Work order generation method and system, electronic equipment and storage medium

Abstract

The invention relates to the technical field of artificial intelligence and data processing, and discloses a work order generation method and system, electronic equipment and a storage medium. The method comprises the steps of obtaining a natural language maintenance instruction input by a user; performing intention recognition and entity information extraction on the natural language maintenance instruction to obtain structured semantic information; performing work order field mapping processing on the structured semantic information to obtain an initial maintenance work order; querying to obtain user archive information based on the user identification information; and generating a target maintenance work order based on the initial maintenance work order and the user file information. Through the method, the information dimension of the user input data extracted by the system can be improved, the generated work order and the identity information of the user can be automatically associated and integrated, and the work order generation efficiency and accuracy are improved.

Description

Technical Field

[0001] This application relates to the fields of artificial intelligence and data processing technology, and in particular to a work order generation method, system, electronic device, and storage medium. Background Technology

[0002] In fields such as property management, equipment maintenance, and customer service, work order systems are core tools for task assignment and process management. However, traditional work order systems are mostly built based on keyword matching or simple template filling, which have obvious problems such as insufficient semantic understanding, limited process optimization, and weak data correlation. This leads to inaccurate information extraction by the work order system, limited overall operating efficiency, and difficulty in automatically, quickly, and accurately generating structured work orders. Summary of the Invention

[0003] The embodiments of this application aim to at least partially address one of the technical problems in the related art. To this end, embodiments of this application propose a work order generation method, system, electronic device, and storage medium.

[0004] Embodiments of this application provide a work order generation method, the method comprising: obtaining a natural language maintenance instruction input by a user; performing intent recognition and entity information extraction on the natural language maintenance instruction to obtain structured semantic information; performing work order field mapping processing on the structured semantic information to obtain an initial maintenance work order; querying user profile information based on user identification information; and generating a target maintenance work order based on the initial maintenance work order and the user profile information.

[0005] In some embodiments, performing intent recognition and entity information extraction on natural language maintenance instructions to obtain structured semantic information includes: in response to the natural language maintenance instructions including voice instructions, performing voice recognition on the voice instructions to obtain a text string; and performing intent recognition and entity information extraction on the text string to obtain structured semantic information.

[0006] In some embodiments, the natural language repair instruction is subjected to intent recognition and entity information extraction to obtain structured semantic information, including: performing intent recognition on the natural language repair instruction to obtain intent recognition result; and extracting entity information from the natural language repair instruction to obtain structured semantic information when the intent recognition result indicates that the user needs to create a repair work order.

[0007] In some embodiments, performing intent recognition and entity information extraction on a text string to obtain structured semantic information includes: extracting entity information from the text string based on a preset instruction to obtain entity information extraction results for a preset entity category, wherein the preset entity category includes at least one of problem location, fault object, fault phenomenon, and urgency level; and obtaining structured semantic information based on the entity information extraction results.

[0008] In some embodiments, the structured semantic information is mapped to work order fields to obtain an initial maintenance work order, including: mapping the structured semantic information to preset work order information based on preset work order mapping rules to obtain an initial maintenance work order, wherein the preset work order information includes at least one of work order type, work order content, and urgency level.

[0009] In some embodiments, natural language maintenance instructions include text instructions and / or voice instructions, and natural language maintenance instructions also include image instructions; performing intent recognition and entity information extraction on natural language maintenance instructions to obtain structured semantic information includes: performing intent recognition and entity information extraction on text instructions and / or voice instructions to obtain text recognition results; performing entity information intent recognition and entity information extraction on image instructions to obtain image recognition results; and performing cross-validation or information complementation on text recognition results and image recognition results to obtain structured semantic information.

[0010] In some embodiments, intent recognition and entity information extraction are performed on the text string to obtain structured semantic information, including: judging the validity information of the text string; if the validity information of the text string is insufficient, converting it into a dialogue mode and outputting a clarifying question to the user; and based on the user's response to the clarifying question, performing intent recognition and entity information extraction to obtain structured semantic information.

