Frozen earth field model operation method and device, storage medium and equipment

Abstract

The invention provides a frozen earth field model operation method and device, a storage medium and equipment, and the method comprises the steps: decomposing a frozen earth field scientific research demand inputted by a user according to a current frozen earth knowledge graph, and constructing a task logic diagram according to a decomposition result; according to the task logic diagram and domain knowledge in the current frozen soil knowledge graph, a task execution standardization template is generated, and the standardization template comprises a task execution standard processing flow and an exception processing flow; and inputting the standardized template and the current frozen soil knowledge graph into a pre-trained semantic model to obtain an output result corresponding to the scientific research demand in the frozen soil field. By introducing a standardized template and a frozen soil knowledge graph, a professional field semantic model is constructed, the understanding ability of the model to professional terms, complex concepts and logical relationships thereof is improved, and the limitation of an existing general large model in terms of semantic understanding and term mastering is overcome.

Description

Technical Field

[0001] This invention relates to the field of big data, and more specifically, to a method, apparatus, storage medium, and equipment for running a model in the field of permafrost. Background Technology

[0002] With the rapid development of natural language processing and artificial intelligence technologies, pre-trained large-scale language models have made breakthroughs in tasks such as language understanding, question answering, dialogue, and text generation. These models, pre-trained on massive amounts of text data, have acquired rich linguistic knowledge and semantic representation capabilities, and their performance continues to improve as the model size and training data increase. However, these general-purpose large models still face many challenges when processing complex scientific knowledge.

[0003] First, scientific research literature contains a large number of technical terms, complex concepts, and intricate logical relationships, making it difficult for general-purpose large-scale models to accurately understand their semantic information and conceptual connections. Second, scientific reasoning requires the ability to think, explore, and synthesize information in multiple rounds, while existing general-purpose large-scale models mainly focus on single-input-output tasks and are inadequate in tasks requiring multiple rounds of interaction and feedback, such as reasoning and planning. Therefore, there is an urgent need for a large-scale model system for scientific research that can understand technical terms, grasp complex conceptual logic, and has the ability to explore and optimize in multiple rounds, along with its corresponding operating method. Summary of the Invention

[0004] The purpose of this invention is to provide a method, apparatus, storage medium, and equipment for running models in the field of permafrost, so as to improve the above-mentioned problems.

[0005] To achieve the above objectives, the technical solutions adopted in the embodiments of the present invention are as follows:

[0006] In a first aspect, embodiments of the present invention provide a method for running a model in the permafrost field, the method comprising:

[0007] Based on the current permafrost knowledge graph, the research needs in the field of permafrost input by the user are decomposed, and a task logic graph is constructed based on the decomposition results. The decomposition results include the task type, the relevant input information of each sub-task, the task output rules, and the connection conditions between each sub-task.

[0008] Based on the task logic diagram and the domain knowledge in the current permafrost knowledge graph, a standardized template for task execution is generated, wherein the standardized template includes a standard processing flow and an exception handling flow for task execution.

[0009] The standardized template and the current permafrost knowledge graph are input into a pre-trained semantic model to obtain the output results corresponding to the scientific research needs in the field of permafrost.

[0010] Secondly, embodiments of the present invention provide a model operation device for permafrost regions, the device comprising:

[0011] The first processing unit is used to decompose the research needs in the field of permafrost input by the user based on the current permafrost knowledge graph, and construct a task logic graph based on the decomposition results; wherein, the decomposition results include task type, relevant input information of each sub-task, task output rules, and connection conditions between each sub-task.

[0012] The first processing unit is further configured to generate a standardized template for task execution based on the task logic diagram and the domain knowledge in the current permafrost knowledge graph, wherein the standardized template includes a standard processing flow and an exception handling flow for task execution.

[0013] The second processing unit is used to input the standardized template and the current permafrost knowledge graph into a pre-trained semantic model to obtain the output results corresponding to the scientific research needs in the field of permafrost.

