Task processing method and apparatus

Abstract

Provided in the embodiments of the present disclosure are a task processing method and apparatus. The task processing method comprises: acquiring task data of a task to be processed and structured information corresponding to said task, wherein the structured information comprises structure data and a plurality of pieces of knowledge data, the structure data comprises a plurality of nodes and a hierarchical relationship between the nodes, and the nodes represent key points of said task; on the basis of the task data, determining a target node from among the plurality of nodes; inputting into a task generation model node data of the target node and the structure data, so as to generate a guided interaction task, wherein the task generation model is a model based on deep learning; and on the basis of an interaction result obtained by means of executing the guided interaction task, determining target knowledge data from among the plurality of pieces of knowledge data, so as to obtain a task processing result of said task. Structure data is decoupled from knowledge data, a task generation model is guided by the structure data to generate a guided interaction task, and a task processing result is determined taking into account the knowledge data, thereby improving the task processing capability.

Description

Task processing method and device

[0001] This disclosure claims priority to Chinese patent application number 202410015103.9, filed with the Patent Office of China on January 4, 2024, entitled “Task Processing Method and Device,” the entire contents of which are incorporated herein by reference. Technical Field

[0002] The present disclosure relates to the field of artificial intelligence technology, and more particularly to a task processing method and device. Background Art

[0003] With the rapid development of the internet, information has experienced exponential growth and fragmentation. This has made it increasingly difficult for professionals to quickly find the information they need, significantly impacting their research and learning efficiency.

[0004] Currently, common methods for acquiring information are search engines and knowledge question-answering systems. These systems can be based on knowledge bases, frequently used question-answer pairs, or information search. However, these systems primarily rely on semantic matching and keyword matching to obtain answers and provide feedback. However, these methods are not only inefficient but also suffer from low accuracy. Therefore, a more efficient task processing method is urgently needed to address these issues. Summary of the Invention

[0005] In view of this, embodiments of the present disclosure provide a task processing method. One or more embodiments of the present disclosure also relate to a task processing apparatus, a question-and-answer processing method, a question-and-answer processing apparatus, a computing device, a computer-readable storage medium, and a computer program to address technical deficiencies in the prior art.

[0006] According to a first aspect of an embodiment of the present disclosure, there is provided a task processing method, comprising:

[0007] Acquire task data of a task to be processed and structured information corresponding to the task to be processed, wherein the structured information includes structure data and a plurality of knowledge data, the structure data includes a plurality of nodes and hierarchical relationships between the nodes, and the nodes represent key points of the task to be processed;

[0008] determining a target node from the plurality of nodes based on the task data;

[0009] Inputting the node data of the target node and the structure data into a task generation model to generate a guided interaction task, wherein the task generation model is a deep learning-based model;

[0010] Based on the interaction result obtained by executing the guided interaction task, target knowledge data is determined from the plurality of knowledge data to obtain a task processing result of the task to be processed.

[0011] According to a second aspect of an embodiment of the present disclosure, there is provided a task processing device, including:

[0012] an acquisition module configured to acquire task data of a task to be processed and structured information corresponding to the task to be processed, wherein the structured information includes structure data and a plurality of knowledge data, the structure data includes a plurality of nodes and hierarchical relationships between the nodes, and the nodes represent key points of the task to be processed;

[0013] a determination module configured to determine a target node from the plurality of nodes based on the task data;

[0014] a generation module configured to input the node data of the target node and the structure data into a task generation model to generate a guided interaction task, wherein the task generation model is a deep learning-based model;

[0015] The execution module is configured to determine target knowledge data from the plurality of knowledge data based on the interaction result obtained by executing the guided interaction task, and obtain the task processing result of the task to be processed.

[0016] According to a third aspect of an embodiment of the present disclosure, a question-answering processing method is provided, including:

[0017] In response to a problem processing request submitted by a front-end user, obtaining problem data of a problem to be processed and structured information corresponding to the problem to be processed, wherein the structured information includes structure data and a plurality of knowledge data, the structure data includes a plurality of nodes and hierarchical relationships between the nodes, and the nodes represent key points of the problem to be processed;

[0018] determining a target node from the plurality of nodes based on the problem data;

[0019] Inputting the node data and the structural data of the target node into the question-answering task generation model to generate a question-answering guidance task, wherein the question-answering task generation model is a model based on deep learning;

[0020] Based on the interactive results obtained by executing the question-answering guidance task, target knowledge data is determined from the plurality of knowledge data, and a problem processing result of the problem to be processed is obtained;

[0021] The problem handling result is sent to the front-end user as feedback of the problem handling request.

[0022] According to a fourth aspect of an embodiment of the present disclosure, there is provided a question-and-answer processing apparatus, comprising:

[0023] an acquisition module configured to acquire, in response to a problem processing request submitted by a front-end user, problem data of a problem to be processed and structured information corresponding to the problem to be processed, wherein the structured information includes structure data and a plurality of knowledge data, the structure data includes a plurality of nodes and hierarchical relationships between the nodes, and the nodes represent key points of the problem to be processed;

[0024] A first determining module is configured to determine a target node from the plurality of nodes based on the problem data;

[0025] a generation module configured to input the node data of the target node and the structural data into the question-answering task generation model to generate a question-answering guidance task, wherein the question-answering task generation model is a deep learning-based model;

[0026] A second determining module is configured to determine target knowledge data from the plurality of knowledge data based on an interaction result obtained by executing the question-answering guidance task, and obtain a problem processing result for the problem to be processed;

[0027] The sending module is configured to send the problem handling result as feedback of the problem handling request to the front-end user.

[0028] According to a fifth aspect of an embodiment of the present disclosure, there is provided a computing device, including:

[0029] memory and processor;

[0030] The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the above-mentioned task processing method are implemented.

[0031] According to a sixth aspect of an embodiment of the present disclosure, a computer-readable storage medium is provided, which stores computer-executable instructions, and when the instructions are executed by a processor, the steps of the above-mentioned task processing method are implemented.

[0032] According to a seventh aspect of an embodiment of the present disclosure, a computer program is provided, wherein when the computer program is executed in a computer, the computer is caused to execute the steps of the above-mentioned task processing method.

