of researchers and their institutions themselves, and the lower-precision 3D models can also be acquired from data sources such as Open Street Map (OSM).
		City 3D model	Information model	LOD0, LOD1
		Building 3D model	Information model	LOD0, LOD1, LOD2, LOD3
		3D model of road	Information model	LOD0, LOD1, LOD2, LOD3
		Three-dimensional model of the site	Information model	LOD0, LOD1, LOD2, LOD3
		3D vegetation model	Information model	LOD0, LOD1, LOD2, LOD3
		Digital orthophoto image	Raster	LOD0, LOD1
		Oblique image	Raster	LOD2, LOD3
		Laser point cloud	Grid	LOD2, LOD3
Uniqueness	Behavioral-perception data	WeChat	Associated to coordinates	LOD0, LOD1, LOD2, LOD3	WeChat data, mobile phone signaling data, and floating car data are generally provided by Tencent, communication operators, and traffic management departments through APIs for researchers’ use. Point of Interest (POI) and Baidu Heatmap data can be obtained via the Web service API, offered by Baidu. Flickr provides an API interface, allowing developers and researchers to access the public photo data on Flickr. Eye-tracker data is often collected by researchers through experiments on specific groups.
		Mobile phone signaling	Associated to coordinates	LOD0, LOD1, LOD2, LOD3
		Floating car	Associated to coordinates	LOD0, LOD1, LOD2, LOD3
		Points of Interest	Vector	LOD0, LOD1, LOD2, LOD3
		Baidu Heatmap	Image, associations to coordinates	LOD0, LOD1, LOD2, LOD3
		Flickr photo hot spots	Image, associate to coordinates	LOD0, LOD1, LOD2, LOD3
		Eye-tracker	Image, correlate to coordinates	LOD2, LOD3

Data Accuracy *: The input data accuracy refers to the data accuracy standard of CityGML [58].

Table 1. The system of spatial-gene input data.

Research Dimension	Data Sources	Specific Data Resources	Data Type	Data Accuracy *	Access Channels
Diachrony	Cultural–historical data	Map	Image	LOD0, LOD1	All kinds of cultural–historical data mainly originate from libraries, museums, and archives in major cities and universities. For instance, the website of the Library of the National University of Singapore openly offers the collation results of ancient books, with those collated by Zhonghua Book Company as the core and covering those of multiple professional publishing houses.
		Eight Sights	Text, image	LOD0, LOD1, LOD2
		Poetry	Text	LOD0, LOD1, LOD2, LOD3
		Photo	Image	LOD0, LOD1, LOD2, LOD3
		Genre Painting	Image	LOD0, LOD1, LOD2, LOD3
Stability	Spatio-temporal basic data	Digital elevation model	grid	LOD0, LOD1, LOD2, LOD3	Digital Elevation Models (DEMs), Digital Orthophoto Images (DOIs), oblique images, and laser point-cloud data can all be downloaded from the Geospatial Data Cloud Platform. The 3D model data of cities, buildings, roads, sites, and vegetation are often constituted by the 3D model databases of researchers and their institutions themselves, and the lower-precision 3D models can also be acquired from data sources such as Open Street Map (OSM).
		City 3D model	Information model	LOD0, LOD1
		Building 3D model	Information model	LOD0, LOD1, LOD2, LOD3
		3D model of road	Information model	LOD0, LOD1, LOD2, LOD3
		Three-dimensional model of the site	Information model	LOD0, LOD1, LOD2, LOD3
		3D vegetation model	Information model	LOD0, LOD1, LOD2, LOD3
		Digital orthophoto image	Raster	LOD0, LOD1
		Oblique image	Raster	LOD2, LOD3
		Laser point cloud	Grid	LOD2, LOD3
Uniqueness	Behavioral-perception data	WeChat	Associated to coordinates	LOD0, LOD1, LOD2, LOD3	WeChat data, mobile phone signaling data, and floating car data are generally provided by Tencent, communication operators, and traffic management departments through APIs for researchers’ use. Point of Interest (POI) and Baidu Heatmap data can be obtained via the Web service API, offered by Baidu. Flickr provides an API interface, allowing developers and researchers to access the public photo data on Flickr. Eye-tracker data is often collected by researchers through experiments on specific groups.
		Mobile phone signaling	Associated to coordinates	LOD0, LOD1, LOD2, LOD3
		Floating car	Associated to coordinates	LOD0, LOD1, LOD2, LOD3
		Points of Interest	Vector	LOD0, LOD1, LOD2, LOD3
		Baidu Heatmap	Image, associations to coordinates	LOD0, LOD1, LOD2, LOD3
		Flickr photo hot spots	Image, associate to coordinates	LOD0, LOD1, LOD2, LOD3
		Eye-tracker	Image, correlate to coordinates	LOD2, LOD3

Data Accuracy *: The input data accuracy refers to the data accuracy standard of CityGML [58].

