Instruction processing method based on training system, training system and storage medium

Abstract

The invention relates to the technical field of training, and particularly discloses an instruction processing method based on a training system, the training system and a storage medium. Comprises: receiving an input instruction of a user; identifying the intention of the user according to the input instruction, and generating a target instruction corresponding to the identified intention; identifying and splitting the target instruction based on a classifier; extracting each instruction parameter and a corresponding execution interface from the splitting result; and calling each execution interface to execute the target instruction. According to the method and the device, the intention of the user can be quickly and accurately pre-judged and identified for the incomplete or indefinite input instruction, and the target instruction corresponding to the identified intention is generated, so that the intelligent degree and the self-adaptive capability of the training system are effectively improved. And meanwhile, the processing efficiency of the target instruction can be effectively improved, a relatively high response speed can be kept even if the training system processes a complex instruction, and the overall use fluency and the user experience feeling of the training system are improved.

Description

Technical Field

[0001] The present application relates to the field of training technology, and in particular to an instruction processing method based on a training system, a training system and a storage medium. Background Art

[0002] With the advancement of technology, automated training systems have become increasingly popular, and users can conduct training independently through the training system. Specifically, users can input relevant instructions into the training system to enable the training system to perform relevant actions.

[0003] However, when using traditional training systems, when the instructions input by the user do not conform to the preset instructions within the system, it is difficult to parse the user's true intentions and cannot output the results the user wants. This shows that the traditional training system has a low level of intelligence and adaptability. Summary of the invention

[0004] Based on this, it is necessary to provide an instruction processing method based on a training system, a training system and a computer-readable storage medium to address the above problems.

[0005] According to a first aspect of an embodiment of the present application, there is provided an instruction processing method based on a training system, comprising: Receive user input instructions; Recognizing the user's intention according to the input instruction, and generating a target instruction corresponding to the recognized intention; Identifying and splitting the target instruction based on a classifier; Extract each instruction parameter and the corresponding execution interface from the split result; Each of the execution interfaces is called to execute the target instruction.

[0006] In one embodiment, the step of identifying the user's intention according to the input instruction and generating a target instruction corresponding to the identified intention includes: When the input instruction is unclear, extracting a number of instructions related to the intention of the input instruction from a pre-stored instruction set according to a preset intelligent model and outputting them; receiving a user's selection instruction for a plurality of output instructions; According to the user's selection instruction, the instruction selected by the user is determined as the target instruction.

[0007] In one embodiment, the step of identifying the user's intention according to the input instruction and generating a target instruction corresponding to the identified intention further includes: When the input instruction is clear, the input instruction is determined as the target instruction.

[0008] In one embodiment, before the step of identifying and splitting the target instruction based on the classifier, the instruction processing method based on the training system further includes: The classifier is pre-trained, and the pre-training includes: constructing instruction sets for several functional areas in the training system, and dynamically generating a large number of instruction processing examples for each instruction in each instruction set in combination with an intelligent model and a classifier.

[0009] In one embodiment, the several functional areas include training management, course management, examination management, survey management, practice management, knowledge management, micro-class management, live broadcast management, decoration operation, student search, to-do query, personal task query and personal information query.

[0010] In one embodiment, the step of extracting each instruction parameter and the corresponding execution interface from the split result includes: When the initial extraction of each instruction parameter fails, the extraction of each instruction parameter is performed in combination with the preset intelligent model and each instruction processing example.

[0011] In one embodiment, after the step of extracting each instruction parameter and the corresponding execution interface from the splitting result, the instruction processing method based on the training system further includes: When the instruction parameter is a time parameter, the time parameter is parsed in combination with a pre-stored intelligent model to convert the time parameter into a date format.

[0012] In one embodiment, before the step of calling each execution interface to execute the target instruction, the instruction processing method based on the training system further includes: Providing a secondary confirmation interface for the user to receive feedback information from the user on each of the command parameters; When receiving a confirmation instruction from the user based on the secondary confirmation interface, executing the step of calling each of the execution interfaces to execute the target instruction; When a correction instruction from the user based on the secondary confirmation interface is received, each of the instruction parameters is corrected according to the correction instruction.

[0013] According to a second aspect of an embodiment of the present application, a training system is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the above-mentioned instruction processing method based on the training system when executing the computer program.