[0033] One embodiment of the present disclosure obtains task data of a pending task and structured information corresponding to the pending task, wherein the structured information includes structured data and multiple knowledge data, the structured data includes multiple nodes and hierarchical relationships between the nodes, and the nodes represent key points of the pending task; based on the task data, a target node is determined from multiple nodes; the node data and structured data of the target node are input into a task generation model to generate a guided interaction task, wherein the task generation model is a deep learning-based model; based on the interaction results obtained by executing the guided interaction task, the target knowledge data is determined from multiple knowledge data to obtain the task processing result of the pending task. The structured data and knowledge data in the structured information are decoupled, the structured data guides the task generation model, the guided interaction task is generated based on the structured data, and then the task processing result of the pending task is determined by combining the interaction results obtained by executing the guided interaction task and the knowledge data, thereby improving the task processing capability while improving the result accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] FIG1 is a schematic diagram of a processing process of a task processing method provided by an embodiment of the present disclosure;

[0035] FIG2 is a flowchart of a task processing method provided by one embodiment of the present disclosure;

[0036] FIG3 is a schematic diagram of structured information of a task processing method provided by one embodiment of the present disclosure;

[0037] FIG4 is a flowchart of a task processing method according to an embodiment of the present disclosure;

[0038] FIG5 is a schematic diagram of a knowledge question-answering structure of a task processing method provided by an embodiment of the present disclosure;

[0039] FIG6 is a schematic structural diagram of a task processing device provided by an embodiment of the present disclosure;

[0040] FIG7 is a flowchart of a question-answering processing method provided by one embodiment of the present disclosure;

[0041] FIG8 is a schematic diagram of the structure of a question-answering processing device provided by one embodiment of the present disclosure;

[0042] FIG9 is a structural block diagram of a computing device provided by an embodiment of the present disclosure. DETAILED DESCRIPTION

[0043] The following description sets forth many specific details to facilitate a full understanding of the present disclosure. However, the present disclosure can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the scope of the present disclosure. Therefore, the present disclosure is not limited to the specific implementations disclosed below.

[0044] The terms used in one or more embodiments of the present disclosure are for the purpose of describing specific embodiments only and are not intended to limit one or more embodiments of the present disclosure. The singular forms "a", "the", and "the" used in one or more embodiments of the present disclosure and the appended claims are also intended to include plural forms, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used in one or more embodiments of the present disclosure refers to and includes any or all possible combinations of one or more associated listed items.

[0045] It should be understood that although the terms first, second, etc. may be used to describe various information in one or more embodiments of the present disclosure, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from each other. For example, without departing from the scope of one or more embodiments of the present disclosure, the first may also be referred to as the second, and similarly, the second may also be referred to as the first. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0046] In addition, 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 one or more embodiments of the present disclosure 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 the relevant laws, regulations and standards of relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0047] In one or more embodiments of the present disclosure, a large model refers to a deep learning model with large-scale model parameters, which usually contains hundreds of millions, tens of billions, hundreds of billions, trillions, or even more than ten trillion model parameters. A large model can also be called a cornerstone model / foundation model (Foundation Model), which is pre-trained by using large-scale unlabeled corpus to produce a pre-trained model with more than 100 million parameters. This model can adapt to a wide range of downstream tasks, and the model has good generalization ability, such as a large-scale language model (LLM), a multi-modal pre-training model, etc.

[0048] In actual applications, large models only require a small number of samples to fine-tune the pre-trained model and can be applied to different tasks. Large models can be widely used in natural language processing (NLP), computer vision and other fields. Specifically, they can be applied to computer vision tasks such as visual question answering (VQA), image caption (IC), and image generation, as well as natural language processing tasks such as text-based sentiment classification, text summary generation, and machine translation. The main application scenarios of large models include digital assistants, intelligent robots, search, online education, office software, e-commerce, and intelligent design.

[0049] First, the terms involved in one or more embodiments of the present disclosure are explained.

[0050] Enterprise knowledge dialogue: A dialogue system based on enterprise knowledge, not limited to enterprise document knowledge.

[0051] DocGraph: A document organization format that can organize documents in a knowledge graph-like format after understanding them.

[0052] An agent is an entity that performs specific tasks within a computer system or network. It can be a software program, a hardware device, or even a person. An agent completes tasks on behalf of users or other systems, such as data collection, data processing, and decision support. Agents can be autonomous and possess a degree of intelligence and adaptability to perform tasks in diverse contexts.

[0053] Chain of Thought (CoT) is a problem-solving and decision-making approach, also known as chain thinking or problem-solving. It emphasizes gradually breaking down a problem, considering its multiple dimensions and causal relationships, and identifying its essence and proposing a reasonable solution.

[0054] Figure 1 is a schematic diagram of the processing process of a task processing method provided by one embodiment of the present disclosure. As shown in Figure 1, after receiving a pending task, the task data and the corresponding DocGraph (i.e., structured information) are obtained. The structured information includes structural data and multiple knowledge data. The structural data includes multiple nodes and the hierarchical relationships between the nodes. The nodes represent the key points of the pending task. The pending task can be a knowledge question-and-answer task or an enterprise knowledge dialogue task. Based on the task data, a target node is determined from multiple nodes according to the node data of each node. The node data and structural data of the target node are input into a task generation model, which generates a guided interaction task. The task generation model can be a deep learning-based model. The guided interaction task is executed, and based on the interaction results obtained from executing the guided interaction task, target knowledge data is determined from multiple knowledge data. The task processing results of the pending task are determined based on the target knowledge data. The task generation model can be a large model with data processing capabilities. During the generation of the guided interaction task by the task generation model and the subsequent execution of the guided interaction task, an agent approach is used to complete environmental perception, planning, and action.

[0055] To summarize, the structural data and knowledge data in structured information are decoupled, the structural data guides the task generation model, and guided interaction tasks are generated based on the structural data. Then, the interaction results obtained by executing the guided interaction tasks and the knowledge data are combined to determine the task processing results of the tasks to be processed, thereby improving the task processing capability and the accuracy of the results.

[0056] In the present disclosure, a task processing method is provided. The present disclosure also relates to a task processing device, a question-and-answer processing method, a question-and-answer processing device, a computing device, and a computer-readable storage medium, which are described in detail one by one in the following embodiments.

[0057] Referring to FIG. 2 , FIG. 2 shows a flow chart of a task processing method provided according to an embodiment of the present disclosure, which specifically includes the following steps.

[0058] Step 202: Obtain task data of the task to be processed and structured information corresponding to the task to be processed, wherein the structured information includes structural data and multiple knowledge data, the structural data includes multiple nodes and hierarchical relationships between the nodes, and the nodes represent key points of the task to be processed.

[0059] Specifically, the pending task can be a knowledge dialogue task, or an information acquisition task such as a question-and-answer task or a human-computer interaction task in which the user obtains information; accordingly, the task data is the question data, dialogue data, and other data submitted by the user, or data triggered and submitted by the user's operation, or scene data in a human-computer interaction scenario; structured data is data containing multiple nodes, and the multiple nodes of the structured data have a hierarchical relationship and are arranged and connected according to the hierarchical structure; knowledge data is the data obtained by integrating the node data corresponding to multiple nodes in the structured data. Structured information can be generated based on documents, historical records, and logs. The key points of the pending task represent the task keywords or task nodes of the pending task.

[0060] Based on this, a task to be processed is received, and task data for the task to be processed, as well as structured information matching the task to be processed, is obtained. The structured information includes structural data and multiple knowledge data. The structural data includes multiple nodes and the hierarchical relationships between the nodes. The nodes represent the key points of the task to be processed. At least two nodes with a hierarchical relationship in the structural data form a node link. The knowledge data is data generated based on the node link, that is, obtained by integrating the node data of the nodes in the node link.

[0061] In practical applications, the structural data and knowledge data contained in structured information are stored separately in a decoupled manner.