The collection of data sources along these three dimensions will strengthen support for integrity and accuracy while facilitating targeted exploration of spatial-gene information.

Simultaneously, in alignment with the requirements for reliable storage and comprehensive data extraction, the aforementioned multi-source heterogeneous data must be converted into a computer-readable format through specific data-exchange standards after collection, subsequently entering specialized data storage centers such as data warehouses and data lakes. This process facilitates real-time monitoring of data integrity while fulfilling traceability requirements for various spatial-gene datasets; concurrently, it ensures high availability and scalability during storage, thereby mitigating potential data disasters resulting from accumulation.

To address the characteristics of structured, semi-structured, and unstructured data types, text processing and graphic recognition software based on platforms like Microsoft Access and ArcGIS can be employed for preprocessing tasks. Following this step, with reference to the semantic information accuracy of CityGML in different spatial hierarchy systems, the feature-matching and geographic-data-merging of multi-port basic data are conducted, generating a new dataset with enhanced spatial and attribute accuracy. Then, it is stored in a specific structure in a dedicated spatial-gene data storage platform, employing an OLE DB connection manager and providing data externally through an interface. Furthermore, all the text, images, and geographic data need to undergo quality inspection, spatial consistency comparison, and pre-storage in ArcGIS to guarantee the accuracy of the data and the dynamics of the system. In this respect, Conflation offers a technique for dealing with the discrepancies in data sources [59].

Thus, through a strict data submission and inspection mechanism, multi-source heterogeneous datasets are integrated into a unified and interoperable resource library that enables high-quality and efficient aggregation of extensive foundational spatial-gene data.

5. Mining: Analysis and Processing of Spatial-Gene Data

Subsequent to data collection and storage, it is essential to utilize computer software for conducting various data analyses and processing, as well as interpreting the spatial-combination patterns, morphological evolution trends, and profound structural insights embedded within the data. This approach aims to fully leverage the potential of spatial-gene data sources. Currently recognized as one of the most extensively pursued research endeavors, scholars often engage in customized identification and interpretation of spatial genes through methods such as landscape pattern index, shape index, image element decomposition, historical information translation, and spatial syntax analysis [60]. Consequently, this practice can lead to identical datasets being assigned divergent meanings, resulting in issues related to a disorganized and biased accumulation of spatial-gene data.

We assert that the apparent randomness in the current application of spatial-gene data mining tools is fundamentally rooted in the inconsistent understanding of spatial-gene ontology. Various scholars articulate their interpretations of spatial genes based on individual perspectives, thereby deviating from the theoretical core and semantic significance inherent to spatial genes.

The conceptual ontology of spatial genes mainly comprises three parts:

(1)

Spatial elements. Spatial genes are composed of various scales of spatial elements serving as fundamental constituents. In accordance with the research logic of urban morphology from the macroscopic to the microscopic, in combination with the identifiable contents at different LOD levels in CityGML, as well as different data-usage environments, such as comprehensive urban design, district urban design, block urban design, and architectural design, we have classified spatial genes into four spatial levels: urban, district, block, and building. We have also clarified the spatial elements, evaluation criteria, and typical cases related to each level of spatial genes (Table 2). This multi-dimensional integration reflects the hierarchical and self-organizing characteristics of spatial genes transitioning from individual components to a cohesive whole.

(2)

Spatial relationships. Spatial genes are embedded within sequences, angles, proportions, and other relational dynamics among different types of spatial elements, such as the reciprocal gaze between urban areas and landscape features; interdependencies between alleys and water systems; or enclosure relationships between buildings and courtyards. As these spatial elements continuously interact with their external environment, their relationships evolve from disorder towards order, thereby establishing genetic laws governing constrained, yet sustainable, urban space organization.