[0011] Embodiments of this application provide a work order generation system, comprising: an acquisition module for acquiring natural language repair instructions input by a user; an extraction module for performing intent recognition and entity information extraction on the natural language repair instructions to obtain structured semantic information; a processing module for performing work order field mapping processing on the structured semantic information to obtain an initial repair work order; an identification module for querying user profile information based on user identification information; and a generation module for generating a target repair work order based on the initial repair work order and the user profile information.

[0012] Embodiments of this application provide an electronic device, which includes: a memory, and one or more processors communicatively connected to the memory; the memory stores instructions executable by the one or more processors, which are executed by the one or more processors to cause the one or more processors to implement the steps of the method of any of the above embodiments.

[0013] Embodiments of this application provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method of any of the above embodiments.

[0014] Embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the steps of the method according to any of the above embodiments.

[0015] In the embodiments of this application, unstructured natural language repair instructions input by the user can be transformed into standardized target repair work orders. First, the system can perform intent recognition and entity information extraction on the acquired natural language repair instructions input by the user, transforming them into machine-understandable structured semantic information. Then, through predefined work order field mapping rules, this structured semantic information is converted into a standardized initial repair work order. Simultaneously, the system can query user profile information in parallel based on user identification information while processing the natural language repair instructions. Finally, the system automatically merges the initial work order with the user profile information to generate a complete, accurate, and standardized target repair work order. The method provided in this application can effectively improve the efficiency and automation of work order generation, effectively avoid the cumbersome interaction mode between the user and the system in traditional technologies, and also avoid errors and information omissions that may occur when manually entering work orders, ensuring the accuracy and completeness of the work order information. Attached Figure Description

[0016] Figure 1 A flowchart illustrating the work order generation method provided in this application embodiment; Figure 2a , 2b A flowchart illustrating a work order generation method provided in an embodiment of this application; Figure 3 A schematic diagram illustrating a "dialogue as a service" work order generation mode provided in an embodiment of this application; Figure 4 This is a schematic diagram illustrating the work order field mapping processing of structured semantic information according to an embodiment of this application; Figure 5 A schematic diagram of the work order generation system provided in the embodiments of this application; Figure 6 A block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0017] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0018] In fields such as property management, equipment maintenance, and customer service, work order systems are core tools for task assignment and process management. However, traditional work order systems are mostly built based on keyword matching or simple template filling, which have obvious problems such as insufficient semantic understanding, limited process optimization, and weak data correlation. This leads to inaccurate information extraction by the work order system, limited overall operating efficiency, and difficulty in automatically, quickly, and accurately generating structured electronic work orders.

[0019] First, keyword-matching-based work order submission schemes struggle to truly understand the context and genuine intent of user input. For example, when a user describes, "Our bathroom sink is leaking, please send a technician over quickly to fix it," traditional systems typically only match the keyword "leaking," failing to accurately deduce that "bathroom" is the specific location of the problem, "sink" is the object of the fault, and it's even more difficult to infer the urgency of the work order from interjections like "quickly." This approach extracts information with limited dimensions and poor accuracy, making it difficult to generate accurate electronic work orders.

[0020] Secondly, traditional technologies offer limited benefits in process optimization, failing to fundamentally change the interaction pattern between users and the system. Although some systems have introduced keyword recommendation features, users still face complex form structures, requiring them to manually fill in or confirm most of the information. The entire interaction process still follows the inherent logic of "human-machine-form," failing to achieve a true improvement in efficiency.

[0021] Furthermore, traditional systems suffer from significant shortcomings in data integration, with weak correlations between data points leading to information silos. Traditional systems typically process user identity information and repair requests as separate data. Users usually need to complete identity verification before entering a separate repair request module to fill in the information. The system fails to automatically and intelligently associate and integrate the repair information contained in the user's natural language with their identity information. This information silos not only increase the user's operational burden but also limit the system's responsiveness and intelligence during the service process.