[0062] Furthermore, considering that the tasks to be processed are of different types and involve different fields, it is necessary to refer to the type and field of the tasks to be processed when determining the structured information. The specific implementation is as follows:

[0063] Acquire task data of a task to be processed and task information of the task to be processed; and determine structured information corresponding to the task to be processed according to the task information.

[0064] Specifically, the task information includes but is not limited to task type information, field information, and region information of the task to be processed.

[0065] Based on this, the task data of the task to be processed is obtained, the task information of the task to be processed is determined, and the structured information corresponding to the task to be processed is determined according to the task information.

[0066] The task information of the task to be processed is determined based on the structured information of the task to be processed, so that the structured information is highly matched with the task to be processed, thereby improving the accuracy of the determination of the structured information.

[0067] Furthermore, considering that there are many types of tasks to be processed and the amount of structured information is limited, when determining the structured information of the task to be processed, not only can structured information be selected from a pre-stored set of structured information, but also structured information matching the task to be processed can be generated after the task to be processed is determined. The specific implementation is as follows:

[0068] Target data is acquired from a database according to the task information, and structured information is generated based on the target data; or structured information is selected from a pre-stored structured information set according to the task information.

[0069] Specifically, the database can be a knowledge base constructed based on information publicly available on the Internet in multiple fields and self-built information; the target data can be target documents in the form of papers, articles, etc., or data segments or question-and-answer data related to task information.

[0070] Based on this, target data matching the keywords in the task information is obtained from the database according to the task information, and structured information is generated based on the target data. Alternatively, a pre-stored set of structured information is used, and structured information matching the task information is selected from the pre-stored set of structured information according to the task information.

[0071] For example, in the knowledge question-answering scenario, the task to be processed is the knowledge question-answering task, and the task data can be the question data raised by the questioner. When the task data is "how to register a birth", relevant text content or documents can be found in the database based on the task data. After determining the document that matches the task data, structured information with a hierarchical structure as shown in Figure 3 is constructed based on the document. The structured data includes nodes such as birth registration (first level), region A and non-region A (second level), legitimate children and illegitimate children (third level); and knowledge data: birth registration policy for legitimate children in region A; birth registration policy for illegitimate children in region A; birth registration policy for legitimate children in non-region A, and birth registration policy for illegitimate children in non-region A.

[0072] In summary, by determining the structured information corresponding to the task information in at least two ways, the structured information with a high degree of matching with the task information is obtained, thereby improving the accuracy of subsequent task processing results.

[0073] Step 204: Determine a target node from the plurality of nodes based on the task data.

[0074] Specifically, in the above-mentioned acquisition of task data of the task to be processed and structured information corresponding to the task to be processed, the structured information includes structural data and multiple knowledge data, the structural data includes multiple nodes and the hierarchical relationship between each node, and after the node represents the key point of the task to be processed, the target node can be determined from multiple nodes based on the task data, wherein the target node is selected from multiple nodes in the structured information, and the node data of the target node has a high degree of matching with the task data.

[0075] Based on this, after obtaining the task data of the task to be processed and the structured information corresponding to the task to be processed, since the structured information includes structural data and multiple knowledge data, the structural data includes multiple nodes and the hierarchical relationship between each node, and the node represents the key point of the task to be processed, it is possible to match the node data from multiple nodes based on the task data, and determine the target node with a higher degree of match with the task data among the multiple nodes.

[0076] Furthermore, considering that the structure data contains multiple nodes and each node has corresponding node data, when selecting the target node, the node data corresponding to each node can be matched with the task data. The specific implementation is as follows:

[0077] Determine the node data of each node in the structural data; match the task data with the node data of each node, and determine the target node from the multiple nodes according to the matching result.

[0078] Specifically, node data is the data represented by the node in the structural data. There is a hierarchical relationship between nodes, and there is also a hierarchical relationship between node data; matching can be semantic matching or feature vector matching.

[0079] Based on this, the node data of each node in the structure data is determined. The task data is matched with the node data of each node respectively to obtain the matching degree between the task data and each node data, and the target node is determined from the multiple nodes according to the matching degree.

[0080] Continuing with the previous example, we match the task data "How to register a birth" with the node data of each node in the structure data. If the node "Birth Registration" in the structure data has a high degree of match with the task data, we select it as the target node.

[0081] In summary, the target node is determined by matching the task data with the node data of each node, thereby improving the efficiency of determining the target node.

[0082] Step 206: Input the node data of the target node and the structural data into a task generation model to generate a guided interaction task, wherein the task generation model is a model based on deep learning.

[0083] Specifically, after determining the target node from multiple nodes based on the task data, the node data and structural data of the target node can be input into the task generation model to generate a guided interaction task, wherein the task generation model is a model based on deep learning, wherein the task generation model can be a large model based on deep learning, and the task generation model is combined with an agent, which has the ability to perceive the environment, plan, and act. A guided interaction task is generated based on the task execution path after the task generation model plans the task execution path based on the node data, structural data, and task data after perceiving the environmental information; the guided interaction task is used to guide the questioner to answer questions in a knowledge question-answering scenario and determine the answer corresponding to the task data based on the answer information submitted by the questioner.

[0084] Based on this, after determining the target node from multiple nodes based on the task data, the node data and structure data of the target node are input into the task generation model to generate a guided interaction task. The task data can also be input into the task generation model together. After generating the initial guided interaction task, the initial guided interaction task is updated based on the task data to obtain the guided interaction task.

[0085] Furthermore, considering the hierarchical relationship between nodes in the structured data, the nodes in the structured data do not exist in isolation. When generating guided interaction tasks, it is also necessary to refer to the hierarchical relationship between nodes and the data association relationship under the hierarchical relationship. The specific implementation is as follows:

[0086] The node data of the target node and the structural data are input into a task generation model, and the task generation model is used to determine structural link data matching the node data in the structural data; and a guided interaction task is generated based on the structural link data.

[0087] Specifically, the structural link data refers to a node link including a target node determined in the structural data, and the arrangement order of the nodes in the node link matches the hierarchical relationship of the nodes in the structural data.

[0088] Based on this, the node data and structure data of the target node are input into the task generation model. The task generation model is then used to determine the structural link data that matches the node data in the structure data. Based on the node data of the nodes contained in the structural link data and the associations between the nodes, a guided interaction task is generated.

[0089] Continuing with the above example, if the task data is "How to register births in Region A", the target node corresponding to the task data is the "Region A" node in the structural data. Then, the structural data link containing the "Region A" node is determined: "Birth Registration" - "Region A" - "Legitimate Children", and "Birth Registration" - "Region A" - "Illegitimate Children", and then a guided interaction task is generated based on these two structural link data.

[0090] In summary, based on the node data of the nodes contained in the structural link data and the association relationship between the nodes, the guided interaction task is generated. The subsequent guided interaction task execution process is the process of gradually determining the task processing results of the tasks to be processed.