(3)

Spatial attributes. Spatial attributes encompass two components: the attributes of the elements (such as height, color, material, etc.) and the attributes of the spatial relationships (such as window-to-wall ratio, street height–width ratio, openness degree of the block, etc.).

However, only through the cross-correlation of these three parts, completing the construction and visualization of the “spatial-combination pattern”, can the information interaction of the semantic value of spatial genes be realized (Figure 3).

Table 2. The spatial elements related to each level of spatial genes.

Level of Spatial Genes	Spatial Elements	Sub-Forms of Spatial Elements	LOD Levels	Data-Usage Environments	Typical Cases
Urban	Mountain	Hill, low mountain, medium mountain, high mountain, etc.	LOD0	Comprehensive urban design	Landscape pattern gene: Mountain—waterbody—settlement
	Waterbody	Ocean, lake, reservoir, stream, river, etc.
	Settlement	City, town, village, etc.
	City wall	Palace city wall, Imperial city wall, Outer city wall, etc.
	City center	Political center, commercial center, cultural center, etc.
	District	Residential district, industrial district, commercial district, educational and cultural district, mixed district, etc.
	Transportation junction	Airport, railway station, highway station, wharf, etc.
	Greenway	Waterfront greenway, mountain and forest greenway, green land greenway, road greenway, rural greenway, historical and cultural greenway, etc.
District	Block	Commercial block, residential block, industrial block, educational block, etc.	LOD1	District urban design	Axis spatial gene: Landmark—square—road—building complex
	Park	Comprehensive park, specialized park, community park, pleasure garden, etc.
	Square	Religious square, assembly square, transportation square, memorial square, commercial square, leisure and entertainment square, etc.
	Building complex	Low-rise building complex, multi-story building complex, high-rise building complex, mixed building complex, etc.
	Road	Expressway, major arterial, minor arterial, local street, etc.
	Landmark	Landmark building, sculpture, geographical mark, monument, etc.
	Greenbelt	Street greenbelt, park greenbelt, protective greenbelt, scenic greenbelt, etc.
	Axis	Historical axis, natural axis, planning axis, etc.
	Transportation station	Subway station, bus station, light rail station, courier station, etc.
Block	Building	Super-tall building, tall building, multi-story building, low-story building, etc.	LOD2	Block urban design	Lifang gene: Building—plant—street—plot
	Street	Commercial street, living street, landscape street, transportation street, comprehensive street, etc.
	Plot	Commercial plot, residential plot, industrial plot, comprehensive plot, etc.
	Street furnishing	Street lamp, traffic indicator, pedestrian bridge, entrance and exit of underground transportation, advertising shop sign, news kiosk, bulletin board, gazebo, chair and stool, trash bin, telephone booth, mailbox, fountain, etc.
	Plant	Shrub, tree, grass, flower, etc.
Building	Door	Mobile door, folding door, closet door, partition door, swing door, etc.	LOD3	Architectural design	Courtyard gene: Courtyard—wall—roof—door—window
	Window	Casement window, hopper window, sliding window, louvre window, etc.
	Wall	Brickwork wall, paneled wall, stonework wall, wooden wall, concrete wall, etc.
	Step	Stone step, brick-built step, concrete step, wooden step, steel step, etc.
	Room	Bedroom, study room, living room, kitchen, dining room, office, classroom, balcony, bathroom, exhibition hall, etc.
	Roof	Flat roof, pitched roof, spatial curved roof, combined roof, etc.
	Courtyard	Enclosed courtyard, open courtyard, garden-style courtyard, patio, etc.
	Column	Square column, circular column, octagonal column, hexagonal column, etc.
	Staircase	Spiral staircase, curved staircase, suspended staircase, straight staircase, zigzag staircase, etc.
	Chimney	Round chimney, elliptical chimney, well-shaped chimney, etc.

Table 2. The spatial elements related to each level of spatial genes.

