Scientific task processing method and device, computer equipment and readable storage medium

Abstract

The invention relates to a scientific task processing method and device, computer equipment and a readable storage medium. The method comprises the following steps: establishing a scientific tool knowledge graph; obtaining a to-be-processed scientific task; based on the to-be-processed scientific task, traversing the scientific tool knowledge graph, and determining a scientific tool chain adapted to the to-be-processed scientific task; the scientific tool chain comprises a calling sequence of at least one optimal scientific tool; and based on the scientific tool chain, executing the to-be-processed scientific task to obtain an execution result of the to-be-processed scientific task. By adopting the method, the scientific tool with enhanced retrieval can be selected and called, the processing efficiency and the global accuracy of the scientific task are improved, and the resource consumption is reduced.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a scientific task processing method, apparatus, computer equipment, and readable storage medium. Background Art

[0002] Scientific research in the fields of biology, chemistry and materials sciences increasingly relies on a diverse range of specialized tools that cover a wide range of scientific tasks, from molecular modeling to gene editing.

[0003] However, due to the complexity of scientific tools and the dependencies between tools, researchers are unable to fully plan scientific tasks globally and make full use of these resources to solve corresponding scientific tasks. This makes it easy for problems such as high consumption of tool resources and low efficiency to occur during the collaborative use of multiple scientific tools. Summary of the invention

[0004] Based on this, it is necessary to provide a scientific task processing method, device, computer equipment and readable storage medium that can integrate scientific tools, better organize and schedule scientific tools to solve scientific tasks in response to the above-mentioned technical problems.

[0005] In a first aspect, the present application provides a scientific task processing method, the method comprising:

[0006] Build a knowledge graph of scientific tools;

[0007] Get pending science tasks;

[0008] Based on the scientific task to be processed, traverse the scientific tool knowledge graph to determine a scientific tool chain suitable for the scientific task to be processed; the scientific tool chain includes at least one calling sequence of an optimal scientific tool;

[0009] Based on the scientific tool chain, the scientific task to be processed is executed to obtain the execution result of the scientific task to be processed.

[0010] In one embodiment, traversing the scientific tool knowledge graph based on the scientific task to be processed to determine the scientific tool chain adapted for the scientific task to be processed includes:

[0011] According to the scientific task to be processed, perform full-graph retrieval and sub-graph retrieval on the scientific tool knowledge graph to obtain at least one suitable scientific tool for the scientific task to be processed;

[0012] Combining and sorting the adapted scientific tools to obtain at least one scientific tool combination;

[0013] Obtaining the combined semantic similarity between each of the scientific tool combinations and the scientific task to be processed, and extracting at least one of the optimal scientific tools from the adapted scientific tools;

[0014] Based on the optimal scientific tool, the scientific tool chain adapted to the scientific task to be processed is generated.

[0015] In one embodiment, the scientific tool knowledge graph includes a plurality of associated tool nodes, each tool node represents a scientific tool; performing a full-graph search and a sub-graph search on the scientific tool knowledge graph for the scientific task to be processed to obtain at least one suitable scientific tool for the scientific task to be processed includes:

[0016] According to the scientific task to be processed, a full-graph search is performed on the scientific tool knowledge graph, and a first tool node is extracted from the scientific tool knowledge graph; a first semantic similarity between the first tool node and the scientific task to be processed is greater than a first set threshold;

[0017] According to the first tool node and the preset subgraph search depth, the subgraph corresponding to each of the first tool nodes is determined from the scientific tool knowledge graph, and a second tool node is extracted from each of the subgraphs; the second semantic similarity between the second tool node and the corresponding first tool node after being spliced ​​and the scientific task to be processed is greater than a second set threshold;

[0018] The adapted scientific tool for the scientific task to be processed is determined according to the first tool node and the second tool node.

[0019] In one embodiment, obtaining the semantic similarity between each combination of the scientific tools and the scientific task to be processed, and extracting at least one optimal scientific tool from the adapted scientific tools comprises:

[0020] Obtaining a first semantic similarity between each of the first tool nodes and the scientific task to be processed;

[0021] Obtaining a third semantic similarity between each of the second tool nodes and the scientific task to be processed;

[0022] Calculating the product of the first semantic similarity and each of the third semantic similarities in each subgraph respectively, to obtain a plurality of the combined semantic similarities in each subgraph;

[0023] Sorting the combined semantic similarities of all subgraphs to obtain a sorting result;

[0024] According to the ranking result, at least one optimal scientific tool is extracted from the adapted scientific tools.

