A wetland analysis adaptive scheduling method and system based on knowledge graph tracing

Abstract

This application belongs to the field of computer science and discloses a wetland analysis adaptive scheduling method and system based on knowledge graph traceability. The method first constructs a wetland computing ontology that integrates OGC API-Processes, wetland ontology and traceability information to describe the semantics and traceability relationship between wetland computing services, tasks and data; and constructs an executable knowledge graph that integrates services and wetland data, executes computing tasks and tracks traceability under the knowledge graph framework, and stores execution information to form a dynamic traceability graph; when wetland data update is detected, depth-first search is used to reversely traverse the traceability graph to deduce the affected tasks and data and obtain a local subgraph; then an adaptive scheduling algorithm based on the Kahn algorithm is used to judge and mark the set of tasks that need to be re-executed according to the timeliness of the task itself and upstream data / tasks; finally, the marked tasks are scheduled and executed according to the determined adaptive order, and the data is updated to achieve dynamic adaptive scheduling of wetland analysis.

Description

Technical Field

[0001] The present invention relates to the fields of computer science, geographic information systems and knowledge engineering, and in particular to a wetland analysis and adaptive scheduling method and system based on knowledge graph tracing. Background Art

[0002] In recent years, scholars have begun to explore the use of graph structures to represent and manage computational entities that can perform specific operations and the relationships between them, allowing traditional knowledge graphs to evolve from static information repositories to dynamic, executable systems. Currently, the main methods for integrating computational entities into knowledge graphs include embedded coding and plug-in services. Embedded coding embeds computational logic directly into the graph, which achieves automation but has maintenance complexity and security risks. Plug-in services connect independent services through semantic Web service descriptions, which improves reusability but traditional descriptions only support syntactic-level integration. Semantic Web services significantly improve service interoperability by introducing semantic information such as functions, I / O parameters, and execution conditions.

[0003] Wetland computing faces the challenge of single functional entities being unable to address complex problems. Business logic is required to coordinate multiple computing entities for collaborative solutions. Traditional approaches rely on expert experience for fixed orchestration (such as icon-guided modeling), while semantic technology enables automated process combination. While existing research has constructed service-integrated knowledge graphs for intelligent combination, applications in the geographic information field remain limited to rigid processes for remote sensing image analysis.

[0004] The dynamic nature of wetland ecosystems and their sensitivity to natural and human disturbances present significant challenges for their monitoring and management. While integrating data and computational entities into knowledge graphs supports dynamic updates, it can trigger cascading effects in data flows. When source data changes, key issues such as synchronization and coordination between upstream and downstream computational entities, matching data availability with execution timing, and context-adaptive solution must be addressed. This poses new technical requirements for ensuring the timeliness and accuracy of wetland computations. For example, the sampling period of wetland monitoring data may not align with the execution time of wetland computational entities. When executing computational entities, linked data may be unavailable or invalid. Therefore, the solution process for the same wetland problem will differ in different contexts, requiring the solution environment of the wetland virtual geographic environment to adapt to dynamic contextual changes. Determining which affected computational entities need to be reactivated to ensure the synchronous update of all derived information and dependencies, thereby maintaining the timeliness and accuracy of wetland computations, remains a current challenge in wetland computations.

[0005] The current wetland computing services have the following problems:

[0006] First, existing semantic Web service ontologies (such as OWL-S and WSMO) are difficult to adapt to the particularity of wetland spatial processing services and lack runtime semantic expression;

[0007] Second, the traditional Web service architecture only supports simple parameter passing and cannot achieve direct interaction between computing entities and wetland monitoring data entities in the knowledge graph;

[0008] Third, the current system lacks an adaptive computing mechanism based on data cascade effects, and it is necessary to establish a traceability-driven dynamic execution framework to ensure information consistency and computing efficiency.