[0014] Thirdly, embodiments of the present invention provide a storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described method.

[0015] Fourthly, embodiments of the present invention provide an electronic device, the electronic device comprising: a processor and a memory, the memory being used to store one or more programs; when the one or more programs are executed by the processor, the above-described method is implemented.

[0016] Compared to existing technologies, the permafrost domain model operation method, apparatus, storage medium, and device provided in this invention decompose user-input permafrost research needs based on the current permafrost knowledge graph, and construct a task logic graph based on the decomposition results. The decomposition results include task types, relevant input information for each sub-task, task output rules, and connection conditions between sub-tasks. Based on the task logic graph and domain knowledge in the current permafrost knowledge graph, a standardized template for task execution is generated, including a standard processing flow and anomaly handling flow. The standardized template and the current permafrost knowledge graph are input into a pre-trained semantic model to obtain the output results corresponding to the permafrost research needs. By introducing standardized templates and a permafrost knowledge graph, a professional domain semantic model is constructed, improving the model's ability to understand professional terminology, complex concepts, and their logical relationships, overcoming the limitations of existing general-purpose large models in semantic understanding and terminology mastery.

[0017] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention.

[0020] Figure 2 This is one of the flowcharts illustrating the permafrost field model operation method provided in this embodiment of the invention.

[0021] Figure 3 This is the second flowchart illustrating the permafrost field model operation method provided in this embodiment of the invention.

[0022] Figure 4 This is the third flowchart illustrating the permafrost field model operation method provided in this embodiment of the invention.

[0023] Figure 5 This is a schematic diagram of a unit for operating a model in the frozen soil field, as provided in an embodiment of the present invention.

[0024] In the diagram: 10-Processor; 11-Memory; 12-Bus; 13-Communication interface; 501-First processing unit; 502-Second processing unit. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0026] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0027] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this invention, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0028] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0029] In the description of this invention, it should be noted that the terms "upper," "lower," "inner," "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship in which the product of this invention is usually placed when in use. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limiting this invention.

[0030] In the description of this invention, it should also be noted that, unless otherwise explicitly specified and limited, the terms "set" and "connection" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0031] The following detailed description of some embodiments of the present invention is provided in conjunction with the accompanying drawings. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0032] This invention provides an electronic device, which may be a computer device or a server device. Please refer to... Figure 1This is a schematic diagram of the structure of an electronic device. The electronic device includes a processor 10, a memory 11, and a bus 12. The processor 10 and the memory 11 are connected via the bus 12. The processor 10 is used to execute executable modules, such as computer programs, stored in the memory 11.

[0033] Processor 10 can be an integrated circuit chip with signal processing capabilities. During implementation, each step of the permafrost domain model operation method can be completed through integrated logic circuits in the hardware or software instructions within processor 10. The aforementioned processor 10 can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0034] The memory 11 may include high-speed random access memory (RAM) and may also include non-volatile memory, such as at least one disk storage device.

[0035] Bus 12 can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus, etc. Figure 1 The symbol is represented by a single double-headed arrow, but this does not mean that there is only one bus 12 or one type of bus 12.

[0036] The memory 11 is used to store programs, such as the program corresponding to the permafrost model running device. The permafrost model running device includes at least one software function module that can be stored in the memory 11 in the form of software or firmware or embedded in the operating system (OS) of the electronic device. After receiving the execution instruction, the processor 10 executes the program to implement the permafrost model running method.

[0037] The electronic device provided in this embodiment of the invention may further include a communication interface 13. The communication interface 13 is connected to the processor 10 via a bus.

[0038] It should be understood that, Figure 1 The structure shown is only a partial schematic diagram of the electronic device; the electronic device may also include components that are larger than... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown. Figure 1 The components shown can be implemented using hardware, software, or a combination thereof.

[0039] The permafrost model operation method provided in this embodiment of the invention can be applied to, but is not limited to, applications in the field of permafrost. Figure 1 For the specific process of the electronic devices shown, please refer to [link / reference]. Figure 2 The model operation methods for the permafrost domain include S21, S22, and S23, which are described in detail below.