[0091] Furthermore, considering that there are many nodes in the structured data and there is a hierarchical relationship between each node, in order to simplify the guided interaction task, the generation of the guided interaction task can be achieved based on the neighboring nodes of the target node. The specific implementation is as follows:

[0092] Determine the node position information of the target node data in the structural link data; determine the neighboring node data of the target node data in the structural link data based on the node position information; generate a guided interaction task based on the neighboring node data, wherein the neighboring node data is the preceding node data and / or subsequent node data of the target node data.

[0093] Specifically, the node position information includes but is not limited to the hierarchical information of the target node in the structural data; the neighboring node data is the node data corresponding to the nodes adjacent to the target node in the hierarchical dimension; the preceding node data is the node data of the node adjacent to the target node; and the subsequent node data is the node data of the node adjacent to the target node.

[0094] Based on this, node position information of the target node data in the structured link data is determined. Based on the node position information, at least one neighboring node data of the target node data is determined in the structured link data. A guided interaction task is generated based on the neighboring node data. The neighboring node data can be the preceding node data of the target node data, the following node data of the target node data, or both the preceding node data and the following node data of the target node data.

[0095] Continuing with the previous example, if the task data is "How to register births in Region A," the target node could be the "Region A" node in the structured data. Combined with the node's location information, the neighboring nodes would be "Birth Registration," "Legitimate Children," and "Illegitimate Children." This allows us to generate guided interaction tasks based on the neighboring node data.

[0096] In summary, guided interaction tasks are generated based on neighboring node data, thereby simplifying guided interaction tasks and improving the processing efficiency of tasks to be processed.

[0097] Furthermore, considering that task data may exist in historical data processed by the task processing model, after the node data and structure data of the target node are input into the task generation model, the task generation model can combine the historical data related to the node data to generate a guided interaction task. The specific implementation is as follows:

[0098] The node data of the target node and the structural data are input into a task generation model, and the historical data associated with the node data is determined using the task generation model; a text processing task is generated based on the historical data, the node data and the structural data, and the text processing task is used as the guided interaction task.

[0099] Specifically, the historical data may be the task data of the previously processed task, the determined structured information, the target node, and the task processing result; the text generation task is the task to be executed corresponding to the task data and used to determine the result matching the task data.

[0100] Based on this, the node data and structural data of the target node are input into the task generation model, and the historical data associated with the node data is determined using the task generation model. Based on the node data and structural data, a text processing task is generated based on the historical data, and the text processing task is used as the guided interaction task. This allows the influence of historical data to be added to the process of guiding the interactive task generation.

[0101] Continuing with the previous example, if the task data is "How to register births in Region A," the target node can be the node "Region A" in the structured data. "Region A" is then input into the task generation model along with the structured data. The task generation model combines the historical task data "How to register births of children born in Region A," the target node corresponding to this historical task data, and the task processing results to generate a text generation task. Upon completion of this task generation, the answer data corresponding to the task data is obtained.

[0102] In summary, text processing tasks are generated based on historical data, node data, and structural data. By referring to historical data, the credibility of guided interaction tasks is improved.

[0103] Step 208: Based on the interaction result obtained by executing the guided interaction task, target knowledge data is determined from the plurality of knowledge data to obtain the task processing result of the task to be processed.

[0104] Specifically, after the node data and structural data of the target node are input into the task generation model to generate a guided interaction task, wherein the task generation model is a model based on deep learning, the target knowledge data can be determined from multiple knowledge data based on the interaction result obtained by executing the guided interaction task, and the task processing result of the task to be processed can be obtained, wherein the interaction result is the task execution result of the guided interaction task; the target knowledge data is the data in the knowledge data that has a higher degree of matching with the task data; accordingly, the task processing result can be the target knowledge data.

[0105] Based on this, in the above-mentioned process, the node data and structural data of the target node are input into the task generation model to generate a guided interaction task, and the guided interaction task is executed. Based on the interaction result obtained by executing the guided interaction task, the target knowledge data is determined from multiple knowledge data, and the target knowledge data is used as the task processing result of the task to be processed, or the task processing result of the task to be processed is generated based on the target knowledge data, so as to achieve the purpose of simplifying the target knowledge data.

[0106] Furthermore, considering that the purpose of executing the guided interaction task is to obtain the task processing result of the task to be processed, and the number of interactive subtasks in the guided interaction task will affect the accuracy of the task processing result, the specific implementation is as follows:

[0107] In the case where the guided interaction task includes one interaction subtask, target knowledge data is determined from the multiple knowledge data based on the sub-interaction result obtained by executing the interaction subtask, and the task processing result of the task to be processed is obtained; in the case where the guided interaction task includes at least two interaction subtasks, based on the first interaction result obtained by executing the first interaction subtask of the at least two interaction subtasks, the second interaction subtask of the at least two interaction subtasks is updated and executed until the interaction result is obtained, and based on the interaction result, target knowledge data is determined from the multiple knowledge data, and the task processing result of the task to be processed is obtained.

[0108] Specifically, the interactive subtask is a subtask in the guided interactive task. The execution result obtained after all subtasks in the guided interactive task are completed is the execution result of the guided interactive task; accordingly, the sub-interaction result is the execution result of the interactive subtask.

[0109] Based on this, when the guided interaction task includes one interactive subtask, the execution result of the interactive subtask is the interaction result of the guided interaction task. Based on the sub-interaction result obtained from executing the interactive subtask, the target knowledge data is determined from the multiple knowledge data to obtain the task processing result of the task to be processed. When the guided interaction task includes at least two interactive subtasks, the execution results obtained after the at least two interactive subtasks are executed in sequence are the interaction result of the guided interaction task. The first interactive subtask of the at least two interactive subtasks is executed to obtain the first interaction result. The second interactive subtask of the at least two interactive subtasks is updated based on the first interaction result, and the second interactive subtask is executed. This process is repeated until the last interactive subtask in the at least two interactive subtasks is completed, at which point the interaction result is obtained. Based on the interaction result, the target knowledge data is determined from the multiple knowledge data to obtain the task processing result of the task to be processed.

[0110] Continuing with the above example, if the task data is "How to register births in Region A," the generated guided interaction task includes the question "Is the birth legitimate?" and this question is fed back to the questioner. If the questioner answers "yes," the guided interaction task is completed, resulting in the interaction result "Birth registration policy for legitimate children in Region A." If the task data is "How to register births," the generated guided interaction task includes two questions: "Is the birth legitimate?" and "Region." The question "Region" is fed back to the questioner first. If the questioner answers "Region A," the question "Is the birth legitimate?" is fed back to the questioner. If the questioner answers "yes," the guided interaction task is completed, resulting in the interaction result "Birth registration policy for legitimate children in Region A."

[0111] In summary, the interactive subtasks included in the guided interactive task are executed, and target knowledge data is determined from multiple knowledge data based on the interactive results, thereby improving the matching degree between the task processing results and the task to be processed.

[0112] Furthermore, when the guided interaction task contains two tasks, the execution of the second task needs to depend on the execution result of the first task. Therefore, after the first task is completed, the second task can be updated based on the execution result before execution. The specific implementation is as follows:

[0113] Determine a first guidance task and a second guidance task included in the guidance interaction task, wherein the first guidance task and the second guidance task have an execution order relationship; execute the first guidance task to obtain a first execution result; based on the first execution result, determine a first node from the multiple nodes, and update the second guidance task to a target guidance task based on the first node; execute the target guidance task to obtain a target execution result, and use the target execution result as the interaction result of the guidance interaction task.