[0009] Therefore, wetland computing currently urgently needs a method that can effectively integrate wetland monitoring data and computing entities into a unified framework, and achieve dynamic adaptation and intelligent scheduling of wetland computing processes by tracking the evolution and provenance of data and tasks. Summary of the Invention

[0010] Based on this, the present invention proposes a wetland analysis adaptive scheduling method and system based on knowledge graph tracing, which can effectively integrate wetland monitoring data and computing entities into a unified framework, and realize dynamic adaptation and intelligent scheduling of wetland computing processes by tracking the evolution and tracing of data and tasks.

[0011] In a first aspect, the present invention provides a wetland analysis and adaptive scheduling method based on knowledge graph tracing, the technical solution of which includes the following steps:

[0012] (a) Construct a wetland computing ontology based on the OGC API-Processes standard, extending and integrating the wetland monitoring ontology and the W3C traceability model. Define the semantics and traceability relationships between wetland computing services, tasks, and data entities through parameter description classes, input / output description classes, and attribute mapping classes.

[0013] (b) Encapsulate the wetland analysis process into an executable wetland computing service that can be called remotely based on OGC API-Processes;

[0014] (c) Integrate wetland computing entities and wetland monitoring data to build an executable knowledge graph and describe entity relationships through ontology;

[0015] (d) Bind computing services and data entities in the knowledge graph to trigger tasks and track the execution process to obtain traceability associations of inputs, outputs, tasks, and service details;

[0016] (e) Store the traceability information to form a directed acyclic wetland computing traceability graph, where nodes are data or task entities and edges represent entity relationships;

[0017] (f) When a data entity update is detected, the affected entities and task entities are deduced based on the reverse traversal of the computational provenance graph to generate a local provenance subgraph;

[0018] (g) Topologically sort the task entities in the subgraph. If the last execution time of a task is earlier than the latest time of the upstream task entity or the input data entity, it is marked as needing to be re-executed;

[0019] (h) Scheduling and executing task entities that need to be re-executed in accordance with the execution order determined by the adaptive scheduling algorithm to update the affected wetland data entities.

[0020] As an optional implementation of the first aspect of the present application, step (a) also includes: defining a wetland computing service class corresponding to the OGC API-Processes processing concept and reusing its data attributes; defining a wetland computing task class as an active entity, which can execute wetland computing services and input and output entities through attribute association, and the entity points to the wetland data entity in the knowledge graph.

[0021] As an optional implementation of the first aspect of the present application, step (c) also includes constructing a knowledge graph framework: constructing a knowledge graph interaction mediator for OGC API-Processes, the knowledge graph interaction mediator is responsible for data interaction between the executable wetland computing service and the knowledge graph, and the data interaction is realized through input mediation algorithm and output mediation algorithm; the knowledge graph framework also includes a knowledge graph job manager, which is responsible for persisting the job information of the OGC API-Processes to the knowledge graph.

[0022] As an optional implementation of the first aspect of the present application, the input mediation algorithm receives the request parameter list of the OGC API-Processes as input, traverses the parameter list, and for complex object parameters identified as pointing to entities in the knowledge graph, retrieves the wetland data entity attribute values ​​corresponding to the complex object parameters from the knowledge graph according to the attribute mapping defined in the wetland computing ontology, and assigns the wetland data entity attribute values ​​to the corresponding attributes of the complex object parameters to generate an input parameter list suitable for execution of the OGC API-Processes.

[0023] As an optional implementation of the first aspect of the present application, the output mediation algorithm receives the execution result list of the OGC API-Processes as input, traverses the output parameters in the result list, and for output parameters identified as complex object forms, queries the output description and its corresponding attribute mapping in the associated wetland computing ontology, creates a new wetland data entity instance in the knowledge graph and sets its type, uses the attribute values ​​of the complex object parameters as the attribute values ​​of the new entity instance according to the attribute mapping, and inserts the triple of the new entity instance and the attribute value into the knowledge graph.

[0024] As an optional implementation of the first aspect of this application, the knowledge graph job manager inherits from the BaseManager class of the Pygeoapi framework, and sends SPARQL statements to the endpoint of the knowledge graph through rewriting functions. The rewriting functions include add_job, update_job, delete_job, get_jobs and get_job_result, thereby realizing persistent management of the OGC API-Processes job in the knowledge graph.