Level of Spatial Genes	Spatial Elements	Sub-Forms of Spatial Elements	LOD Levels	Data-Usage Environments	Typical Cases
Urban	Mountain	Hill, low mountain, medium mountain, high mountain, etc.	LOD0	Comprehensive urban design	Landscape pattern gene: Mountain—waterbody—settlement
	Waterbody	Ocean, lake, reservoir, stream, river, etc.
	Settlement	City, town, village, etc.
	City wall	Palace city wall, Imperial city wall, Outer city wall, etc.
	City center	Political center, commercial center, cultural center, etc.
	District	Residential district, industrial district, commercial district, educational and cultural district, mixed district, etc.
	Transportation junction	Airport, railway station, highway station, wharf, etc.
	Greenway	Waterfront greenway, mountain and forest greenway, green land greenway, road greenway, rural greenway, historical and cultural greenway, etc.
District	Block	Commercial block, residential block, industrial block, educational block, etc.	LOD1	District urban design	Axis spatial gene: Landmark—square—road—building complex
	Park	Comprehensive park, specialized park, community park, pleasure garden, etc.
	Square	Religious square, assembly square, transportation square, memorial square, commercial square, leisure and entertainment square, etc.
	Building complex	Low-rise building complex, multi-story building complex, high-rise building complex, mixed building complex, etc.
	Road	Expressway, major arterial, minor arterial, local street, etc.
	Landmark	Landmark building, sculpture, geographical mark, monument, etc.
	Greenbelt	Street greenbelt, park greenbelt, protective greenbelt, scenic greenbelt, etc.
	Axis	Historical axis, natural axis, planning axis, etc.
	Transportation station	Subway station, bus station, light rail station, courier station, etc.
Block	Building	Super-tall building, tall building, multi-story building, low-story building, etc.	LOD2	Block urban design	Lifang gene: Building—plant—street—plot
	Street	Commercial street, living street, landscape street, transportation street, comprehensive street, etc.
	Plot	Commercial plot, residential plot, industrial plot, comprehensive plot, etc.
	Street furnishing	Street lamp, traffic indicator, pedestrian bridge, entrance and exit of underground transportation, advertising shop sign, news kiosk, bulletin board, gazebo, chair and stool, trash bin, telephone booth, mailbox, fountain, etc.
	Plant	Shrub, tree, grass, flower, etc.
Building	Door	Mobile door, folding door, closet door, partition door, swing door, etc.	LOD3	Architectural design	Courtyard gene: Courtyard—wall—roof—door—window
	Window	Casement window, hopper window, sliding window, louvre window, etc.
	Wall	Brickwork wall, paneled wall, stonework wall, wooden wall, concrete wall, etc.
	Step	Stone step, brick-built step, concrete step, wooden step, steel step, etc.
	Room	Bedroom, study room, living room, kitchen, dining room, office, classroom, balcony, bathroom, exhibition hall, etc.
	Roof	Flat roof, pitched roof, spatial curved roof, combined roof, etc.
	Courtyard	Enclosed courtyard, open courtyard, garden-style courtyard, patio, etc.
	Column	Square column, circular column, octagonal column, hexagonal column, etc.
	Staircase	Spiral staircase, curved staircase, suspended staircase, straight staircase, zigzag staircase, etc.
	Chimney	Round chimney, elliptical chimney, well-shaped chimney, etc.

Upon the completion of ontology construction and visualization, it is essential to conduct corresponding data analyses in accordance with the attributes of spatial elements and spatial relationships, thereby achieving effective spatial-gene output data. Initially, spatial analysis techniques such as element analysis, structural analysis, and network analysis are employed to compute the attributes of various elements and relationships, such as window-to-wall ratio, street height–width ratio, openness degree of the block, etc. This process transforms high-dimensional urban genetic morphemes, syntax, and semantics into low-dimensional data while facilitating loose coupling and comprehensive analyses of multi-source heterogeneous data that encompass text, images, raster formats, vector representations, and information models. Building on this foundation, clustering algorithms—including tagging methods, fuzzy logic approaches, K-means clustering, and density-based techniques—are utilized to categorize and summarize the attribute values associated with spatial elements and relationships. Concurrently, the “space-nature-culture” interactive analytical framework is applied to discern underlying patterns as well as requisite conditions for the formation of spatial-combination patterns while extracting derivative information and new insights (

Table 3. The main analysis methods of the spatial gene.