[0114] Specifically, the first guidance task and the second guidance task are two adjacent subtasks, and the node level of the node corresponding to the first guidance task is higher than the node level of the node corresponding to the second guidance task; the first node is the node determined in the structural data based on the first execution result after the first guidance task is completed, and the node level of the first node is lower than the node level of the node corresponding to the first guidance task.

[0115] Based on this, a first guidance task and a second guidance task, which have an execution order relationship, are determined within the guidance interaction task. The first guidance task is executed to obtain a first execution result. Based on the first execution result, a first node is determined from a plurality of nodes in the structure data, where the node level of the first node is lower than the node level of the node corresponding to the first guidance task. Based on the first node, the second guidance task is updated as a target guidance task. The target guidance task is executed to obtain a target execution result. The target execution result is used as the interaction result of the guidance interaction task.

[0116] Continuing with the above example, when the task data is "How to register a birth", the generated guided interaction task includes the two tasks of "legitimate birth" and "location". The first guided task "location" is executed, and the question of "location" is fed back to the questioner. When the questioner answers "Region A", the next-level nodes "legitimate children" and "illegitimate children" corresponding to the node "Region A" are determined in the result data, and "legitimate birth" is updated to the second guided task of "whether it is legitimate", and the question of "whether it is legitimate" is fed back to the questioner. When the questioner answers "yes", the guided interaction task is completed, and the interactive result of "Birth Registration Policy for Legitimate Children in Region A" is obtained.

[0117] In summary, the second guidance task is updated and then executed, so that the influence of the execution result of the first guidance task is incorporated into the execution process of the second guidance task, which plays a correction role on the second guidance task.

[0118] Furthermore, after the guided interaction task is completed, the target knowledge data needs to be determined based on the knowledge data. The determination of the target knowledge data depends on the node data of each node in the structural data. The specific implementation is as follows:

[0119] The guided interaction task is executed to obtain an interaction result; a target link corresponding to the interaction result is determined in the structure data, and target knowledge data is determined from the plurality of knowledge data based on the target link.

[0120] Specifically, the link nodes in the target link are node links with a connection relationship in the structural data, and the hierarchical structures of the nodes between the link nodes included in the target link are different.

[0121] Based on this, a guided interaction task is executed to obtain an interaction result; a target link corresponding to the interaction result is determined in the structure data, and target knowledge data is determined in the plurality of knowledge data based on the target link.

[0122] Continuing with the previous example, if we obtain the interaction result "Birth registration of legitimate children in Region A," we determine the target link in the structured data based on the interaction result: "Birth registration" - "Region A" - "legitimate children." Based on this target link, we determine the target knowledge data item "Birth registration policy for legitimate children in Region A" in the knowledge data.

[0123] In summary, target knowledge data is determined from multiple pieces of knowledge data based on the target link, thereby improving the accuracy of determining the target knowledge data.

[0124] One embodiment of the present disclosure obtains task data of a pending task and structured information corresponding to the pending task, wherein the structured information includes structured data and multiple knowledge data, the structured data includes multiple nodes and hierarchical relationships between the nodes, and the nodes represent key points of the pending task; based on the task data, a target node is determined from multiple nodes; the node data and structured data of the target node are input into a task generation model to generate a guided interaction task, wherein the task generation model is a deep learning-based model; based on the interaction results obtained by executing the guided interaction task, the target knowledge data is determined from multiple knowledge data to obtain the task processing result of the pending task. The structured data and knowledge data in the structured information are decoupled, the structured data guides the task generation model, the guided interaction task is generated based on the structured data, and then the task processing result of the pending task is determined by combining the interaction results obtained by executing the guided interaction task and the knowledge data, thereby improving the task processing capability while improving the result accuracy.

[0125] The following further illustrates the task processing method provided by the present disclosure using the application of the task processing method in knowledge question answering as an example, in conjunction with FIG4 . FIG4 shows a flowchart of the processing process of a task processing method provided by an embodiment of the present disclosure, which specifically includes the following steps.

[0126] Step 402: Construct a structured information set, wherein the structured information set includes a plurality of structured information, the structured information includes structure data and knowledge data, the structured data includes a plurality of nodes, and a hierarchical relationship between the nodes.

[0127] As shown in Figure 5, the knowledge construction module constructs structured information (DocGraph). This structured information includes both structural data and knowledge data, and the two are decoupled. The structural data, which can be a graph structure, guides the planning of the large model, while the knowledge data provides the knowledge source for end-to-end answer generation. Structural data has distinct nodes and levels. Levels refer to hierarchical information. For example, the "Birth Registration" node in the structural data has a level of 1.

[0128] Step 404: Obtain the task to be processed, and determine the target link data in the link data corresponding to the structure data based on the task data of the task to be processed.

[0129] The tasks to be processed can be knowledge dialogue tasks, or information acquisition tasks such as question-and-answer tasks and human-computer interaction tasks in which users obtain information; accordingly, the task data is data submitted by users such as question data and dialogue data, or data triggered and submitted by user operations, and can also be data in human-computer interaction scenarios.

[0130] When the task data of the task to be processed is the question "How to register a birth", the agent corresponding to the large model is used to search in the structured information set based on the question to determine the target link data that matches the task data.

[0131] Step 406: Determine the target structure data corresponding to the target link data.

[0132] The graph structure of the target link data is used as the target structure data.

[0133] Step 408: Determine a target node from among the multiple nodes included in the target structure data based on the task data.

[0134] Step 410: Determine environmental data based on the knowledge dialogue model.

[0135] The knowledge dialogue model is a large model that uses its corresponding agent to perceive environmental data (such as the current time and project-level parameters) and form memory data (multi-round context short-term memory and long-term memory of structured data and knowledge data). Environmental data can include multi-round user conversation information, system time, project-level parameter configuration, etc.

[0136] Step 412: Call the knowledge dialogue model to generate a guided interaction task based on the environment data, structure data, target node and task data.

[0137] Guided interaction tasks are tasks generated based on planning. Planning includes, but is not limited to, clarification, questioning, recommendation, guidance, search, and direct output. The agent corresponding to the large model will utilize the large model's powerful planning capabilities to plan and determine the task to be performed based on acquired environmental data, structural data, target nodes, and task data. In practical applications, CoT decisions (chain thinking) can be used to determine the next step or task to be executed.

[0138] Step 414: Based on the interaction result obtained by executing the guided interaction task, determine whether the interaction result matches the task to be processed; if so, execute step 416; if not, execute step 418.

[0139] Step 416: Use the interaction result as the task execution result of the task to be processed.

[0140] Step 418: Generate a target interactive task based on the interaction result and execute it until a task execution result matching the task to be processed is obtained.

[0141] Execute the target interactive task, use the agent corresponding to the big model to complete the knowledge search, guidance, or based on the determined counter-question information, request the big model again to obtain the task execution result. The task execution result is the answer to the task data question.