[0025] In one embodiment, generating the scientific tool chain adapted for the scientific task to be processed based on the optimal scientific tool includes:

[0026] Extracting key information of the scientific task to be processed, wherein the key information includes a plurality of semantic parameters;

[0027] The optimal scientific tools are prioritized according to the semantic parameters to generate the scientific tool chain adapted to the scientific task to be processed.

[0028] In one embodiment, executing the scientific task to be processed based on the scientific tool chain to obtain the execution result of the scientific task to be processed includes:

[0029] According to the scientific tool chain, each of the semantic parameters is input into each of the optimal scientific tools in turn, and the execution result of the scientific task to be processed is output.

[0030] In one embodiment, the method further comprises:

[0031] If the execution result is different from the preset result, a scientific tool chain adapted to the scientific task to be processed is regenerated according to the semantic parameters and the optimal scientific tool until the execution result is the same as the preset result.

[0032] In a second aspect, the present application also provides a scientific task processing device, comprising:

[0033] Graph building module, used to build scientific tool knowledge graph;

[0034] Task acquisition module, used to obtain scientific tasks to be processed;

[0035] A task planning module, used to traverse the scientific tool knowledge graph based on the scientific task to be processed, and determine a scientific tool chain suitable for the scientific task to be processed; the scientific tool chain includes at least one calling sequence of an optimal scientific tool;

[0036] The task execution module is used to execute the scientific task to be processed based on the scientific tool chain and obtain the execution result of the scientific task to be processed.

[0037] In a third aspect, the present application further provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps corresponding to the method described in the first aspect above are implemented.

[0038] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps corresponding to the method described in the first aspect above are implemented.

[0039] The above-mentioned scientific task processing method, apparatus, computer equipment and readable storage medium construct a scientific tool knowledge graph; obtain the scientific task to be processed; based on the scientific task to be processed, traverse the scientific tool knowledge graph to determine the scientific tool chain adapted for the scientific task to be processed; the scientific tool chain includes at least one calling sequence of the optimal scientific tool; based on the scientific tool chain, execute the scientific task to be processed and obtain the execution result of the scientific task to be processed, thereby realizing retrieval-enhanced scientific tool selection and calling, improving the processing efficiency and global accuracy of scientific tasks, and reducing resource consumption. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0041] Figure 1 An application environment diagram of a scientific task processing method in one embodiment;

[0042] Figure 2 A schematic diagram of a flow chart of a scientific task processing method in one embodiment;

[0043] Figure 3 is a flow chart of step 203 in one embodiment;

[0044] Figure 4 is a flow chart of step 301 in one embodiment;

[0045] Figure 5 is a flow chart of step 303 in one embodiment;

[0046] Figure 6 is a flow chart of step 304 in one embodiment;

[0047] Figure 7 is a structural block diagram of a scientific task processing device in one embodiment;

[0048] Figure 8 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0050] Unless otherwise defined, the technical terms or scientific terms involved in this application should be understood by people with ordinary skills in the technical field to which this application belongs. The words "one", "a", "a", "the" and the like involved in this application do not indicate a quantitative limitation, and may represent the singular or plural. The terms "include", "comprise", "have" and any of their variations involved in this application are intended to cover non-exclusive inclusions; for example, a process, method, system, product or device that includes a series of steps or modules (units) is not limited to the listed steps or units, but may also include steps or units that are not listed, or may also include other steps or units inherent to these processes, methods, products or devices. The words "connect", "connected", "coupled" and the like involved in this application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The "multiple" involved in this application refers to two or more. "And / or" describes the association relationship of associated objects, indicating that there may be three relationships, for example, "A and / or B" can represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the objects before and after are in an "or" relationship. The terms "first", "second", "third", etc. involved in this application are only used to distinguish similar objects and do not represent a specific ordering of the objects.

[0051] The scientific task processing method provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown, the terminal 102 communicates with the server 104 through a network. The data storage system can store data that the server 104 needs to process. The data storage system can be integrated on the server 104, or placed on the cloud or other network servers.