Analysis Methods	Commentary
Element Analysis	Investigate the elemental composition within a specific spatial system.
Structural Analysis	Calculate the proportion of each component in a particular spatial system.
Network Analysis	Abstract the relationships between spatial elements into networks and employ computational methods and indicators to describe the network structures.
Clustering Algorithm	Acquire the cross-relationships among different spatial elements and characteristic factors in a quantitative manner.
“Space-Nature-Culture” Interactive Analysis	Disclose the mechanism of the interactive influence of various urban systems behind spatial genes.

). Subsequently, through advanced data fusion and integration technology in accordance with the data output rules of gene entries, indexing scenarios, feature factors, and functional mechanisms, the encoding of multi-scale, multi-type, and multi-temporal spatial-gene data is accomplished effectively (

Table 4. The data output rules of the spatial gene.

Ontology Elements	Commentary	Data Type
Gene entries	As the category criterion for the attribute characteristics of space genes, it is indicated in textual form and is the basic element composition or spatial organizational object of spatial genes.	Text
Indexing scenarios	It is employed to depict the spatio-temporal coordinates of typical and representative urban scenes controlled by spatial genes, which is of vital significance for comprehending the degree to which spatial genes exert their effects and the conditions they demand.	Vector, Image
Feature factors	It is composed of the configuration rules of several elements refined by spatial genes and their index values.	Text, value
Functional Mechanisms	It is necessary to employ the “space-nature-culture” system to disclose the mechanism of the interactive influence of various urban systems behind spatial genes.	Text

). Ultimately, in conjunction with a well-structured workflow intervention, such extraction processes transform into repeatable and highly scalable operations, enabling comprehensive analyses of multi-source heterogeneous datasets within a unified system, resulting in standardized outputs for spatial-gene data products.

It is worthy to mention that with the continuous improvement of the types, speed, and quantity of urban basic data, it brings about significant challenges to spatial-gene analysis and reflection capabilities. To enhance the analysis efficiency, accuracy, and insights, we propose a spatial-gene data-analysis framework based on big data technology, specifically encompassing both in-database analysis and out-of-database analysis [61]. (1) In-database analysis refers to a mode of establishing a data warehouse within the organization and mining spatial-gene big data through a data analysis platform. This is applicable in scenarios dominated by structured data, with low real-time requirements and privacy demands. (2) Out-of-database analysis is the trend in spatial-gene research and practice of transforming data types from structured data to unstructured data. Therefore, it is necessary to resort to distributed computing services that support efficient real-time storage and analysis, such as stream computing, MySQL, MapReduce based on Hadoop, etc., to facilitate cross-organizational collaborative processing of spatial-gene big data by researchers. Currently, the Urban Spatial Research Institute of Southeast University has established a dedicated data center and computing platform for spatial genes. The platform offers a theoretical peak performance of 1085 TFLOPS in FP32 precision and features petabyte-scale storage through a distributed cluster architecture for the high-performance computing and data storage requirements of spatial-gene research. Additionally, it employs a data security protection platform based on backup all-in-one machines to ensure the high performance and stable operation of the platform. Meanwhile, this platform adopts virtualization and cloud technologies, enabling researchers to obtain computing resources through remote access and achieve collaborative work [62]. In the future, if there are higher computing demands, we will expand this platform and leverage cloud computing service platforms to promote the transformation of spatial-gene analysis capabilities from the current transformed level to the experienced level [63,64].

6. Application: Circulation and Sharing of Spatial-Gene Data

Spatial-gene data can only realize its practical value when it is transformed, applied, and circulated within specific contexts, such as data compilation in the realm of planning formulation, plan generation, and scenario simulation; land use regulation, planning permissions, and monitoring and evaluation in the domain of planning management; as well as preventive protection and intelligent restoration in the field of planning and construction. Historically, these scenarios have often been governed by disparate departments, leading to fragmentation among their respective operations. Consequently, issues such as inconsistent data standards and misaligned data products have arisen in practical applications. Therefore, it is imperative to clarify the collaborative relationships among multiple entities and establish a multi-agent application framework for spatial-gene data. Drawing upon the actual efficacy of spatial genes in urban planning and construction processes, we propose a comprehensive perspective on the application and transformation of spatial-gene data with the Planning Support System at its core while utilizing CIM and Spatial Gene Pool as supportive elements to facilitate the external circulation of spatial-gene data products (Figure 4).