[0025] As an optional implementation of the first aspect of the present application, the SPARQL statements sent by the knowledge graph job manager to the endpoint of the knowledge graph include: using the INSERT statement to create a new job instance; using the DELETE statement to delete the original job instance and using the INSERT statement to add the updated job instance; using the DELETE statement to delete the job instance; using the SELECT statement to request a list of job instances; and requesting the execution results of the job instance.

[0026] In a second aspect, an embodiment of the present application provides a wetland analysis and adaptive scheduling system based on knowledge graph tracing, including:

[0027] The wetland computing semantics and service construction module is configured to: build a wetland computing ontology, extend and integrate the wetland monitoring ontology and the W3C traceability model based on the OGC API-Processes standard, define the semantics and traceability relationships between wetland computing services, tasks, and data entities through parameter description classes, input / output description classes, and attribute mapping classes; and encapsulate the wetland analysis process into an executable wetland computing service that can be remotely called based on the OGC API-Processes standard.

[0028] An executable knowledge graph and traceability record module is connected to the wetland computing semantics and service construction module and is configured to: integrate wetland computing entities and wetland monitoring data to build an executable knowledge graph, and describe entity relationships through ontology; bind computing services and data entities to trigger tasks in the knowledge graph, and track the execution process to obtain traceability associations of inputs, outputs, tasks, and service details; store traceability information to form a directed acyclic wetland computing traceability graph, where nodes are data or task entities and edges represent entity relationships;

[0029] The adaptive update and scheduling execution module is connected to the executable knowledge graph and traceability record module, and is configured to: when a data entity update is detected, deduce the affected entities and task entities based on the reverse traversal of the calculated traceability graph to generate a local traceability subgraph; topologically sort the task entities in the subgraph, and if the last execution time of the task is earlier than the latest time of the upstream task entity or the input data entity, mark it as needing to be re-executed; schedule and execute the task entities that need to be re-executed in accordance with the execution order determined by the adaptive scheduling algorithm to update the affected wetland data entities.

[0030] In a third aspect, an embodiment of the present application provides an electronic device comprising a processor, a memory, and a program or instruction stored in the memory and executable on the processor, wherein when the program or instruction is executed by the processor, the steps of the method described in the first aspect are implemented.

[0031] In a fourth aspect, an embodiment of the present application provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps of the method described in the first aspect are implemented.

[0032] Compared with the prior art, the present invention has the following beneficial effects:

[0033] (1) An innovative traceability perception wetland computing ontology was constructed. By integrating the OGC API-P standard, PROV model and wetland monitoring ontology, a unified semantic description framework was established, achieving human-computer understandable service interface description and efficient integration of computing entities.

[0034] (2) Based on the OGC API-Processes framework, we innovatively implemented the service encapsulation of wetland computing entities. Through the independently developed knowledge graph interaction mediator, we broke through the technical bottleneck of traditional Web services and built an executable knowledge graph architecture that supports two-way data interaction between API-P services and knowledge graphs.

[0035] (3) A traceability-driven adaptive computing method is proposed. By dynamically tracking data dependencies and analyzing timeliness, it intelligently identifies and executes only the affected computing tasks, effectively solving the data cascade problem in wetland environments and significantly improving computing efficiency while ensuring the consistency of computing results. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 This is a sequence diagram of synchronous and asynchronous execution of OGC API-Processes in an embodiment of the present invention;

[0037] Figure 2 An executable knowledge graph framework in an embodiment of the present invention;

[0038] Figure 3 This is a diagram showing the entities involved in the traceability diagram in an embodiment of the present invention;

[0039] Figure 4 A structural diagram of a wetland analysis and adaptive scheduling system based on knowledge graph tracing provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0040] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0041] The terms "first", "second", etc. in the specification and claims of this application are used to distinguish similar objects, and are not used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable under appropriate circumstances, so that the embodiments of the application can be implemented in a sequence other than those illustrated or described here. In addition, "and / or" in the specification and claims represents at least one of the connected objects, and the character " / " generally represents that the objects associated before and after are in a kind of "or" relationship. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically limited.