[0014] According to a third aspect of an embodiment of the present application, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the above-mentioned instruction processing method based on the training system is implemented.

[0015] The instruction processing method, training system and storage medium based on the training system provided by the embodiment of the present application can identify the user's intention according to the user's input instruction, and generate a target instruction corresponding to the identified intention, then identify and split the target instruction based on the classifier, extract each instruction parameter and the corresponding execution interface from the split result, and finally call each execution interface to execute the target instruction to output the data required by the user. Compared with the traditional technology, in the scheme of the present application, the user's intention can be quickly and accurately predicted and identified for incomplete or unclear input instructions, and a target instruction corresponding to the identified intention is generated, which effectively improves the intelligence and adaptive ability of the training system. At the same time, the target instruction can be identified and split based on the classifier, and then each target parameter and the corresponding execution interface are extracted respectively, and then the execution interface is called to execute the target instruction, thereby effectively improving the processing efficiency of the target instruction, even if the training system can maintain a high response speed when processing complex instructions, improving the overall use fluency and user experience of the training system. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A flowchart of an instruction processing method based on a training system provided in one embodiment of the present application; Figure 2 A flowchart of an instruction processing method based on a training system provided in one embodiment of the present application; Figure 3 A flowchart of an instruction processing method based on a training system provided in one embodiment of the present application; Figure 4 A schematic diagram of the structure of a training system provided in one embodiment of the present application. DETAILED DESCRIPTION

[0017] In order to facilitate the understanding of the present application, the present application will be described more fully below with reference to the relevant drawings. The preferred embodiments of the present application are given in the drawings. However, the present application can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to understand the disclosure of the present application more thoroughly and comprehensively.

[0018] In this application, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements, unless otherwise clearly defined. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0019] The terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein in the specification of this application are only for the purpose of describing specific embodiments and are not intended to limit this application. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0021] As described in the background art, traditional training systems lack sufficient intelligence and adaptive capabilities when it comes to understanding user intentions and executing specific instructions.

[0022] In order to solve the above problems, the embodiments of the present application provide an instruction processing method based on a training system, a training system, and a computer-readable storage medium.

[0023] In one embodiment, a training system-based instruction processing method is provided, through which the user's intention can be quickly and accurately identified, thereby outputting correct results, improving the overall user satisfaction, and improving the intelligence and adaptability of the training system.

[0024] Reference Figure 1 The instruction processing method based on the training system provided in this embodiment includes the following steps: Step S100: receiving an input instruction from a user.

[0025] First, when using the training system, the user can input instructions in the user interface provided by the training system. In actual applications, the user's input instructions may be complete instructions, or may be incomplete instructions or instructions with unclear intentions.

[0026] Step S300: Identify the user's intention according to the input instruction, and generate a target instruction corresponding to the identified intention.

[0027] After receiving the user's input instruction, the user's intention can be identified based on the input instruction, and a target instruction corresponding to the intention can be generated. As mentioned above, the input instruction is not necessarily a complete instruction or an instruction with a clear intention. For such instructions, in this embodiment, the user's intention can first be identified based on the input instruction. The identification process can be to predict the user's intention based on the user's input instruction and in combination with a pre-stored instruction set, and generate an instruction corresponding to the intention. In this embodiment, the instruction corresponding to the identified intention is defined as a target instruction. In this embodiment, the user's intention can also be predicted in other ways in combination with the user's input instruction, so as to generate an instruction corresponding to the intention.

[0028] Step S500: Identify and split the target instruction based on a classifier.

[0029] After the target instruction is generated, the target instruction can be parsed based on the classifier, and the target instruction can be specifically identified and split, and the target instruction can be split into several instruction parameters for subsequent processing. For example, if the target instruction is "query my courseware list", the target instruction can be split into two instruction parameters "my courseware" and "list" based on the classifier.

[0030] In this embodiment, the classifier may be pre-trained so that it can be applicable to the rapid and accurate segmentation of various target instructions.

[0031] Step S700: extract each instruction parameter and the corresponding execution interface from the split result.

[0032] After the target instruction is split, each instruction parameter can be extracted from the split result, and the execution interface corresponding to each instruction parameter can also be extracted. As in the above example, the two instruction parameters "My Courseware" and "List" can be extracted, and the execution interfaces corresponding to "My Courseware" and "List" can be extracted. When each execution interface is called, the specific data corresponding to each instruction parameter can be obtained.