6.1. Planning Support System

For urban planning practice, a cognitive psychological process dominated by urban planners is requisite, encompassing cognition, learning, problem-solving, decision-making, and design [65]. Planning Support System is a comprehensive framework that synthesizes a variety of data, methodologies, and models through the integration of information technology and management science. This system establishes a mapping relationship between data and planning problems, thereby enabling precise and efficient decision-making in urban planning and design. To address the issues of fragmented and ineffective spatial-gene data in practice, it is essential to leverage the Planning Support System to facilitate the seamless integration of the “description—explanation—evaluation—design” workflow within urban planning while achieving the effective transmission of spatial genes in urban development through a targeted approach.

Specifically, this operational process should be propelled by the requirements for spatial-gene data in planning and design undertakings. It regards the harmonious interaction among urban spaces, natural environments, and historical and cultural elements as a constraint objective. Moreover, it is essential to clarify the supporting relationships between upstream and downstream operations while integrating the functionalities of GIS with the capabilities of Web servers for comprehensive data collection, analysis, and exchange. This will enable an assessment of the developmental trends in spatial-combination patterns and their adaptability within complex urban systems. Ultimately, this approach aims to anchor local information-bearing spatial genes through multi-level reconstructions of spatial systems while ensuring full-process alignment between spatial genes and planning practical applications in various dimensions such as goal orientation, foundational platforms, data systems, and implementation strategies (Figure 5).

6.2. City Information Model (CIM)

As the foundational operational platform for urban planning, construction, management, and operation in the current context, CIM can function as a critical information infrastructure for the transformation of spatial-gene data. However, since spatial genes are inherently more challenging to integrate into the basic data framework of the CIM platform compared to 2D GIS data, 3D model data, BIM data, and others, there is a pressing need to explore more flexible technical interfacing methods. For example, as the Ministry of Housing and Urban-Rural Development of China proposed the deployment requirements for development interfaces (Web API), we propose that supporting the application of spatial-gene data in planning practice through network application programming interfaces (APIs) categorized as “external references” for resource access, data analysis, and project integration within CIM represents a more pragmatic solution.

Specifically regarding resource access, various types of spatial-gene information should be encoded to elucidate their regional distribution, element composition, and functional mechanisms across multiple scales and temporal dimensions; this information can then be accessed by planners, managers, and members of the public alike. In terms of data analysis—leveraging ArcGIS10.8 software’s spatial-analysis techniques—different categories of spatial data and models can be integrated to facilitate intuitive representation and efficient transmission of spatial-gene information via visualization techniques. Concerning projects related to spatial genes, linking relevant planning project datasets provided by regulatory authorities, will furnish essential case studies necessary for advancing urban planning and design initiatives across diverse regions (Figure 6).

6.3. Spatial Gene Pool

In addition to the CIM platform, spatial-gene data products require the support of specialized data fusion and sharing hubs to facilitate information reutilization and collaborative efforts among various institutions. In this context, pioneering fields such as genomics and space science have accumulated substantial experience in developing collaborative platforms and gene banks. Notably, the establishment of the National Genomics Data Center and the National Space Science Data Center has resulted in a national-level gene bank for the aggregation and management of extensive genetic data, leveraging collective advantages to enhance professional specialization and knowledge sharing within related domains.

Spatial-gene research and practice can utilize this framework to establish a socially-driven, non-profit Spatial Gene Pool in the form of an integrated application platform that provides remote network data services encompassing unified storage, integration and extraction, as well as shared applications for a diverse array of research participants.

(1)

The unified storage module includes regional and urban sample libraries alongside multi-source data archives;

(2)

The integration and extraction module features various spatial-gene analysis tools software along with online analytical platforms;

(3)

The shared application module comprises spatial-gene banks, research project repositories, engineering project databases, scientific literature collections, etc. (Figure 7).

Regarding the multi-agent interaction mechanism of the Spatial Gene Pool, specific social entities, such as the Spatial Genetics Society, undertake its specialized management, while technology companies, as auxiliary forces, offer the necessary data support, and the government-led CIM provides a considerable amount of fundamental data. Based on this platform’s functionalities, researchers from around the globe can obtain developer or accessor permissions for spatial-gene data but must upload their contributions to the database for verification prior to publication. This process not only ensures high standards of data quality and security but also facilitates seamless integration between data resources and databases. Ultimately, through the multi-dimensional application of spatial-gene data, a partnership between the public and the government is established, allowing social forces to engage in urban planning and construction based on spatial genes and contributing collective wisdom to the urban cultural diversity and sustainable development (Figure 8).