[0052] On server 104, a scientific tool knowledge graph is constructed; a scientific task to be processed is obtained; based on the scientific task to be processed, the scientific tool knowledge graph is traversed to determine a scientific tool chain suitable for the scientific task to be processed; the scientific tool chain includes at least one calling sequence of an optimal scientific tool; based on the scientific tool chain, the scientific task to be processed is executed to obtain an execution result of the scientific task to be processed.

[0053] The terminal 102 may be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, etc. The server 104 may be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0054] In an exemplary embodiment, Figure 2 As shown, a scientific task processing method is provided, which is applied to Figure 1 The server side in is taken as an example to illustrate, including the following steps 202 to 206.

[0055] in:

[0056] Step 201, construct a scientific tool knowledge graph.

[0057] The scientific tool knowledge graph includes a plurality of associated tool nodes, each of which represents a scientific tool. The scientific tool knowledge graph also includes attribute nodes connected to each of the tool nodes, and the attribute nodes store information such as the name, function, input, and output of the scientific tool.

[0058] Step 202, obtaining scientific tasks to be processed.

[0059] Step 203, based on the scientific task to be processed, traverse the scientific tool knowledge graph to determine a scientific tool chain suitable for the scientific task to be processed; the scientific tool chain includes at least one optimal scientific tool calling sequence.

[0060] Specifically, according to the scientific task to be processed, the scientific tool knowledge graph is traversed to obtain the semantic similarity between the scientific task to be processed and the tool node of the scientific tool knowledge graph, and according to the semantic similarity, the adapted scientific tool of the scientific task to be processed is extracted from the scientific tool knowledge graph. Then, the optimal scientific tool is selected from the adapted scientific tools to generate the scientific tool chain of the scientific task to be processed.

[0061] Step 204: Based on the scientific tool chain, execute the scientific task to be processed to obtain the execution result of the scientific task to be processed.

[0062] Specifically, the large language model is used to extract the input parameters of each optimal scientific tool from the context of the scientific task to be processed, and the input is formatted to meet the requirements of the specific tool. According to the scientific tool chain, the corresponding optimal scientific tools are called in sequence to execute the scientific task to be processed. The results output by each optimal scientific tool are integrated and analyzed to obtain the final execution result of the scientific task to be processed.

[0063] Optionally, if the execution result is different from the preset result, a scientific tool chain adapted to the scientific task to be processed is regenerated according to the semantic parameters and the optimal scientific tool until the execution result is the same as the preset result.

[0064] In the above-mentioned scientific task processing method, by constructing a scientific tool knowledge graph, the scientific task to be processed is allowed to traverse the scientific tool knowledge graph, and the scientific tool chain suitable for the scientific task to be processed is determined. Based on the scientific tool chain, the scientific task to be processed is executed to obtain the execution result of the scientific task to be processed, thereby realizing retrieval-enhanced scientific tool selection and calling, improving the processing efficiency and global accuracy of scientific tasks, and reducing resource consumption.

[0065] In an exemplary embodiment, Figure 3 As shown, step 203 traverses the scientific tool knowledge graph based on the scientific task to be processed to determine the scientific tool chain adapted to the scientific task to be processed, specifically including the following steps 301 to 304. Among them:

[0066] Step 301: perform a full-graph search and a sub-graph search on the scientific tool knowledge graph according to the scientific task to be processed, and obtain at least one suitable scientific tool for the scientific task to be processed.

[0067] Step 302: sort the adapted scientific tools in combination to obtain at least one scientific tool combination.

[0068] The adapted scientific tools include the first scientific tool retrieved from the full graph and the second scientific tool retrieved from the sub-graph. The first scientific tool and the second scientific tool are combined and sorted to obtain at least one scientific tool combination.

[0069] Step 303, obtaining the combined semantic similarity between each of the scientific tool combinations and the scientific task to be processed, and extracting at least one of the optimal scientific tools from the adapted scientific tools.

[0070] Specifically, the combined semantic similarity between each of the scientific tool combinations and the scientific task to be processed is calculated, and each of the scientific tool combinations is sorted from high to low according to the combined semantic similarity, and the n scientific tool combinations with the highest combined semantic similarity are selected, and the first scientific tool and the second scientific tool contained in the n scientific tool combinations are taken as the optimal scientific tools.

[0071] Step 304: Generate the scientific tool chain adapted to the scientific task to be processed based on the optimal scientific tool.