[0142] In summary, an embodiment of the present disclosure obtains the task data of the task to be processed and the structured information corresponding to the task to be processed, wherein the structured information includes structured data and multiple knowledge data, the structured data includes multiple nodes and the hierarchical relationship between each node, and the node represents the key point of the task to be processed; based on the task data, the target node is determined from the multiple nodes; the node data and structured data of the target node are input into the task generation model to generate a guided interaction task, wherein the task generation model is a model based on deep learning; based on the interaction result obtained by executing the guided interaction task, the target knowledge data is determined from the multiple knowledge data to obtain the task processing result of the task to be processed. The structured data and the knowledge data in the structured information are decoupled, the task generation model is guided by the structured data, the guided interaction task is generated based on the structured data, and then the task processing result of the task to be processed is determined in combination with the interaction result and the knowledge data obtained by executing the guided interaction task, thereby improving the task processing capability while improving the result accuracy.

[0143] Corresponding to the above method embodiment, the present disclosure also provides an embodiment of a task processing device. FIG6 shows a schematic diagram of the structure of a task processing device provided by an embodiment of the present disclosure. As shown in FIG6, the device includes:

[0144] An acquisition module 602 is configured to acquire task data of a task to be processed and structured information corresponding to the task to be processed, wherein the structured information includes structure data and a plurality of knowledge data, the structure data includes a plurality of nodes and hierarchical relationships between the nodes, and the nodes represent key points of the task to be processed;

[0145] A determination module 604 is configured to determine a target node from the plurality of nodes based on the task data;

[0146] A generation module 606 is configured to input the node data of the target node and the structure data into a task generation model to generate a guided interaction task, wherein the task generation model is a deep learning-based model;

[0147] The execution module 608 is configured to determine target knowledge data from the plurality of knowledge data based on the interaction result obtained by executing the guided interaction task, and obtain the task processing result of the task to be processed.

[0148] In an optional embodiment, the acquisition module 602 is further configured to:

[0149] Obtaining task data of a task to be processed and task information of the task to be processed;

[0150] Determine structured information corresponding to the task to be processed according to the task information.

[0151] In an optional embodiment, the acquisition module 602 is further configured to:

[0152] Acquire target data from a database according to the task information, and generate structured information based on the target data; or,

[0153] Structural information is selected from a pre-stored set of structural information according to the task information.

[0154] In an optional embodiment, the determining module 604 is further configured to: determine node data of each node in the structural data;

[0155] The task data is matched with the node data of each node, and a target node is determined from the multiple nodes according to the matching result.

[0156] In an optional embodiment, the generating module 606 is further configured to:

[0157] Inputting the node data of the target node and the structure data into a task generation model, and using the task generation model to determine the structure link data matching the node data in the structure data;

[0158] A guided interaction task is generated based on the structure link data.

[0159] In an optional embodiment, the generating module 606 is further configured to:

[0160] Determining node position information of the target node data in the structure link data;

[0161] Determining neighboring node data of the target node data in the structure link data according to the node location information;

[0162] A guided interaction task is generated based on the neighboring node data, wherein the neighboring node data is preceding node data and / or subsequent node data of the target node data.

[0163] In an optional embodiment, the generating module 606 is further configured to:

[0164] Inputting the node data of the target node and the structure data into a task generation model, and using the task generation model to determine historical data associated with the node data;

[0165] A text processing task is generated based on the historical data, the node data and the structural data, and the text processing task is used as the guided interaction task.

[0166] In an optional embodiment, the execution module 608 is further configured to:

[0167] In the case where the guided interaction task includes an interaction subtask, determining target knowledge data from the plurality of knowledge data based on a sub-interaction result obtained by executing the interaction subtask, and obtaining a task processing result of the task to be processed;

[0168] In the case where the guided interaction task includes at least two interaction sub-tasks, based on the first interaction result obtained by executing the first interaction sub-task among the at least two interaction sub-tasks, the second interaction sub-task among the at least two interaction sub-tasks is updated and executed until the interaction result is obtained, and based on the interaction result, the target knowledge data is determined from the multiple knowledge data to obtain the task processing result of the task to be processed.

[0169] In an optional embodiment, the execution module 608 is further configured to:

[0170] Determining a first guidance task and a second guidance task included in the guidance interaction task, wherein the first guidance task and the second guidance task have an execution order relationship;

[0171] Executing the first guiding task to obtain a first execution result;

[0172] determining a first node from the plurality of nodes based on the first execution result, and updating the second guided task to a target guided task based on the first node;

[0173] The target guidance task is executed to obtain a target execution result, and the target execution result is used as the interaction result of the guidance interaction task.

[0174] In an optional embodiment, the execution module 608 is further configured to:

[0175] Executing the guided interaction task to obtain an interaction result;

[0176] A target link corresponding to the interaction result is determined in the structure data, and target knowledge data is determined from the plurality of knowledge data based on the target link.

[0177] In summary, an embodiment of the present disclosure obtains the task data of the task to be processed and the structured information corresponding to the task to be processed, wherein the structured information includes structured data and multiple knowledge data, the structured data includes multiple nodes and the hierarchical relationship between each node, and the node represents the key point of the task to be processed; based on the task data, the target node is determined from the multiple nodes; the node data and structured data of the target node are input into the task generation model to generate a guided interaction task, wherein the task generation model is a model based on deep learning; based on the interaction result obtained by executing the guided interaction task, the target knowledge data is determined from the multiple knowledge data to obtain the task processing result of the task to be processed. The structured data and the knowledge data in the structured information are decoupled, the task generation model is guided by the structured data, the guided interaction task is generated based on the structured data, and then the task processing result of the task to be processed is determined in combination with the interaction result and the knowledge data obtained by executing the guided interaction task, thereby improving the task processing capability while improving the result accuracy.

[0178] The above is a schematic scheme of a task processing device of this embodiment. It should be noted that the technical scheme of the task processing device and the technical scheme of the task processing method described above are of the same concept. For details not described in detail in the technical scheme of the task processing device, please refer to the description of the technical scheme of the task processing method described above.

[0179] Referring to FIG. 7 , FIG. 7 shows a flowchart of a question-answering processing method provided according to an embodiment of the present disclosure, which specifically includes the following steps.

[0180] Step 702: Responding to a problem processing request submitted by a front-end user, obtaining problem data of a problem to be processed and structured information corresponding to the problem to be processed, wherein the structured information includes structure data and a plurality of knowledge data, the structure data includes a plurality of nodes and hierarchical relationships between the nodes, and the nodes represent key points of the problem to be processed;

[0181] Step 704: Determine a target node from the plurality of nodes based on the problem data;

[0182] Step 706: Input the node data and the structural data of the target node into the question-answering task generation model to generate a question-answering guidance task, wherein the question-answering task generation model is a deep learning-based model;

[0183] Step 708: Based on the interactive results obtained by executing the question-answering guidance task, target knowledge data is determined from the plurality of knowledge data, and a problem processing result of the problem to be processed is obtained;

[0184] Step 710: Send the problem handling result as feedback of the problem handling request to the front-end user.