[0040] S21. Based on the current permafrost knowledge graph, decompose the research needs in the field of permafrost input by the user, and construct a task logic graph based on the decomposition results.

[0041] The decomposition results include the task type (e.g., retrieval task, paper writing task), the relevant input information of each subtask, the task output rules, and the connection conditions between each subtask.

[0042] The following examples illustrate the types of tasks, including literature analysis tasks, experimental design generation tasks, data analysis tasks, and text generation, etc.

[0043] Literature analysis employs natural language processing technology to extract and classify content from scientific literature, extracting key information and structuring it.

[0044] Experimental scheme generation: Based on task requirements, experimental schemes are automatically generated and report guidance is provided. This adopts a template-based generation algorithm and a rule-based scheme evaluation algorithm.

[0045] Data analysis provides data preprocessing and intelligent analysis functions, employing statistical and machine learning techniques for data mining.

[0046] Text generation, based on a generative model to assist in writing scientific research papers, employs template-based generation algorithms and semantic content filling algorithms.

[0047] S22, Based on the task logic diagram and the domain knowledge in the current permafrost knowledge graph, generate a standardized template for task execution.

[0048] Domain knowledge includes image features and text features, while standardized templates include the standard processing flow and exception handling flow for task execution.

[0049] Standardized processing flow refers to the routine operational procedures performed according to standard steps and processes during task execution. It ensures that the task proceeds smoothly according to predetermined steps and achieves the expected goals. For example, in the operation of a permafrost domain model, each step of the task is guided by domain knowledge from the task logic diagram and knowledge graph. Standardized processing flow typically includes clearly defined input, processing, and output steps to ensure the system can complete the task stably and effectively.

[0050] Anomaly handling procedures refer to the emergency response process a system employs when anomalies occur during task execution (such as incorrect input, missing data, or processing failure). These procedures typically include detecting the anomaly, identifying the anomaly type, taking corrective measures (e.g., reacquiring data, correcting model parameters, or adjusting execution strategies), and restoring normal operation. The purpose of this process is to ensure the system continues to function even when problems arise, preventing task failure due to anomalies.

[0051] S23 inputs the standardized template and the current permafrost knowledge graph into the pre-trained semantic model to obtain the output results corresponding to the scientific research needs in the field of permafrost.

[0052] In the permafrost domain model operation method provided in this embodiment of the invention, by introducing standardized templates and permafrost knowledge graphs, a professional domain semantic model is constructed, which improves the model's ability to understand professional terms, complex concepts and their logical relationships, and overcomes the limitations of existing general large models in terms of semantic understanding and terminology mastery.

[0053] Building upon the preceding text, this embodiment of the invention also provides an optional implementation method for the content in S23, as detailed below. S23 involves inputting a standardized template and the current permafrost knowledge graph into a pre-trained semantic model to obtain output results corresponding to research needs in the permafrost field, including: S231 and S232, as detailed below.

[0054] S231, based on the current permafrost knowledge graph, uses a heuristic algorithm to calculate the reasoning results of each subtask.

[0055] S232: Based on the reasoning algorithm in the current permafrost knowledge graph, combine the reasoning results of each subtask to obtain the output result.

[0056] In an optional implementation, the permafrost domain model operation method further includes: S24, which is described in detail below.

[0057] S24, Generate a visualization report of the reasoning process corresponding to the output results.

[0058] By graphically displaying the reasoning process and results, the model helps users understand its decision-making process. The interpretable reasoning layer and dynamic knowledge base enable reasoning and result backtracking based on task logic chains, significantly improving the model's interpretability and repeatability, and meeting the basic requirements of logical verification in scientific research.

[0059] Please refer to Figure 3 In an optional implementation, the permafrost domain model operation method further includes: S31 and S32, as detailed below.