[0042] Example 1

[0043] The present invention provides a wetland analysis adaptive scheduling method based on knowledge graph traceability, including constructing a traceability-aware wetland computing ontology, constructing an executable knowledge graph framework based on API-P, and implementing a traceability-driven knowledge graph adaptive computing method.

[0044] (1) Wetland computing ontology with traceability perception

[0045] Constructing a unified semantic description model for wetland computing is a primary challenge in realizing an executable knowledge graph for wetlands. To optimize the semantic expression of wetland computing, enabling both humans and machines to more intuitively understand service interface functionality, facilitate computing entity integration, and subsequently enable provenance-based adaptive execution, this paper designs a provenance-aware wetland computing ontology. This ontology builds on and expands upon concepts from the OGC API-Processes (API-P) standard, integrating a wetland monitoring ontology and the W3C Data Provenance (PROV) model.

[0046] It's important to note that Process, Job, and Result are three key concepts in the API-P architecture. A Process defines a remotely callable processing service and its metadata; a Job records the status of a specific execution instance; and a Result stores output in JSON format. These three elements form the execution framework of the Wetland Computing service.

[0047] To describe the semantics of provenance information, this paper reuses the W3C PROV model. The PROV model primarily includes three core concepts: activity (prov:Activity), entity (prov:Entity), and agent (prov:Agent), along with seven interdependent relationships between these concepts. This model allows for describing provenance information from the perspectives of data flow, processing flow, and responsibility flow. This paper reuses the PROV model and optimizes its concepts and relationships, minimizing information redundancy while fully documenting provenance semantics to meet the specific needs of wetland computing.

[0048] This paper proposes an ontology specifically designed for wetland computing executable knowledge graphs. This ontology incorporates the OGC API-Processes standard and integrates with the wetland monitoring ontology to facilitate the effective association of computational entities with monitoring data. Furthermore, the wetland computing ontology emphasizes the importance of provenance information. By reusing the PROV ontology, it ensures transparency and traceability of data and computational entities during wetland computing.

[0049] The Wetland Computing Service (wco:Process) represents a network service that can process, analyze, and simulate wetland data and dynamically update the knowledge graph. It corresponds to the API-P Process. The data attributes of the wco:Process directly reuse the attributes of the API-P Process. The Parameter Description (wco:ParameterDescription) class has two direct subclasses: the Input Description (wco:InputDescription) class and the Output Description (wco:OutputDescription) class. The Parameter Description class is linked to the class in the Wetland Monitoring Ontology through the wco:kg_uri attribute, which indicates the type of the wetland data entity corresponding to the parameter in the knowledge graph. The Property Mapping class (wco:PropertyMap) specifies the mapping rules between API-P input and output parameter names (wco:prop_name) and wetland monitoring ontology property URIs (wco:prop_uri). Specifically, wco:prop_uri corresponds to the attributes of wco:kg_uri, and wco:prop_name corresponds to the parameters directly used by the Wetland Computing Service.

[0050] The Wetland Computing Service provides the ability to perform wetland analysis and knowledge graph updates, and is an abstract computational entity within the knowledge graph. A wetland computing task (wco:Task) is an activity (prov:Activity) that leverages this capability and is a specific computational entity. Tasks can be directly executed by binding (prov:used) a wco:ProcessExecute entity to a specific data entity within the knowledge graph. The wco:ProcessExecute entity can define multiple specific inputs (wco:Input). Each input (wco:Input) entity is described by a corresponding wco:InputDescription entity and points to a specific wetland data entity within the knowledge graph through the wco:kg_uri attribute. This entity belongs to the wetland monitoring ontology class pointed to by the wco:kg_uri attribute of the corresponding input description entity. The wco:Job class describes the runtime semantics of the task entity and is generated (prov:wasGeneratedBy) by the task entity, corresponding to an API-P job, thus reusing the job's data attributes. At the same time, the job class can connect to multiple outputs (wco:Output). Similar to the input, the output points to the specific wetland data entity in the knowledge graph through the wco:kg_uri attribute.