[0022] Therefore, the purpose of this application is to provide a work order generation method to overcome the above-mentioned defects in traditional technologies and realize the automatic, fast and accurate generation of structured electronic work orders through a single natural language interaction.

[0023] Figure 1 This is a flowchart illustrating the work order generation method provided in an embodiment of this application.

[0024] like Figure 1 As shown, the work order generation method 100 provided in this application embodiment includes steps S110-S140.

[0025] Step S110: Obtain the natural language maintenance instructions input by the user.

[0026] For example, a user can input a natural language repair command describing their warranty needs into the system through an input interface on their user terminal. For instance, the user terminal could be a smartphone application, a mini-program, or a web application. The natural language repair command could be input information in various formats to express the user's repair intentions, such as text or voice.

[0027] Step S120: Perform intent recognition and entity information extraction on the natural language maintenance instructions to obtain structured semantic information.

[0028] For example, natural language repair instructions are typically unstructured text or voice commands used by users in their daily lives, which cannot be directly understood and invoked by machines. Therefore, in order to create work orders efficiently and accurately, it is necessary to first perform intent recognition and entity information extraction on the natural language repair instructions.

[0029] Step S130: Perform work order field mapping processing on the structured semantic information to obtain the initial maintenance work order.

[0030] For example, the system's work order field mapping rules can be set in advance, and then the obtained structured semantic information can be mapped according to the pre-set work order field mapping rules to obtain standardized and normalized initial maintenance work orders.

[0031] Step S140: Based on the user identification information, retrieve the user profile information.

[0032] For example, while processing the natural language maintenance instructions input by the user, the system can also initiate a query request to the user information database (such as the customer profile in the main data center) based on the current user's login status or unique identity (User ID) to obtain the user's fixed basic information, such as: community name, building number, room number, default contact number, etc.

[0033] Step S150: Generate the target repair work order based on the initial repair work order and user profile information.

[0034] For example, the system's service module can merge the initial maintenance work order mapped in step S130 with the user profile information queried in step S140 to generate a target maintenance work order with complete and structured information.

[0035] In the embodiments of this application, unstructured natural language maintenance instructions input by the user can be converted into standardized target maintenance work orders.

[0036] First, the system can perform intent recognition and entity information extraction on the natural language repair instructions input by the user, transforming them into machine-understandable structured semantic information. Then, through predefined work order field mapping rules, this structured semantic information is converted into a standardized initial repair work order.

[0037] Simultaneously, the system can query user profile information based on user identification information while processing natural language repair instructions. Finally, the system automatically merges the initial work order with the user profile information to generate a complete, accurate, and standardized target repair work order.

[0038] The work order generation method provided in this application can effectively improve the efficiency and automation of work order generation, effectively avoid the cumbersome interaction mode between users and the system in traditional technologies, and also avoid errors and information loss that may occur when manually entering work orders, thus ensuring the accuracy and completeness of work order information.

[0039] Figure 2a , 2b This is a flowchart illustrating a work order generation method provided in an embodiment of this application.

[0040] Figure 3 This is a schematic diagram of a "dialogue as a service" work order generation mode provided in an embodiment of this application.

[0041] In one embodiment, the process of performing intent recognition and entity information extraction on natural language maintenance instructions to obtain structured semantic information includes: in response to the natural language maintenance instructions including voice instructions, performing voice recognition on the voice instructions to obtain a text string; and performing intent recognition and entity information extraction on the text string to obtain structured semantic information.

[0042] For example, such as Figure 2a , 2b , Figure 3As shown, when the system recognizes that the natural language maintenance command input by the user includes a voice command, the system can first use the Automatic Speech Recognition (ASR) module to perform speech recognition on the real-time voice command input by the user, converting the voice command into a text string. Then, the text string is further processed for intent recognition and entity information extraction to obtain structured semantic information.

[0043] like Figure 3 As shown, the embodiments provided in this application propose a "dialogue-as-a-service" work order generation mode. Users only need to complete tasks that previously required multiple, discrete form filling operations through a single, continuous natural language input, fundamentally simplifying the human-computer interaction process.