[0033] Step S900: calling each of the execution interfaces to execute the target instruction.

[0034] After determining the execution interface corresponding to each instruction parameter, each execution interface can be called to execute the target instruction, thereby outputting the data required by the user.

[0035] The instruction processing method based on the training system provided in this embodiment can identify the user's intention according to the user's input instruction, generate a target instruction corresponding to the identified intention, and then identify and split the target instruction based on the classifier, extract each instruction parameter and the corresponding execution interface from the split result, and finally call each execution interface to execute the target instruction to output the data required by the user. Compared with the traditional technology, the scheme of the present application can quickly and accurately predict the user's intention for incomplete or unclear input instructions, and generate a target instruction corresponding to the identified intention, which effectively improves the intelligence and adaptability of the training system. At the same time, it can identify and split the target instruction based on the classifier, and then extract each target parameter and the corresponding execution interface respectively, and then call the execution interface to execute the target instruction, thereby effectively improving the processing efficiency of the target instruction, even if the training system can maintain a high response speed when processing complex instructions, improving the overall use fluency and user experience of the training system.

[0036] Reference Figure 2 In one embodiment, step S300, i.e., the step of identifying the user's intention according to the input instruction and generating a target instruction corresponding to the identified intention, includes: Step S310: When the input instruction is unclear, a number of instructions related to the intention of the input instruction are extracted from a pre-stored instruction set according to a preset intelligent model and output.

[0037] Unclear input instructions refer to situations where the input instructions are incomplete or the intention of the input instructions is unclear. For example, when the input instruction is "courseware", it is not a complete input instruction and the user's intention is unclear. When it is determined that the input instruction is unclear, several instructions related to the intention of the input instruction can be automatically extracted from the pre-stored instruction set according to the preset intelligent model, so that the user can choose the instruction that is closest to his or her true intention. Back to the above example, when the input instruction is "courseware", the intelligent model can be used to select instructions close to "courseware" from the instruction set to expand the input instruction "courseware". For example, the input instruction "courseware" can be expanded to multiple instructions such as "query my courseware", "create a new courseware process", and "courseware management and update".

[0038] The number of instructions extracted that are related to the intention of the input instruction may be at least three, thereby providing the user with room for selection and avoiding missing instructions that reflect the user's true intention as much as possible.

[0039] The extracted several instructions can be output to the human-computer interaction interface of the training system so that the user can select them.

[0040] That is to say, when the user's input information is limited, it is possible to quickly capture and respond to the user's potential needs, thereby improving the efficiency and accuracy of instruction processing.

[0041] Step S320: receiving a user's selection instruction for the outputted plurality of instructions.

[0042] When a plurality of instructions are output to the human-computer interaction interface of the training system, the user can select an instruction that meets his / her own needs from the plurality of instructions through the human-computer interaction interface.

[0043] Step S330: According to the user's selection instruction, the instruction selected by the user is determined as the target instruction.

[0044] When the user selects an instruction that meets his / her needs from the output instructions, the instruction selected by the user can be used as the target instruction, that is, the instruction that reflects the user's true intention, so as to facilitate the subsequent query and output of data according to the target instruction.

[0045] In one embodiment, step S300, i.e., the step of identifying the user's intention according to the input instruction and generating a target instruction corresponding to the identified intention, further includes: Step S310': when the input instruction is clear, determine the input instruction as the target instruction.

[0046] When the user's input instruction is a clear instruction, for example, the user's input instruction is "query my courseware", the input instruction is a complete instruction with a clear intention. At this time, there is no need to expand the input instruction, and the input instruction can be directly determined as the target instruction.

[0047] In one embodiment, before step S500, that is, the step of identifying and splitting the target instruction based on the classifier, the instruction processing method based on the training system provided in this embodiment further includes: The classifier is pre-trained, and the pre-training includes: constructing instruction sets for several functional areas in the training system, and dynamically generating a large number of instruction processing examples for each instruction in each instruction set in combination with an intelligent model and a classifier.