[0072] In this embodiment, by combining the first scientific tool with the second scientific tool, the optimal scientific tool is selected from the adapted scientific tools based on the combined semantic similarity between the scientific tool combination and the scientific task to be processed. This can make the extracted scientific tool and the scientific task strongly correlated, improve the accuracy of scientific tool extraction, and thus improve the execution efficiency of scientific tasks.

[0073] In one embodiment, if Figure 4 As shown, step 301 performs full-graph retrieval and sub-graph retrieval on the scientific tool knowledge graph for the scientific task to be processed to obtain at least one adapted scientific tool for the scientific task to be processed, specifically including the following steps 401 to 403.

[0074] Step 401: perform a full-graph search on the scientific tool knowledge graph according to the scientific task to be processed, and extract a first tool node from the scientific tool knowledge graph.

[0075] Among them, a first semantic similarity between the first tool node and the scientific task to be processed is greater than a first set threshold.

[0076] Specifically, according to the scientific task to be processed, the scientific tool knowledge graph is searched in its entirety, the first semantic similarity between the scientific task to be processed and each tool node in the scientific tool knowledge graph is calculated, and the first tool node whose first semantic similarity is greater than a first set threshold is extracted. The first tool node is the first scientific tool obtained by the full-graph search.

[0077] Step 402, according to the first tool nodes and a preset subgraph search depth, determine the subgraphs corresponding to each of the first tool nodes from the scientific tool knowledge graph, and extract the second tool nodes from each of the subgraphs.

[0078] Among them, the second semantic similarity between the second tool node and the corresponding first tool node after splicing and the scientific task to be processed is greater than a second set threshold.

[0079] In detail, taking each first tool node as the origin, according to the preset subgraph search depth, determine the subgraph corresponding to each first tool node in the scientific tool knowledge graph. Obtain all tool nodes from each subgraph as the preparatory tool nodes of the second tool node. After splicing and combining each preparatory tool node with the corresponding first tool node, calculate the second semantic similarity with the scientific task to be processed, and take the preparatory tool node whose second semantic similarity is greater than the second set threshold as the second tool node. The second tool node is the second scientific tool obtained by the subgraph search.

[0080] Step 403: Determine the adapted scientific tool for the scientific task to be processed according to the first tool node and the second tool node.

[0081] Specifically, each first scientific tool corresponding to each first tool node and each second scientific tool corresponding to each second tool node are combined to obtain an adapted scientific tool for the scientific task to be processed.

[0082] In this embodiment, by extracting the first tool node whose first semantic similarity is greater than the first set threshold in the global search, and extracting the preliminary tool node whose second semantic similarity is greater than the second set threshold in the subgraph search as the second tool node, the screening of adapted scientific tools is completed, and the relevance of the adapted scientific tools to the tasks to be processed is improved, which is conducive to the selection of the optimal scientific tool and further improves the accuracy of the scientific tool chain in processing scientific tasks.

[0083] In one embodiment, if Figure 5 As shown, step 303 obtains the combined semantic similarity between each of the scientific tool combinations and the scientific task to be processed, and extracts at least one of the optimal scientific tools from the adapted scientific tools, specifically including the following steps 501 to 505.

[0084] Step 501: Obtain a first semantic similarity between each of the first tool nodes and the scientific task to be processed.

[0085] Step 502: Obtain a third semantic similarity between each of the second tool nodes and the scientific task to be processed.

[0086] Step 503 : calculating the product of the first semantic similarity and each of the third semantic similarities in each subgraph respectively, to obtain a plurality of the combined semantic similarities in each subgraph.

[0087] Step 504: sort the combined semantic similarities of all subgraphs to obtain a sorting result.

[0088] Step 505: extract at least one optimal scientific tool from the adapted scientific tools according to the ranking result.

[0089] In detail, after full-graph retrieval and sub-graph retrieval, a series of adapted scientific tools for the scientific task to be processed are obtained. The first semantic similarity between the first scientific tool in the adapted scientific tools and the scientific task to be processed, and the third semantic similarity between the second scientific tool and the scientific task to be processed are calculated respectively. In each sub-graph, the product of the first semantic similarity and each third semantic similarity is calculated respectively to obtain multiple combined semantic similarities in each sub-graph, that is, the combined semantic similarity of each scientific tool combination in each sub-graph. The combined semantic similarities of all sub-graphs are sorted from high to low to obtain a sorting result. Select the n scientific tool combinations with the highest combined semantic similarity, and take the first scientific tool and the second scientific tool contained in the n scientific tool combinations as the optimal scientific tools.