[0060] S31, obtain user feedback information relative to the output results.

[0061] S32, update the permafrost knowledge graph based on user feedback.

[0062] This includes, but is not limited to, adding new knowledge, updating knowledge relationships, and deleting outdated knowledge for the permafrost knowledge graph.

[0063] Please continue to refer to the figure. After the permafrost knowledge graph is updated, the permafrost domain model operation method also includes: S33, as detailed below.

[0064] S33 redefines features at specific levels in the semantic model by combining parameter fine-tuning and adaptation techniques with domain knowledge from the updated permafrost knowledge graph.

[0065] Among them, parameter fine-tuning and adaptation techniques can be, but are not limited to, LoRA, Adapter, and PEFT techniques.

[0066] Semantic models are multi-layered. For example, the base layer primarily handles simple features, including lexical expressions and basic image pixels; these features do not involve complex domain knowledge. The intermediate layer mainly identifies specific concepts, sentence structures, or simple logical relationships. The upper layers of the semantic model perform complex reasoning and synthesis, such as domain-specific rules, connections between concepts, and completing complex reasoning tasks. For instance, in an experimental environment, the base layer of the semantic model primarily extracts basic information from permafrost research (such as permafrost temperature and humidity); the intermediate layer mainly uses a domain corpus to identify domain-specific terms and the interrelationships between concepts (such as the thermodynamic properties of animated images and the physical properties of permafrost); and the upper layer learns the reasoning abilities of domain experts to understand complex causal relationships or influences (such as the need to consider the potential risk of ground subsidence in engineering construction projects in permafrost regions). Therefore, the redefinition of features at a specific level in a model, as described in S33, means: at the base layer, improvements are directly achieved through pre-trained word vectors or simple knowledge graphs; at the intermediate and higher levels, adjustments are made based on feedback from domain experts and new research findings to better adapt to specific tasks in the permafrost field.

[0067] By redefining features at specific levels in the semantic model, domain knowledge is effectively integrated with model capabilities, enhancing the model's sensitivity to specific tasks in the permafrost domain. This is equivalent to pre-training and fine-tuning.

[0068] Parameter fine-tuning and adaptation techniques enable domain transfer and knowledge adaptation, reduce reliance on high-quality data, effectively solve the problems of data closure and scarcity in the scientific research field, and enable the model to capture and analyze cutting-edge scientific research knowledge in a timely manner.

[0069] The semantic model can be, but is not limited to, large models such as Deepseek, OpenAI GPT, and Meta LLaMA.

[0070] Building upon the preceding text, this invention also provides an optional implementation method for generating an initial permafrost knowledge graph, or for dynamically updating the permafrost knowledge graph. Please refer to [link / reference needed]. Figure 4 The model operation methods for the permafrost field also include: S11, S12, S13, S14 and S15, which are described in detail below.

[0071] S11, acquire data in the permafrost region.

[0072] Optionally, data can be downloaded from public databases, extracted from internal enterprise systems, or collected through web crawlers. Public databases include the China Frozen Soil Engineering Database and the China Coal Industry Database. The web crawler uses the Scrapy framework and supports scheduled crawling and data supplementation.

[0073] S12, Preprocessing data in the permafrost field.

[0074] S13 involves deep processing of the preprocessed data to build a domain corpus.

[0075] The domain corpus includes text features and image features from the preprocessed data.

[0076] Optionally, the BERT model is used for text feature extraction to extract semantic information from the text. Simultaneously, a recurrent neural network (RNN) is used to extract image features, which are then combined with the text features to construct a domain-specific corpus.

[0077] S14. Based on the domain corpus, compile rules specific to the permafrost domain to generate a rule template library.

[0078] The rule template library includes physical law rules and engineering experience rules for the field of frozen soil.

[0079] S15 uses a knowledge graph construction tool to process the domain corpus and rule template library to construct the Frozen Soil Knowledge Graph, which is used to store the structured representation of domain knowledge.