[0048] Specifically, the instruction set can contain information such as instruction name, function description, target URL, required parameters and calling examples to ensure the comprehensiveness and accuracy of the instructions. Special emphasis is placed on comprehensive coverage of professional terms. For example, key words unique to the training field such as "micro-course", "project", "courseware", and "learning hours" are all included in the instruction set. After the instruction set is built, a large number of instruction processing examples can be dynamically generated for each instruction in each instruction set in combination with AI intelligent models and classifiers, which can incorporate a large number of professional situation simulations, such as "querying the learning hours information of the micro-course "Python Basics"", "uploading the courseware "Data Analysis Case", etc., to ensure that the intelligent model can not only widely understand general instructions, but also accurately identify and respond to the complex instruction structure unique to the training system, so that when users raise related needs such as "micro-course management", "project progress tracking", and "learning plan", it can provide efficient natural language processing support.

[0049] Among them, several functional areas may include training management, course management, examination management, survey management, practice management, knowledge management, micro-class management, live broadcast management, decoration operation, student search, to-do query, personal task query and personal information query, etc., that is, each of the above core functional areas of the training system can be deeply trained so that the classifier can recognize and process instructions related to the above core functional areas.

[0050] In step S500, the classifier can accurately distinguish whether the target instruction is a regular question exchange or a request containing a specific instruction. If it is a request containing a specific instruction, it can distinguish whether it comes from a management instruction or a query instruction, and then call the corresponding special classifier to perform in-depth analysis. Under this framework, each classifier can be customized according to the source of the instruction, aiming to keenly capture the subtle differences in instruction features.

[0051] In one embodiment, step S700, i.e., the step of extracting each instruction parameter and the corresponding execution interface from the split result, includes: When the initial extraction of each instruction parameter fails, the extraction of each instruction parameter is performed in combination with the preset intelligent model and each instruction processing example.

[0052] After the classifier identifies and splits the target instruction, it can extract the instruction parameters from the split results, and at the same time extract the execution interface corresponding to each instruction parameter. In this step, if the initial extraction fails, it can be extracted with the help of the trained intelligent model and the instruction processing examples. The intelligent model can deeply analyze the semantic connection between user instructions and instruction processing examples, and can not only accurately capture the key information fragments in the instructions, but also identify the implicit elements that are critical to the execution of instructions in a complex language environment.

[0053] In one embodiment, after step S700, i.e., the step of extracting each instruction parameter and the corresponding execution interface from the split result, the instruction processing method based on the training system provided by this embodiment further includes: When the instruction parameter is a time parameter, the time parameter is parsed in combination with a pre-stored intelligent model to convert the time parameter into a date format.

[0054] For example, when the extracted instruction parameters contain time parameters, the time parameters can be parsed in conjunction with the intelligent model. Specifically, time parameters may include descriptions such as "today", "tomorrow", "yesterday", and "last year", which are not machine-readable dates. Therefore, before executing the target instruction, such time parameters can be parsed and converted into a date format so that the machine can execute it directly. For another example, the time parameter may also be an unconventional expression covering a complex time span, such as "the past year". For such an expression, the reasoning and understanding capabilities of the intelligent model can be used to accurately capture the relative relationship between the current date and the time span, and convert it into an interval consisting of clear and calculable dates, thereby ensuring the accuracy of parameter extraction and the executable nature of the instruction.

[0055] In step S900, when the number of target instructions exceeds 1, each target instruction can be executed in sequence. When executing each target instruction, the corresponding API execution interface can be called to achieve the connection between the natural language instruction and the training system. At the same time, after executing the target instruction to obtain the corresponding data, the data can be formatted and beautified, and then output and displayed to the human-computer interaction interface in the form of a card, thereby improving the intuitiveness and fluency of the interaction between the user and the training system.

[0056] Reference Figure 3 In one embodiment, before step S900, i.e., the step of calling each execution interface to execute the target instruction, the instruction processing method based on the training system further includes: Step S910: providing a secondary confirmation interface for the user to receive feedback information from the user on each of the instruction parameters; Step S920: when receiving a confirmation instruction from the user based on the secondary confirmation interface, executing the step of calling each execution interface to execute the target instruction; Step S930: When receiving a correction instruction from the user based on the secondary confirmation interface, correct each of the instruction parameters according to the correction instruction.