[0090] In this embodiment, the combined semantic similarity of the scientific tool combination is obtained by multiplying the first semantic similarity of the first scientific tool with the third semantic similarity of the second tool, and the first scientific tool and the second scientific tool contained in the n scientific tool combinations with the highest combined semantic similarity are taken as the optimal scientific tools, thereby realizing the extraction of the optimal scientific tools and improving the integrity of the scientific tool chain and its relevance to the scientific tasks to be processed.

[0091] In an exemplary embodiment, Figure 6 As shown, step 304 generates the scientific tool chain adapted to the scientific task to be processed based on the optimal scientific tool, which specifically includes the following steps 601 to 602.

[0092] Step 601: extract key information of the scientific task to be processed, where the key information includes multiple semantic parameters.

[0093] Step 602: Prioritize each of the optimal scientific tools according to the semantic parameters, and generate the scientific tool chain adapted to the scientific task to be processed.

[0094] In detail, the key information of the scientific task to be processed is extracted using a large language model, and the key information includes multiple semantic parameters. The optimal scientific tools are prioritized according to the semantic parameters to generate the scientific tool chain adapted to the scientific task to be processed. Furthermore, according to the scientific tool chain, each of the semantic parameters is used as an input parameter of the optimal scientific tool, and is input into each of the optimal scientific tools in turn. The results output by each optimal scientific tool are integrated and analyzed, and the execution results of the scientific task to be processed are output.

[0095] In this embodiment, key information of the scientific tasks to be processed is extracted through a large language model, so as to sort the optimal scientific tools and generate a scientific tool chain to achieve the optimal processing order of the scientific tasks to be processed and improve the accuracy of the execution results.

[0096] In a preferred embodiment, a scientific task processing method is provided, which specifically includes the following contents:

[0097] S1, build a knowledge graph of scientific tools.

[0098] S2, obtain pending scientific tasks.

[0099] S3, according to the scientific task to be processed, perform a full-graph search on the scientific tool knowledge graph, calculate the first semantic similarity between the scientific task to be processed and each tool node in the scientific tool knowledge graph, and extract the first tool node whose first semantic similarity is greater than a first set threshold. The first tool node is the first scientific tool obtained by the full-graph search.

[0100]

[0101] Among them, T full represents the k first scientific tools retrieved from the whole graph, S(q, T i ) represents the pending scientific task q and the tool node T i The first semantic similarity of , G represents the scientific tool knowledge graph, and top-k{} represents the k scientific tools with the highest first semantic similarity.

[0102] S4, taking each first tool node as the origin, according to the preset subgraph search depth, determine the subgraph corresponding to each first tool node in the scientific tool knowledge graph, and perform subgraph search on the scientific tool knowledge graph. Obtain all tool nodes from each subgraph as preparatory tool nodes for the second tool node. After splicing and combining each preparatory tool node with the corresponding first tool node, calculate the second semantic similarity with the scientific task to be processed, and take the preparatory tool node whose second semantic similarity is greater than the second set threshold as the second tool node.

[0103]

[0104] Among them, T sub represents the m second science tools retrieved from the subgraph, Represents the pending scientific task q and the current first scientific tool T i and Preparatory Science Tools T j The second semantic similarity after combination, Indicates the first scientific tool T i The subgraph of

[0105] top-m{} means taking the m scientific tools with the second highest semantic similarity.

[0106] S5, combining the first scientific tools corresponding to the first tool nodes and the second scientific tools corresponding to the second tool nodes to obtain an adapted scientific tool for the scientific task to be processed. Combining and sorting the first scientific tools and the second scientific tools in each subgraph to obtain at least one scientific tool combination.

[0107] S6, calculating a first semantic similarity between the first scientific tool and the scientific task to be processed, and a third semantic similarity between the second scientific tool and the scientific task to be processed.

[0108] S7, calculating the product of the first semantic similarity and the third semantic similarity in each scientific tool combination, and the combined semantic similarity of each scientific tool combination.

[0109] S8, sorting all combinations of semantic similarity from high to low to obtain a sorting result, and selecting n scientific tool combinations with the highest combination semantic similarity.