[0051] (2) Executable knowledge graph framework based on API-P

[0052] The wetland computing entities referred to in this paper differ from common network services in the computer field. On the one hand, wetland computing entities are executed directly within the knowledge graph, rather than being static services. On the other hand, wetland computing entities are not just tools for processing and executing specific data; they also have the ability to transform the knowledge graph.

[0053] like Figure 1 As shown, API-P supports both asynchronous and synchronous execution modes. The former is targeted at simple, short-running computations, while the latter is suitable for complex, resource-intensive computations that require longer processing times. Synchronous mode establishes a persistent connection and directly returns computation results, making it suitable for complex computations. Asynchronous mode immediately returns the Job ID and disconnects, supporting status queries and result retrieval, making it suitable for lightweight computations. Both modes manage computation tasks through Job objects.

[0054] The present invention designs a knowledge graph interaction mediator to solve the data compatibility problem between API-P (JSON) and knowledge graph (RDF), such as Figure 2 As shown in the figure, its core features include: 1) a mediator built on PygeoAPI to implement bidirectional data format conversion; 2) a knowledge graph job manager (KGManager) that rewrites five key functions (add_job / update_job / delete_job / get_jobs / get_job_result) and uses SPARQL statements to implement graph persistence management for jobs. Specifically, INSERT / DELETE / SELECT statements are used to implement job creation, status updates, and result queries, respectively, completing the full lifecycle management of API-P jobs to the knowledge graph.

[0055] The wetland computation entity of this invention requires wetland data entities in the knowledge graph to perform computations. Neither inline nor reference modes can directly process knowledge graph entities. Therefore, when describing an APIP-P, each input description (wco:InputDescription) and output description of a wco:Process defines the wetland data entity concept to which the parameter points (wco:kg_uri), as well as a mapping (wco:propertyMap) between the wetland data entity attributes and the API-P complex object input properties. This invention designs input and output mediation algorithms to implement data interaction between the API-P and the knowledge graph. The input mediation algorithm receives an API-P and its associated request parameter list as input, identifies parameters that are complex structured objects and associated with entities in the knowledge graph, accesses the knowledge graph based on the property mapping in the input description, retrieves the actual values ​​associated with the data attribute URIs, assigns these values ​​to the corresponding input object attributes of the API-P, and returns a list of input parameters suitable for API-P execution. The output mediation algorithm integrates the API-P execution results into the knowledge graph and traverses the output parameters in the execution results.

[0056] (3) Tracing-driven adaptive computing method for knowledge graphs

[0057] like Figure 3 As shown in the figure, based on the wetland computing ontology and executable graph framework constructed in the previous article, the wetland knowledge graph forms a continuously evolving dynamic knowledge graph. In this dynamic knowledge graph, data entities follow a specific sequence and flow from one processing task to the next, thus forming an orderly data flow. According to the direction of the data flow, it can be divided into upstream and downstream. When the upstream data entity is updated, the downstream entity should also be recalculated. For example, when , indicating that the data entity d2 is outdated and the task needs to be re-executed Correspondingly, when an updated downstream data entity needs to be obtained, the evolution path diagram of the data entity needs to be traced back. This process is called traceability calculation.

[0058] The goal of provenance computing is to construct a provenance graph (PG). The provenance graph can be viewed as a four-tuple , which records the source, evolution process and mutual relationship of information in the knowledge graph. Represents a collection of data entities in the knowledge graph, Represents a collection of task entities in the knowledge graph; Is a computing unit, representing the task entity set To the data entity collection Mapping of ordered pairs of elements, Indicates a task The original data entity, Represents a derived data entity; Is a traceability unit, indicating that the data entity set To the task entity collection Mapping of ordered pairs of elements, represents the upstream task of data entity d, Represents downstream tasks. The topological structure of the traceability graph can be constructed by tracing back the computing units and traceability units in the dynamic knowledge graph. The traceability graph is in the form of a directed acyclic graph (DAG), whose nodes represent wetland computing task entities or wetland data entities, and edges represent derivative relationships or dependency relationships between entities. The present invention proposes a traceability calculation method based on the depth-first search (DFS) algorithm, which performs reverse traversal on the knowledge graph, starting from a specified data entity, searching for the derivative path of the data entity, and deducing the evolution process of the data entity to construct a local traceability subgraph of the traceability graph to show the topological structures of three traceability graphs at different generalization levels, including from the simplest linear chain to a multi-branch tree to a graph structure.