[0185] In actual applications, after receiving a question processing request submitted by a front-end user, the problem to be processed is determined based on the question processing request. The problem data of the problem to be processed and the structured information corresponding to the problem to be processed are obtained. The structured information includes structural data and multiple knowledge data. The structural data includes multiple nodes and the hierarchical relationship between each node. The node represents the key point of the problem to be processed. Based on the problem data, the target node is determined from multiple nodes, and the node data and structural data of the target node are input into the question-answering task generation model to generate a question-answering guidance task. Based on the interaction results obtained by executing the question-answering guidance task, the target knowledge data is determined from the multiple knowledge data to obtain the problem processing result of the problem to be processed. The problem processing result is sent to the front-end user as feedback for the question processing request. In this way, when the front-end user submits an enterprise knowledge question, enterprise knowledge question and answer can be realized.

[0186] In summary, an embodiment of the present disclosure obtains the task data of the task to be processed and the structured information corresponding to the task to be processed, wherein the structured information includes structured data and multiple knowledge data, the structured data includes multiple nodes and the hierarchical relationship between each node, and the node represents the key point of the task to be processed; based on the task data, the target node is determined from the multiple nodes; the node data and structured data of the target node are input into the task generation model to generate a guided interaction task, wherein the task generation model is a model based on deep learning; based on the interaction result obtained by executing the guided interaction task, the target knowledge data is determined from the multiple knowledge data to obtain the task processing result of the task to be processed. The structured data and the knowledge data in the structured information are decoupled, the task generation model is guided by the structured data, the guided interaction task is generated based on the structured data, and then the task processing result of the task to be processed is determined in combination with the interaction result and the knowledge data obtained by executing the guided interaction task, thereby improving the task processing capability while improving the result accuracy.

[0187] Corresponding to the above method embodiment, the present disclosure also provides an embodiment of a question-answering processing device. FIG8 shows a schematic diagram of the structure of a question-answering processing device provided by an embodiment of the present disclosure. As shown in FIG8, the device includes:

[0188] The acquisition module 802 is configured to obtain, in response to a problem processing request submitted by a front-end user, problem data of a problem to be processed and structured information corresponding to the problem to be processed, wherein the structured information includes structure data and a plurality of knowledge data, the structure data includes a plurality of nodes and hierarchical relationships between the nodes, and the nodes represent key points of the problem to be processed;

[0189] A first determining module 804 is configured to determine a target node from the plurality of nodes based on the problem data;

[0190] A generation module 806 is configured to input the node data of the target node and the structural data into the question-answering task generation model to generate a question-answering guidance task, wherein the question-answering task generation model is a deep learning-based model;

[0191] The second determining module 808 is configured to determine target knowledge data from the plurality of knowledge data based on the interaction result obtained by performing the question-answering guidance task, and obtain a problem processing result for the problem to be processed;

[0192] The sending module 810 is configured to send the problem handling result as feedback of the problem handling request to the front-end user.

[0193] In summary, an embodiment of the present disclosure obtains the task data of the task to be processed and the structured information corresponding to the task to be processed, wherein the structured information includes structured data and multiple knowledge data, the structured data includes multiple nodes and the hierarchical relationship between each node, and the node represents the key point of the task to be processed; based on the task data, the target node is determined from the multiple nodes; the node data and structured data of the target node are input into the task generation model to generate a guided interaction task, wherein the task generation model is a model based on deep learning; based on the interaction result obtained by executing the guided interaction task, the target knowledge data is determined from the multiple knowledge data to obtain the task processing result of the task to be processed. The structured data and the knowledge data in the structured information are decoupled, the task generation model is guided by the structured data, the guided interaction task is generated based on the structured data, and then the task processing result of the task to be processed is determined in combination with the interaction result and the knowledge data obtained by executing the guided interaction task, thereby improving the task processing capability while improving the result accuracy.

[0194] The above is a schematic diagram of a question-and-answer processing device according to this embodiment. It should be noted that the technical solution of this question-and-answer processing device and the technical solution of the aforementioned question-and-answer processing method are based on the same concept. For details not described in detail in the technical solution of the question-and-answer processing device, please refer to the description of the technical solution of the aforementioned question-and-answer processing method.

[0195] Figure 9 shows a block diagram of a computing device 900 according to one embodiment of the present disclosure. Components of the computing device 900 include, but are not limited to, a memory 910 and a processor 920. The processor 920 is connected to the memory 910 via a bus 930, and a database 950 is used to store data.

[0196] The computing device 900 also includes an access device 940 that enables the computing device 900 to communicate via one or more networks 960. Examples of such networks include a public switched telephone network (PSTN), a local area network (LAN), a wide area network (WAN), a personal area network (PAN), or a combination of communication networks such as the Internet. The access device 940 may include one or more of any type of network interface (e.g., a network interface card (NIC)) whether wired or wireless, such as an IEEE 902.11 wireless local area network (WLAN) wireless interface, a Worldwide Interoperability for Microwave Access (Wi-MAX) interface, an Ethernet interface, a universal serial bus (USB) interface, a cellular network interface, a Bluetooth interface, or a near field communication (NFC) interface.

[0197] In one embodiment of the present disclosure, the aforementioned components of the computing device 900 and other components not shown in FIG9 may also be connected to each other, for example, via a bus. It should be understood that the computing device structure block diagram shown in FIG9 is for illustrative purposes only and does not limit the scope of the present disclosure. Those skilled in the art may add or replace other components as needed.

[0198] The computing device 900 may be any type of stationary or mobile computing device, including a mobile computer or mobile computing device (e.g., a tablet computer, a personal digital assistant, a laptop computer, a notebook computer, a netbook computer, etc.), a mobile phone (e.g., a smartphone), a wearable computing device (e.g., a smartwatch, smart glasses, etc.), or other types of mobile devices, or a stationary computing device such as a desktop computer or a personal computer (PC). The computing device 900 may also be a mobile or stationary server.

[0199] The processor 920 is configured to execute the following computer-executable instructions, which implement the steps of the above method when executed by the processor.

[0200] The above is a schematic solution of a computing device of this embodiment. It should be noted that the technical solution of the computing device and the technical solution of the above method belong to the same concept. For details not described in detail in the technical solution of the computing device, please refer to the description of the technical solution of the above method.

[0201] An embodiment of the present disclosure further provides a computer-readable storage medium storing computer-executable instructions, which implement the steps of the above method when executed by a processor.

[0202] The above is a schematic solution of a computer-readable storage medium of this embodiment. It should be noted that the technical solution of the storage medium and the technical solution of the above method belong to the same concept. For details not described in detail in the technical solution of the storage medium, please refer to the description of the technical solution of the above method.

[0203] An embodiment of the present disclosure further provides a computer program, wherein when the computer program is executed in a computer, the computer is caused to execute the steps of the above method.

[0204] The above is an illustrative solution of a computer program of this embodiment. It should be noted that the technical solution of the computer program and the technical solution of the above method belong to the same concept, and any details not described in detail in the technical solution of the computer program can be referred to the description of the technical solution of the above method.