[0110]

[0111] Among them, T comb represents the n scientific tool combinations with the highest semantic similarity extracted, S(q, T i )×S(q′,T j ) is the combination semantic similarity. top-n{} means taking the n scientific tool combinations with the highest combination semantic similarity.

[0112] S9, taking the first scientific tool and the second scientific tool included in the n scientific tool combinations as optimal scientific tools, extracting multiple semantic parameters of the scientific task to be processed using the large language model, prioritizing each optimal scientific tool according to the semantic parameters, and generating the scientific tool chain adapted to the scientific task to be processed.

[0113] T chain ={T1→T2→…→T a}∈T comb ;

[0114] T chain The generated scientific tool chain includes a tools selected in sequence from a tool set of n scientific tool combinations, and the arrows indicate the execution order.

[0115] S10, according to the scientific tool chain, the corresponding optimal scientific tools are called in sequence to execute the scientific tasks to be processed. The results output by each optimal scientific tool are integrated and analyzed to obtain the final execution results of the scientific tasks to be processed.

[0116] If the execution result is different from the preset result, the scientific tool chain adapted to the scientific task to be processed is regenerated using the large language model according to the semantic parameters and the optimal scientific tools until the execution result is the same as the preset result.

[0117] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0118] Based on the same inventive concept, the embodiment of the present application also provides a scientific task processing device for implementing the above-mentioned scientific task processing method. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above-mentioned method, so the specific limitations in one or more scientific task processing device embodiments provided below can refer to the limitations of the scientific task processing method above, and will not be repeated here.

[0119] In an exemplary embodiment, Figure 7 As shown, a scientific task processing device is provided, including: a graph construction module 701, a task acquisition module 702, a task planning module 703, and a task execution module 704, wherein:

[0120] The graph construction module 701 is used to construct the scientific tool knowledge graph.

[0121] The task acquisition module 702 is used to acquire scientific tasks to be processed.

[0122] The task planning module 703 is used to traverse the scientific tool knowledge graph based on the scientific task to be processed, and determine the scientific tool chain suitable for the scientific task to be processed; the scientific tool chain includes at least one calling sequence of the optimal scientific tool.

[0123] The task execution module 704 is used to execute the scientific task to be processed based on the scientific tool chain and obtain the execution result of the scientific task to be processed.

[0124] In one embodiment, the task planning module 703 is also used to: perform full-graph retrieval and sub-graph retrieval on the scientific tool knowledge graph according to the scientific task to be processed, and obtain at least one adapted scientific tool for the scientific task to be processed; perform combination sorting on the adapted scientific tools to obtain at least one scientific tool combination; obtain the combined semantic similarity between each of the scientific tool combinations and the scientific task to be processed, and extract at least one of the optimal scientific tools from the adapted scientific tools; and generate the scientific tool chain adapted to the scientific task to be processed based on the optimal scientific tool.

[0125] In one embodiment, the task planning module 703 is also used to: perform a full-graph search on the scientific tool knowledge graph according to the scientific task to be processed, and extract a first tool node from the scientific tool knowledge graph; the first semantic similarity between the first tool node and the scientific task to be processed is greater than a first set threshold; determine the subgraph corresponding to each of the first tool nodes from the scientific tool knowledge graph according to the first tool node and a preset subgraph retrieval depth, and extract a second tool node from each of the subgraphs; the second semantic similarity between the second tool node and the corresponding first tool node after splicing and the scientific task to be processed is greater than a second set threshold; determine the adapted scientific tool for the scientific task to be processed according to the first tool node and the second tool node.

[0126] In one embodiment, the task planning module 703 is also used to: obtain the first semantic similarity between each of the first tool nodes and the scientific task to be processed; obtain the third semantic similarity between each of the second tool nodes and the scientific task to be processed; calculate the product between the first semantic similarity and each of the third semantic similarities in each subgraph respectively, to obtain multiple combined semantic similarities in each subgraph; sort the combined semantic similarities of all subgraphs to obtain a sorting result; and extract at least one of the optimal scientific tools from the adapted scientific tools according to the sorting result.

[0127] In one embodiment, the task planning module 703 is also used to: extract key information of the scientific task to be processed, wherein the key information includes multiple semantic parameters; prioritize each of the optimal scientific tools according to the semantic parameters, and generate the scientific tool chain adapted to the scientific task to be processed.