[0059] After obtaining the traceability graph of the data entity, the task execution can be adaptively adjusted according to the timeliness of the task entity to obtain accurate derivative data. The present invention designs a method for adaptive task execution based on the Kahn algorithm. The improved Kahn algorithm first calculates the in-degree of all task entities in the traceability graph and puts the task entities with in-degree 0 into the list S. Then, a task entity without upstream tasks is taken from S in a loop. , remove, find its downstream tasks. If removed_ After the in-degree of the adjacent downstream task becomes 0 after the related edges, S is added. The key is to compare the last execution time (or the latest time of the input data) of the downstream task entity with the upstream task entity directly connected to it. If the time of the downstream task entity is earlier than the latest time of any directly upstream task entity or input data, the downstream task entity is marked as needing to be re-executed. Repeat the above process until the list S is empty and all tasks that need to be re-executed are completed. According to the traceability graph of data entity d7 obtained by the traceability algorithm, the number above the task entity represents the last execution time, and the smaller the number, the earlier it is. According to the adaptive execution algorithm, the topological sorting of the traceability graph can be obtained. By comparing the last execution time of upstream and downstream tasks, the algorithm can determine which tasks need to be re-executed because the upstream data or tasks are outdated.

[0060] Example 2

[0061] See also Figure 4 , shown is a structural diagram of a wetland analysis and adaptive scheduling system based on knowledge graph tracing proposed in the second embodiment of this application. The system includes the following key modules:

[0062] The wetland computing semantics and service construction module 100 is configured to: construct a wetland computing ontology, extend and integrate the wetland monitoring ontology and the W3C traceability model based on the OGC API-Processes standard, define the semantics and traceability relationships between wetland computing services, tasks, and data entities through parameter description classes, input / output description classes, and attribute mapping classes; encapsulate the wetland analysis process into an executable wetland computing service that can be remotely called based on the OGC API-Processes standard;

[0063] The executable knowledge graph and traceability recording module 200 is connected to the wetland computing semantics and service construction module 100 and is configured to: integrate wetland computing entities and wetland monitoring data to construct an executable knowledge graph, and describe entity relationships through ontology; bind computing services and data entities to trigger tasks in the knowledge graph, and track the execution process to obtain traceability associations of inputs, outputs, tasks, and service details; store traceability information to form a directed acyclic wetland computing traceability graph, where nodes are data or task entities and edges represent entity relationships;

[0064] The adaptive update and scheduling execution module 300 is connected to the executable knowledge graph and traceability record module 200, and is configured to: when a data entity update is detected, deduce the affected entities and task entities based on the reverse traversal of the calculated traceability graph to generate a local traceability subgraph; topologically sort the task entities in the subgraph, and if the last execution time of the task is earlier than the latest time of the upstream task entity or the input data entity, mark it as needing to be re-executed; schedule and execute the task entities that need to be re-executed in accordance with the execution order determined by the adaptive scheduling algorithm to update the affected wetland data entities.

[0065] In the embodiment of the present application, a wetland analysis and adaptive scheduling system based on knowledge graph tracing can be a device, or a component, integrated circuit, or chip in a terminal. The device can be a mobile electronic device or a non-mobile electronic device. For example, the mobile electronic device can be a mobile phone, tablet computer, laptop computer, PDA, vehicle-mounted electronic device, wearable device, ultra-mobile personal computer (UMPC), netbook or personal digital assistant (PDA), etc. The non-mobile electronic device can be a server, network attached storage (NAS), personal computer (PC), etc., which is not specifically limited in the embodiment of the present application.