[0057] That is, after successfully extracting each instruction parameter and the corresponding execution interface, it can be further verified to ensure the accuracy of the operation. This step is a user interactive confirmation link, that is, providing a secondary confirmation interface for the user, and displaying the parsed instruction parameters in the secondary confirmation interface. The user can confirm whether the parsed instruction parameters are consistent with their own intentions. If they are consistent, they can enter the confirmation instruction so that the system executes the next step. If they are not consistent, they can directly correct the instruction parameters on the secondary confirmation interface, and the system will execute the subsequent steps according to the corrected instruction parameters. Through the above-mentioned secondary confirmation process, the smoothness of human-computer interaction and the convenience of correction are further guaranteed.

[0058] In one embodiment, a training system is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0059] Figure 4 This is a structural diagram of a training system provided in an embodiment of the present application. The training system may be a server, and its internal structure diagram may be as follows: Figure 4 As shown. The training system includes a processor, a memory and a network interface connected through a system bus. Among them, the processor of the training system is used to provide computing and control capabilities. The memory of the training system 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 training system is used to store various types of data involved in the instruction processing method. The network interface of the training system is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, an instruction processing method is implemented.

[0060] Those skilled in the art will understand that Figure 4 The structure shown in the figure is only a block diagram of a part of the structure related to the present application scheme, and does not constitute a limitation on the training system to which the present application scheme is applied. The specific training system may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0061] 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 in the above-mentioned method embodiments are implemented.

[0062] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0063] 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 memory, storage, database or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), tape, floppy disk, flash memory or optical memory, etc. Volatile memory can include random access memory (RAM) or external cache memory. 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).

[0064] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described 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 specification.

[0065] 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 construed as limiting the scope of the patent. It should be noted that, for a person of ordinary skill in the art, several modifications 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 command processing method based on a training system, characterized in that: include: Receive user input instructions; Identifying the user's intention according to the input instruction and generating a target instruction corresponding to the identified intention, including: when the input instruction is unclear, extracting and outputting a plurality of instructions related to the intention of the input instruction from a pre-stored instruction set according to a preset intelligent model; receiving a user's selection instruction for the outputted plurality of instructions; and determining the instruction selected by the user as the target instruction according to the user's selection instruction; Identifying and splitting the target instruction based on a classifier; Extract each instruction parameter and the corresponding execution interface from the split result; Each of the execution interfaces is called to execute the target instruction.

2. The instruction processing method based on the training system according to claim 1, characterized in that: The step of identifying the user's intention according to the input instruction and generating a target instruction corresponding to the identified intention also includes: When the input instruction is clear, the input instruction is determined as the target instruction.

3. The instruction processing method based on the training system according to claim 1, characterized in that: Before the step of identifying and splitting the target instruction based on the classifier, the instruction processing method based on the training system further includes: The classifier is pre-trained, and the pre-training includes: constructing instruction sets for several functional areas in the training system, and dynamically generating a large number of instruction processing examples for each instruction in each instruction set in combination with an intelligent model and a classifier.

4. The instruction processing method based on the training system according to claim 3, characterized in that: The several functional areas include training management, course management, examination management, survey management, practice management, knowledge management, micro-class management, live broadcast management, decoration operation, student search, to-do query, personal task query and personal information query.

5. The instruction processing method based on the training system according to claim 3, characterized in that: The step of extracting each instruction parameter and the corresponding execution interface from the split result includes: When the initial extraction of each instruction parameter fails, the extraction of each instruction parameter is performed in combination with the preset intelligent model and each instruction processing example.

6. The instruction processing method based on the training system according to claim 1, characterized in that: After the step of extracting each instruction parameter and the corresponding execution interface from the split result, the instruction processing method based on the training system further includes: When the instruction parameter is a time parameter, the time parameter is parsed in combination with a pre-stored intelligent model to convert the time parameter into a date format.

7. The instruction processing method based on the training system according to claim 1, characterized in that: Before the step of calling each of the execution interfaces to execute the target instruction, the instruction processing method based on the training system further includes: Providing a secondary confirmation interface for the user to receive feedback information from the user on each of the command parameters; When receiving a confirmation instruction from the user based on the secondary confirmation interface, executing the step of calling each of the execution interfaces to execute the target instruction; When a correction instruction from the user based on the secondary confirmation interface is received, each of the instruction parameters is corrected according to the correction instruction.

8. A training system, comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the instruction processing method based on the training system according to any one of claims 1 to 7 is implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the instruction processing method based on a training system according to any one of claims 1 to 7 is implemented.

Images