[0128] In one embodiment, the task execution module 704 is also used to: input each of the semantic parameters into each of the optimal scientific tools in turn according to the scientific tool chain, and output the execution result of the scientific task to be processed.

[0129] In one embodiment, the device further includes a result summary module, which is used to: integrate and analyze the results output by each optimal scientific tool to obtain the final execution result of the scientific task to be processed. If the execution result is different from the preset result, the scientific tool chain adapted to the scientific task to be processed is regenerated using the large language model according to the semantic parameters and the optimal scientific tool until the execution result is the same as the preset result.

[0130] Each module in the above scientific task processing device can be implemented in whole or in part by software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each module.

[0131] In an exemplary embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. Figure 8 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store scientific tool data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a scientific task processing method is implemented.

[0132] Those skilled in the art will understand that Figure 8 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0133] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0134] Build a knowledge graph of scientific tools;

[0135] Get pending science tasks;

[0136] Based on the scientific task to be processed, traverse the scientific tool knowledge graph to determine a scientific tool chain suitable for the scientific task to be processed; the scientific tool chain includes at least one calling sequence of an optimal scientific tool;

[0137] Based on the scientific tool chain, the scientific task to be processed is executed to obtain the execution result of the scientific task to be processed.

[0138] In one embodiment, when the processor executes the computer program, the following steps are also implemented: according to the scientific task to be processed, the scientific tool knowledge graph is searched in the whole graph and in the sub-graph to obtain at least one adapted scientific tool for the scientific task to be processed; the adapted scientific tools are combined and sorted to obtain at least one scientific tool combination; the combined semantic similarity between each scientific tool combination and the scientific task to be processed is obtained, and at least one optimal scientific tool is extracted from the adapted scientific tools; based on the optimal scientific tool, the scientific tool chain adapted to the scientific task to be processed is generated.

[0139] In one embodiment, when the processor executes the computer program, the following steps are also implemented: according to the scientific task to be processed, the scientific tool knowledge graph is searched in its entirety, and a first tool node is extracted from the scientific tool knowledge graph; a first semantic similarity between the first tool node and the scientific task to be processed is greater than a first set threshold; according to the first tool node and a preset subgraph retrieval depth, a subgraph corresponding to each of the first tool nodes is determined from the scientific tool knowledge graph, and a second tool node is extracted from each of the subgraphs; a second semantic similarity between the second tool node and the corresponding first tool node after being spliced ​​and the scientific task to be processed is greater than a second set threshold; according to the first tool node and the second tool node, the adapted scientific tool for the scientific task to be processed is determined.

[0140] In one embodiment, when the processor executes the computer program, the following steps are also implemented: obtaining the first semantic similarity between each of the first tool nodes and the scientific task to be processed; obtaining the third semantic similarity between each of the second tool nodes and the scientific task to be processed; calculating the product between the first semantic similarity and each of the third semantic similarities in each subgraph respectively to obtain a plurality of the combined semantic similarities in each subgraph; sorting the combined semantic similarities of all subgraphs to obtain a sorting result; and extracting at least one of the optimal scientific tools from the adapted scientific tools according to the sorting result.

[0141] In one embodiment, when the processor executes the computer program, the following steps are also implemented: extracting key information of the scientific task to be processed, wherein the key information includes multiple semantic parameters; prioritizing each of the optimal scientific tools according to the semantic parameters, and generating the scientific tool chain adapted to the scientific task to be processed.

[0142] In one embodiment, when the processor executes the computer program, the following steps are also implemented: according to the scientific tool chain, each of the semantic parameters is input into each of the optimal scientific tools in turn, and the execution result of the scientific task to be processed is output.

[0143] In one embodiment, when the processor executes the computer program, the following steps are also implemented: if the execution result is different from the preset result, a scientific tool chain adapted to the scientific task to be processed is regenerated according to the semantic parameters and the optimal scientific tool until the execution result is the same as the preset result.

[0144] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps corresponding to the scientific task processing method described in the above embodiments are implemented.

[0145] In one embodiment, a computer program product is provided, including a computer program, which, when executed by a processor, implements the steps corresponding to the scientific task processing method described in the above embodiments.