[0080] It integrates various forms of information such as text, images, and structural data, improving the expressiveness and interpretability of scientific research tasks, and realizing closed-loop support for the scientific research process from "reading literature" to "producing results".

[0081] Based on the foregoing, regarding the content of S12, this embodiment of the invention also provides an optional implementation method, please refer to the following. S12, preprocessing data in the permafrost field, includes: S121, S122, and S123, which are specifically described below.

[0082] S121, Cleaning and processing of data in the field of permafrost.

[0083] The data cleaning process includes deduplication, missing data filling, anomaly repair (unrepairable anomalies can be deleted), format standardization, and data unification. This process removes invalid, abnormal, and inaccurate data. Algorithms such as decision trees and K-Means can be used, but are not limited to, to complete the cleaning process.

[0084] In one alternative implementation, when the acquired permafrost data includes unstructured data, the unstructured data can be converted into structured data first using technologies such as OCR and content recognition, and then cleaned.

[0085] S122, Check the permafrost data after cleaning according to preset rules, confirm whether the permafrost data is qualified, delete the unqualified permafrost data or re-clean it.

[0086] S123 After all the data in the permafrost field are qualified, the data in the permafrost field are classified according to the research area, research theme and source type of the original permafrost field data. The classified permafrost field data is the preprocessed permafrost field data.

[0087] The types of sources include, but are not limited to, online data, news information, journal articles, and scientific reports.

[0088] Please see Figure 5 , Figure 5 The present invention provides a permafrost field model running device, which is optionally applied to the electronic device described above.

[0089] The permafrost field model operation device includes: a first processing unit 501 and a second processing unit 502.

[0090] The first processing unit 501 is used to decompose the research needs in the field of permafrost input by the user according to the current permafrost knowledge graph, and construct a task logic graph based on the decomposition results; wherein, the decomposition results include task type, relevant input information of each sub-task, task output rules, and connection conditions between each sub-task.

[0091] The first processing unit 501 is also used to generate a standardized template for task execution based on the task logic diagram and the domain knowledge in the current permafrost knowledge graph. The standardized template includes the standard processing flow and exception handling flow for task execution.

[0092] The second processing unit 502 is used to input the standardized template and the current permafrost knowledge graph into the pre-trained semantic model to obtain the output results corresponding to the scientific research needs in the field of permafrost.

[0093] Optionally, the second processing unit 502 may execute S23 and S24 as described above, and the first processing unit 501 may execute other steps in the above method embodiments.

[0094] It should be noted that the permafrost model running device provided in this embodiment can execute the method flow shown in the above-described method flow embodiment to achieve the corresponding technical effects. For the sake of brevity, any parts not mentioned in this embodiment can be referred to the corresponding content in the above-described embodiments.

[0095] This invention also provides a storage medium storing computer instructions and programs, which, when read and run, execute the permafrost field model running method described above. The storage medium may include memory, flash memory, registers, or a combination thereof.

[0096] The following provides an electronic device, which may be a computer device or a server device, such as... Figure 1 As shown, the above-described method for running a permafrost model can be implemented. Specifically, the electronic device includes a processor 10, a memory 11, and a bus 12. The processor 10 may be a CPU. The memory 11 is used to store one or more programs, which, when executed by the processor 10, execute the permafrost model running method of the above embodiment.

[0097] In summary, the permafrost domain model operation method, apparatus, storage medium, and device provided by this invention decompose user-inputted permafrost research needs based on the current permafrost knowledge graph, and construct a task logic graph based on the decomposition results. The decomposition results include task types, relevant input information for each sub-task, task output rules, and connection conditions between sub-tasks. Based on the task logic graph and domain knowledge in the current permafrost knowledge graph, a standardized template for task execution is generated, including a standard processing flow and an exception handling flow for task execution. The standardized template and the current permafrost knowledge graph are input into a pre-trained semantic model to obtain the output results corresponding to the permafrost research needs. By introducing standardized templates and a permafrost knowledge graph, a professional domain semantic model is constructed, improving the model's ability to understand professional terminology, complex concepts, and their logical relationships, overcoming the limitations of existing general-purpose large models in semantic understanding and terminology mastery.