[0205] The foregoing description describes specific embodiments of the present disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in an order different from that described in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order shown or the sequential order to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0206] The computer instructions include computer program code, which may be in source code form, object code form, executable file, or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal, and software distribution medium. It should be noted that the content contained in the computer-readable medium may be appropriately increased or decreased according to the requirements of patent practice. For example, in some regions, according to patent practice, computer-readable media does not include electric carrier signals and telecommunication signals.

[0207] It should be noted that for the aforementioned method embodiments, for the sake of simplicity of description, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the embodiments of the present disclosure are not limited by the order of the actions described, because according to the embodiments of the present disclosure, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the embodiments of the present disclosure.

[0208] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0209] The preferred embodiments of the present disclosure disclosed above are only used to help illustrate the present disclosure. The optional embodiments do not describe all details in detail, nor do they limit the invention to only the specific embodiments described. Obviously, many modifications and variations can be made based on the content of the embodiments of the present disclosure. The present disclosure selects and specifically describes these embodiments in order to better explain the principles and practical applications of the embodiments of the present disclosure, so that those skilled in the art can better understand and utilize the present disclosure. The present disclosure is limited only by the claims and their full scope and equivalents.

Claims

1. A task processing method, comprising: Obtaining task data of a task to be processed and structured information corresponding to the task to be processed, wherein the structured information includes structure data and a plurality of knowledge data, the structure data includes a plurality of nodes and hierarchical relationships between the nodes, and the nodes represent key points of the task to be processed; Determining a target node from the plurality of nodes based on the task data; Inputting the node data of the target node and the structure data into a task generation model to generate a guided interaction task, wherein the task generation model is a model based on deep learning; Determining target knowledge data from the plurality of knowledge data based on an interaction result obtained by executing the guided interaction task, to obtain a task processing result of the task to be processed.

2. The task processing method according to claim 1, wherein the obtaining task data of the task to be processed and the structured information corresponding to the task to be processed comprises: Obtaining task data of the task to be processed, and task information of the task to be processed; Determining the structured information corresponding to the task to be processed according to the task information.

3. The task processing method according to claim 2, wherein the determining the structured information corresponding to the task to be processed according to the task information comprises: Obtaining target data in a database according to the task information, and generating structured information based on the target data; Or, Selecting structured information from a pre-stored set of structured information according to the task information.

4. The task processing method according to any one of claims 1 to 3, wherein the determining a target node from the plurality of nodes based on the task data comprises: Determining node data of each node in the structure data; Matching the task data with the node data of each node, and determining a target node from the plurality of nodes according to the matching result.

5. The task processing method according to any one of claims 1 to 4, wherein the inputting the node data of the target node and the structure data into a task generation model to generate a guided interaction task comprises: Inputting the node data of the target node and the structure data into a task generation model, and using the task generation model to determine structure link data matching the node data in the structure data; Generating a guided interaction task based on the structure link data.

6. The task processing method according to claim 5, wherein the generating a guided interaction task based on the structure link data comprises: Determining node position information of the target node data in the structure link data; Determining neighbor node data of the target node data in the structure link data according to the node position information; Generating a guided interaction task based on the neighbor node data, wherein the neighbor node data is pre-order node data and / or post-order node data of the target node data.

7. The task processing method according to claim 6, wherein the inputting the node data of the target node and the structure data into a task generation model to generate a guided interaction task comprises: Input the node data of the target node and the structure data into the task generation model, and use the task generation model to determine the historical data associated with the node data; Generate a text processing task based on the historical data, the node data, and the structure data, and use the text processing task as the guided interaction task.

8. The task processing method according to any one of claims 1 to 7, wherein determining the target knowledge data from the multiple knowledge data based on the interaction result obtained by executing the guided interaction task to obtain the task processing result of the to-be-processed task includes: When the guided interaction task includes one interaction subtask, determining the target knowledge data from the multiple knowledge data based on the sub-interaction result obtained by executing the interaction subtask to obtain the task processing result of the to-be-processed task; When the guided interaction task includes at least two interaction subtasks, updating and executing the second interaction subtask among the at least two interaction subtasks based on the first interaction result obtained by executing the first interaction subtask among the at least two interaction subtasks until an interaction result is obtained, and determining the target knowledge data from the multiple knowledge data based on the interaction result to obtain the task processing result of the to-be-processed task.

9. The task processing method according to any one of claims 1 to 8, wherein when the guided interaction task includes a first guided task and a second guided task, the execution of the guided interaction task includes: Determine the first guided task and the second guided task included in the guided interaction task, wherein there is an execution sequence relationship between the first guided task and the second guided task; Execute the first guided task to obtain a first execution result; Based on the first execution result, determine a first node from the multiple nodes, and update the second guided task to a target guided task based on the first node; Execute the target guided task to obtain a target execution result, and use the target execution result as the interaction result of the guided interaction task.

10. The task processing method according to any one of claims 1 to 9, wherein determining the target knowledge data from the multiple knowledge data based on the interaction result obtained by executing the guided interaction task includes: Execute the guided interaction task to obtain an interaction result; Determine a target link corresponding to the interaction result in the structure data, and determine the target knowledge data from the multiple knowledge data based on the target link.

11. A question and answer processing method, comprising: Obtain the question data of the to-be-processed question and the structured information corresponding to the to-be-processed question in response to a question processing request submitted by a front-end user, wherein the structured information includes structure data and multiple knowledge data, the structure data includes multiple nodes and the hierarchical relationship between each node, and the node represents the key point of the to-be-processed question; Determine a target node from the multiple nodes based on the question data; Input the node data of the target node and the structure data into the Q&A task generation model to generate a Q&A guidance task, where the Q&A task generation model is a deep learning-based model; Based on the interaction result obtained by executing the Q&A guidance task, determine the target knowledge data from the multiple knowledge data to obtain the problem processing result of the problem to be processed; Send the problem processing result as the feedback of the problem processing request to the front-end user.

12. A task processing device, comprising: An acquisition module configured to acquire the task data of the task to be processed and the structured information corresponding to the task to be processed, where the structured information includes structure data and multiple knowledge data, the structure data includes multiple nodes and the hierarchical relationships between the nodes, and the nodes represent the key points of the task to be processed; A determination module configured to determine a target node from the multiple nodes based on the task data; A generation module configured to input the node data of the target node and the structure data into a task generation model to generate a guided interaction task, where the task generation model is a deep learning-based model; An execution module configured to determine the target knowledge data from the multiple knowledge data based on the interaction result obtained by executing the guided interaction task to obtain the task processing result of the task to be processed.

13. A computing device, comprising: A memory and a processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the method according to any one of claims 1 to 11 are implemented.

14. A computer-readable storage medium storing computer-executable instructions, and when the computer-executable instructions are executed by a processor, the steps of the method according to any one of claims 1 to 11 are implemented.

15. A computer program, when the computer program is executed on a computer, causes the computer to execute the steps of the method according to any one of claims 1 to 11.

Images