[0146] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0147] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0148] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0149] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A scientific task processing method, characterized in that: The method comprises: Build a knowledge graph of scientific tools; Get pending science tasks; Based on the scientific task to be processed, traverse the scientific tool knowledge graph to determine a scientific tool chain suitable for the scientific task to be processed; the scientific tool chain includes at least one calling sequence of an optimal scientific tool; Based on the scientific tool chain, the scientific task to be processed is executed to obtain the execution result of the scientific task to be processed.

2. The scientific task processing method according to claim 1, characterized in that: The traversing the scientific tool knowledge graph based on the scientific task to be processed to determine the scientific tool chain adapted for the scientific task to be processed includes: According to the scientific task to be processed, perform full-graph retrieval and sub-graph retrieval on the scientific tool knowledge graph to obtain at least one suitable scientific tool for the scientific task to be processed; Combining and sorting the adapted scientific tools to obtain at least one scientific tool combination; Obtaining the combined semantic similarity between each of the scientific tool combinations and the scientific task to be processed, and extracting at least one of the optimal scientific tools from the adapted scientific tools; Based on the optimal scientific tool, the scientific tool chain adapted to the scientific task to be processed is generated.

3. The scientific task processing method according to claim 2, characterized in that: The scientific tool knowledge graph includes a plurality of associated tool nodes, each tool node represents a scientific tool; performing full-graph retrieval and sub-graph retrieval on the scientific tool knowledge graph according to the scientific task to be processed to obtain at least one adapted scientific tool for the scientific task to be processed includes: According to the scientific task to be processed, a full-graph search is performed on the scientific tool knowledge graph, and a first tool node is extracted from the scientific tool knowledge graph; a first semantic similarity between the first tool node and the scientific task to be processed is greater than a first set threshold; According to the first tool node and the preset subgraph search depth, the subgraph corresponding to each of the first tool nodes is determined from the scientific tool knowledge graph, and a second tool node is extracted from each of the subgraphs; the second semantic similarity between the second tool node and the corresponding first tool node after being spliced ​​and the scientific task to be processed is greater than a second set threshold; The adapted scientific tool for the scientific task to be processed is determined according to the first tool node and the second tool node.

4. The scientific task processing method according to claim 3, characterized in that: The obtaining of the semantic similarity between the combination of each scientific tool and the scientific task to be processed, and extracting at least one optimal scientific tool from the adapted scientific tools comprises: Acquire a first semantic similarity between each of the first tool nodes and the scientific task to be processed; Obtaining a third semantic similarity between each of the second tool nodes and the scientific task to be processed; Calculating the product of the first semantic similarity and each of the third semantic similarities in each subgraph respectively, to obtain a plurality of the combined semantic similarities in each subgraph; Sorting the combined semantic similarities of all subgraphs to obtain a sorting result; According to the ranking result, at least one optimal scientific tool is extracted from the adapted scientific tools.

5. The scientific task processing method according to claim 2, characterized in that: The step of generating the scientific tool chain adapted for the scientific task to be processed based on the optimal scientific tool comprises: Extracting key information of the scientific task to be processed, wherein the key information includes a plurality of semantic parameters; The optimal scientific tools are prioritized according to the semantic parameters to generate the scientific tool chain adapted to the scientific task to be processed.

6. The scientific task processing method according to claim 5, characterized in that: The executing the scientific task to be processed based on the scientific tool chain to obtain the execution result of the scientific task to be processed includes: According to the scientific tool chain, each of the semantic parameters is input into each of the optimal scientific tools in turn, and the execution result of the scientific task to be processed is output.

7. The scientific task processing method according to claim 5, characterized in that: The method further comprises: If the execution result is different from the preset result, a scientific tool chain adapted to the scientific task to be processed is regenerated according to the semantic parameters and the optimal scientific tool until the execution result is the same as the preset result.

8. A scientific task processing device, characterized in that: The device comprises: Graph building module, used to build scientific tool knowledge graph; Task acquisition module, used to obtain scientific tasks to be processed; A task planning module, used to traverse the scientific tool knowledge graph based on the scientific task to be processed, and determine a scientific tool chain suitable for the scientific task to be processed; the scientific tool chain includes at least one calling sequence of an optimal scientific tool; The task execution module is used to execute the scientific task to be processed based on the scientific tool chain and obtain the execution result of the scientific task to be processed.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images