[0066] In the embodiments of the present application, a wetland analysis and adaptive scheduling system based on knowledge graph tracing can be a device having an operating system. The operating system can be an Android operating system, an iOS operating system, or other possible operating systems, which are not specifically limited in the embodiments of the present application.

[0067] The embodiment of the present application provides a wetland analysis adaptive scheduling system based on knowledge graph tracing, which can realize each process of a wetland analysis adaptive scheduling method based on knowledge graph tracing in the method embodiment. To avoid repetition, they will not be described here.

[0068] Optionally, an embodiment of the present application also provides an electronic device, including a processor, a memory, and a program or instruction stored in the memory and executable on the processor. When the program or instruction is executed by the processor, each process of the above-mentioned embodiment of the adaptive scheduling method for wetland analysis based on knowledge graph tracing is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be described here.

[0069] An embodiment of the present application also provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the various processes of the above-mentioned embodiment of the wetland analysis adaptive scheduling method based on knowledge graph tracing are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0070] The processor is the processor in the electronic device described in the above embodiment. The readable storage medium includes a computer-readable storage medium, such as a computer read-only memory (ROM), random access memory (RAM), a magnetic disk, or an optical disk.

[0071] It should be noted that, in this article, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the statement "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or device comprising the element. In addition, it should be noted that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in the opposite order according to the functions involved. For example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0072] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus the necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases the former is a more preferred embodiment. Based on this understanding, the technical solution of this application, or the part that contributes to the existing technology, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for enabling a terminal (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of this application.

[0073] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of this application, ordinary technicians in this field can also make many forms without departing from the purpose of this application and the scope of protection of the claims, all of which are within the protection of this application.

Claims

1. A wetland analysis and adaptive scheduling method based on knowledge graph tracing, characterized by: The following steps are involved: (a) Construct a wetland computing ontology based on the OGC API-Processes standard, extending and integrating the wetland monitoring ontology and the W3C traceability model. Define the semantics and traceability relationships between wetland computing services, tasks, and data entities through parameter description classes, input / output description classes, and attribute mapping classes. (b) Encapsulate the wetland analysis process into an executable wetland computing service that can be called remotely based on OGC API-Processes; (c) Integrate wetland computing entities and wetland monitoring data to build an executable knowledge graph and describe entity relationships through ontology; (d) Bind computing services and data entities in the knowledge graph to trigger tasks and track the execution process to obtain traceability associations of inputs, outputs, tasks, and service details; (e) Store the traceability information to form a directed acyclic wetland computing traceability graph, where nodes are data or task entities and edges represent entity relationships; (f) When a data entity update is detected, the affected entities and task entities are deduced based on the reverse traversal of the computational provenance graph to generate a local provenance subgraph; (g) Topologically sort the task entities in the subgraph. If the last execution time of a task is earlier than the latest time of the upstream task entity or the input data entity, it is marked as needing to be re-executed; (h) Scheduling and executing task entities that need to be re-executed in accordance with the execution order determined by the adaptive scheduling algorithm to update the affected wetland data entities.

2. The method according to claim 1, characterized in that Step (a) further comprises: Define the wetland computing service class corresponding to the OGC API-Processes processing concept and reuse its data attributes; The wetland computing task class is defined as an active entity, which can execute wetland computing services and input and output entities through attribute association. The entities point to wetland data entities in the knowledge graph.

3. The method for adaptive scheduling of wetland analysis based on knowledge graph tracing according to claim 1 is characterized in that: The step (c) also includes constructing a knowledge graph framework: Construct a knowledge graph interaction mediator for OGC API-Processes. The knowledge graph interaction mediator is responsible for data interaction between the executable wetland computing service and the knowledge graph. The data interaction is achieved through input mediation algorithm and output mediation algorithm. The knowledge graph framework also includes a knowledge graph job manager, which is responsible for persisting the job information of the OGC API-Processes into the knowledge graph.