[0098] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0099] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A method for running a model in the permafrost field, characterized in that, The method includes: Based on the current permafrost knowledge graph, the research needs in the field of permafrost input by the user are decomposed, and a task logic graph is constructed based on the decomposition results. The decomposition results include the task type, the relevant input information of each sub-task, the task output rules, and the connection conditions between each sub-task. Based on the task logic diagram and the domain knowledge in the current permafrost knowledge graph, a standardized template for task execution is generated, wherein the standardized template includes a standard processing flow and an exception handling flow for task execution. The standardized template and the current permafrost knowledge graph are input into a pre-trained semantic model to obtain the output results corresponding to the scientific research needs in the field of permafrost.

2. The method for running a model in the permafrost region as described in claim 1, characterized in that, The step of inputting the standardized template and the current permafrost knowledge graph into a pre-trained semantic model to obtain the output results corresponding to the research needs in the permafrost field includes: Based on the current permafrost knowledge graph, a heuristic algorithm is used to calculate the reasoning results of each subtask; Based on the reasoning algorithm in the current permafrost knowledge graph, the reasoning results of each subtask are combined to obtain the output result.

3. The method for running a model in the permafrost region as described in claim 1, characterized in that, The method further includes: Obtain user feedback information relative to the output results; The permafrost knowledge graph is updated based on the user feedback information.

4. The method for running a model in the permafrost region as described in claim 1, characterized in that, The method further includes: Generate a visualization report of the reasoning process corresponding to the output results.

5. The method for running a model in the permafrost region as described in claim 1, characterized in that, After the permafrost knowledge graph is updated, the method further includes: By combining parameter fine-tuning and adaptation techniques with domain knowledge from the updated permafrost knowledge graph, features at specific levels in the semantic model are redefined.

6. The method for running a model in the permafrost region as described in claim 5, characterized in that, The method further includes: Acquire data in the permafrost region; The data from the permafrost region are preprocessed. The preprocessed data is subjected to deep processing to construct a domain corpus, wherein the domain corpus includes text features and image features from the preprocessed data; Based on the domain corpus, specific rules for the frozen soil domain are compiled to generate a rule template library, wherein the rule template library includes physical law rules and engineering experience rules for the frozen soil domain. A knowledge graph construction tool is used to process the domain corpus and the rule template library to construct a permafrost knowledge graph.

7. The method for running a model in the permafrost region as described in claim 6, characterized in that, The preprocessing of the data in the permafrost region includes: Data in the field of permafrost is cleaned and processed. The data on the permafrost region after cleaning is checked according to the preset rules to confirm whether the data is qualified. Unqualified data is deleted or cleaned again. After all the data in the permafrost field are qualified, they are classified according to the research area, research theme and source type of the original permafrost field data. The classified permafrost field data is the preprocessed permafrost field data.

8. A model operation device for permafrost, characterized in that, The device includes: The first processing unit is used to decompose the research needs in the field of permafrost input by the user based on the current permafrost knowledge graph, and construct a task logic graph based on the decomposition results; wherein, the decomposition results include task type, relevant input information of each sub-task, task output rules, and connection conditions between each sub-task. The first processing unit is further configured to generate a standardized template for task execution based on the task logic diagram and the domain knowledge in the current permafrost knowledge graph, wherein the standardized template includes a standard processing flow and an exception handling flow for task execution. The second processing unit is used to input the standardized template and the current permafrost knowledge graph into a pre-trained semantic model to obtain the output results corresponding to the scientific research needs in the field of permafrost.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the method as described in any one of claims 1-7.

10. An electronic device, characterized in that, include: Processor and memory, the memory being used to store one or more programs; When the one or more programs are executed by the processor, the method as described in any one of claims 1-7 is implemented.

Images