7. Conclusions

Data collaboration has emerged as a fundamental undertaking for the sharing and value creation of current spatial-gene data. Against the backdrop of the urban planning data-system transformation triggered by the digital age, a comprehensive understanding of the remarkable capabilities of vast amounts of data in comprehending, predicting, and adapting to human preferences, along with reducing costs and enhancing work efficiency, is indispensable for addressing data-level issues such as data scarcity, time-consuming data cleaning and processing, professional threshold barriers, and unclear decision impacts in the application of digital technologies. Subsequently, the systematic construction of the data collaboration path for spatial-gene research and practice holds dual exigency in both theoretical and practical respects. This system is also of paramount importance for the dissemination and development of the academic concept.

By conducting a comprehensive review of the literature, standards, and practical projects, we analyzed the specific issues existing in the spatial-gene data architecture. Firstly, there was inconsistency in data input and output among scholars regarding spatial genes, which gave rise to widespread disorderly accumulation of information and consequently led to “logical data isolation” in their research data. Secondly, since the existing spatial-gene analysis and decision-making system is rather general and fragmented, researchers and planners carried out their work separately, resulting in insufficient continuity in many planning studies and practices and giving rise to “physical data isolation”. Thirdly, due to the ambiguity of data products and application scenarios, spatial-gene data have difficulty entering subsequent construction engineering links and are even more challenged to play a more specific role in urban construction. Finally, although the number of scholars and institutions engaged in spatial-gene research is increasing rapidly, the absence of a collaborative working environment has, to a certain extent, impeded data sharing and academic communication.

Based on this, we have constructed a data collaboration framework for spatial-gene research and practice from the perspective of the data value chain. This framework encompasses three primary components: data convergence, data mining, and data application. Through the systematic establishment of data resources in a top-down and bottom-up manner, it integrates the multi-source and heterogeneous data scattered throughout, conducts data analysis and processing based on certain rules, and forms highly available and scalable data products that can be flexibly applied in planning and design practices. From the perspective of the welfare effects of multiple entities, institutions from different regions can better utilize the data resources of spatial genes and enhance the efficiency and innovation of research and practice; management departments can also take spatial genes as the object to better implement the protection of urban historical culture. This innovation in the “decentralized” data system model actually consists of three aspects: data collaboration, business collaboration, and entities’ collaboration. Through the comprehensive planning of the three, it ultimately promotes the information integration, business optimization, decision support, and cooperation expansion of spatial genes and perfects the digital system of spatial-gene research and practice.

Although the aforementioned contributions have been achieved, owing to the considerable complexity of spatial-gene data collaboration, it must be conceded that our research still harbors certain limitations. Firstly, spatial-gene data collaboration not only encompasses theoretical innovation and technical implementation but also necessitates multifaceted support, such as that from society, organizations, and policies. Evidently, this paper has not overly dwelt on the aspects related to them. Subsequently, the methods of element analysis, structural analysis, and network analysis proposed in the data analysis aspect are undoubtedly not exhaustive. Particularly, emerging technologies like big data and AI are requisite to enhance the accuracy and efficiency of spatial-gene data mining. Finally, the spatial-gene data collaboration path proposed in this paper is primarily based on the workflow relationship in urban design and has not comprehensively contemplated the disparities in its application in other types of planning.

Amid the advancement of digital planning, the path of spatial-gene data collaboration still demands further expansion. The foreseeable development tendencies in the future encompass the following: establishing a data sharing ecosystem to facilitate cross-border, cross-field, and cross-disciplinary intercommunication of spatial-gene data; reinforcing the technical measures for data security supervision and right confirmation to preclude the diversion and alienation of spatial-gene data; etc. As the quantity of institutions participating in spatial-gene research keeps escalating, it is believed that through continuous research and adjustment, the data collaboration path of spatial genes will be gradually optimized, and the digitalization level will be continuously enhanced. This will also foster the development and dissemination of the spatial-gene concept itself.