4. The method for adaptive scheduling of wetland analysis based on knowledge graph tracing according to claim 3 is characterized in that: The input mediation algorithm receives the request parameter list of the OGC API-Processes as input, traverses the parameter list, and for complex object parameters identified as pointing to entities in the knowledge graph, retrieves the wetland data entity attribute values ​​corresponding to the complex object parameters from the knowledge graph according to the attribute mapping defined in the wetland computing ontology, and assigns the wetland data entity attribute values ​​to the corresponding attributes of the complex object parameters to generate an input parameter list suitable for execution of the OGC API-Processes.

5. The method for adaptive scheduling of wetland analysis based on knowledge graph tracing according to claim 3 is characterized in that: The output mediation algorithm receives the execution result list of the OGC API-Processes as input, traverses the output parameters in the result list, and for output parameters identified as complex object forms, queries the output description and its corresponding attribute mapping in the associated wetland computing ontology, creates a new wetland data entity instance in the knowledge graph and sets its type, uses the attribute values ​​of the complex object parameters as the attribute values ​​of the new entity instance according to the attribute mapping, and inserts the triple of the new entity instance and the attribute value into the knowledge graph.

6. The method for adaptive scheduling of wetland analysis based on knowledge graph tracing according to claim 3 is characterized in that: The knowledge graph job manager inherits from the BaseManager class of the Pygeoapi framework, and sends SPARQL statements to the endpoint of the knowledge graph by rewriting functions. The rewriting functions include add_job, update_job, delete_job, get_jobs and get_job_result, to achieve persistent management of the OGC API-Processes job in the knowledge graph.

7. The method for adaptive scheduling of wetland analysis based on knowledge graph tracing according to claim 6 is characterized in that: The SPARQL statement sent by the knowledge graph job manager to the endpoint of the knowledge graph includes: Use the INSERT statement to create a new job instance; Use the DELETE statement to delete the original job instance and use the INSERT statement to add the updated job instance; Use the DELETE statement to delete the job instance; Request a list of job instances using a SELECT statement; and Request the execution result of a job instance.

8. A wetland analysis and adaptive scheduling system based on knowledge graph tracing, characterized by: include: The wetland computing semantics and service construction module is configured to: build a wetland computing ontology, extend and integrate the wetland monitoring ontology and the W3C traceability model based on the OGC API-Processes standard, define the semantics and traceability relationships between wetland computing services, tasks, and data entities through parameter description classes, input / output description classes, and attribute mapping classes; and encapsulate the wetland analysis process into an executable wetland computing service that can be remotely called based on the OGC API-Processes standard. An executable knowledge graph and traceability record module is connected to the wetland computing semantics and service construction module and is configured to: integrate wetland computing entities and wetland monitoring data to build an executable knowledge graph, and describe entity relationships through ontology; bind computing services and data entities to trigger tasks in the knowledge graph, and track the execution process to obtain traceability associations of inputs, outputs, tasks, and service details; store traceability information to form a directed acyclic wetland computing traceability graph, where nodes are data or task entities and edges represent entity relationships; The adaptive update and scheduling execution module is connected to the executable knowledge graph and traceability record module, and is configured to: when a data entity update is detected, deduce the affected entities and task entities based on the reverse traversal of the calculated traceability graph to generate a local traceability subgraph; topologically sort the task entities in the subgraph, and if the last execution time of the task is earlier than the latest time of the upstream task entity or the input data entity, mark it as needing to be re-executed; schedule and execute the task entities that need to be re-executed in accordance with the execution order determined by the adaptive scheduling algorithm to update the affected wetland data entities.

9. An electronic device, characterized in that: It includes a processor, a memory, and a program or instruction stored in the memory and executable on the processor. When the program or instruction is executed by the processor, the steps of a wetland analysis adaptive scheduling method based on knowledge graph tracing as described in any one of claims 1 to 7 are implemented.

10. A readable storage medium, characterized in that: The readable storage medium stores a program or instruction, and when the program or instruction is executed by the processor, the steps of the wetland analysis adaptive scheduling method based on knowledge graph tracing as described in any one of claims 1 to 7 are implemented.

Images