[0044] In this embodiment, the solution provided by this application can simultaneously support different types of data input, such as voice and text, significantly reducing the user's learning curve and improving ease of use. Furthermore, by unifying the data format of user-inputted natural language repair commands to text format, the system only needs to maintain one set of text processing procedures to handle both text and voice input methods simultaneously, greatly simplifying the system architecture and reducing maintenance costs. Simultaneously, it can also provide a standardized and efficient processing flow for subsequent work order generation.

[0045] In one embodiment, the natural language repair instruction is subjected to intent recognition and entity information extraction to obtain structured semantic information, including: performing intent recognition on the natural language repair instruction to obtain intent recognition result; and extracting entity information from the natural language repair instruction to obtain structured semantic information when the intent recognition result indicates that the user needs to create a repair work order.

[0046] For example, such as Figure 2b As shown, when the application service module of the server obtains a text string, it can send the text string and preset instructions to the Large Language Model (LLM) for processing. This LLM is pre-trained on massive amounts of text data and has powerful semantic understanding, contextual reasoning, and information extraction capabilities.

[0047] By calling upon large AI models for natural language processing, it can perform intent recognition and entity information extraction based on preset input instructions and text strings.

[0048] Specifically, during the intent recognition process, the AI ​​model determines the user's core purpose based on the text string. For example, for the message "Our bathroom sink is leaking," the model identifies the user's intent as "to create an engineering repair work order."

[0049] When the intent recognition result indicates that the user needs to create a repair work order, entity information is extracted from the natural language repair instruction to obtain structured semantic information. In one embodiment, performing intent recognition and entity information extraction on the text string to obtain structured semantic information further includes: extracting entity information from the text string based on a preset instruction to obtain entity information extraction results for preset entity categories, wherein the preset entity categories include at least one of problem location, fault object, fault phenomenon, and urgency level; and obtaining structured semantic information based on the entity information extraction results.

[0050] During entity information extraction, the AI ​​model can extract key information fragments of predefined entity categories from text strings. These predefined entity categories include at least the problem location, the faulty object, the fault phenomenon, and the urgency level. For example, the problem location could be "bathroom," "kitchen," or "balcony"; the faulty object could be "washbasin," "toilet," or "light bulb"; the fault phenomenon could be "leaking," "blocked," or "not working"; and the urgency level could be inferred from adverbs or tone such as "quickly," "immediately," or "as soon as possible."

[0051] In the embodiments provided in this application, the efficiency and intelligence level of work order generation are significantly improved by introducing a large AI model for natural language processing. The system performs intent recognition on natural language repair instructions. If the intent recognition result indicates that the user needs to create a repair work order, entity information is extracted from the natural language repair instructions. This allows the system to accurately determine whether the user's core need is to create a work order. This judgment mechanism can effectively optimize the allocation of computational resources.

[0052] In the process of extracting entity information from text strings, this embodiment can utilize the deep semantic understanding capabilities of AI big data models. Based on preset instructions and according to preset entity categories, the AI ​​big data models can accurately extract multi-dimensional and interrelated key business information (such as location, object, phenomenon, urgency) from unstructured spoken descriptions, thereby improving the accuracy and multidimensionality of information extraction, far exceeding the implementation schemes that rely on keyword matching in traditional technologies.

[0053] In one embodiment, the natural language maintenance instruction includes text instructions and / or voice instructions, and further includes image instructions; the natural language maintenance instruction is subjected to intent recognition and entity information extraction to obtain structured semantic information, including: performing intent recognition and entity information extraction on the text instructions and / or voice instructions to obtain text recognition results; performing entity information intent recognition and entity information extraction on the image instructions to obtain image recognition results; and performing cross-validation or information complementation on the text recognition results and image recognition results to obtain structured semantic information.

[0054] For example, such as Figure 2b As shown, in addition to inputting text and voice commands, users can also upload images. The system can integrate a large multimodal AI model or a standalone computer vision (CV) model. This model can analyze image content, automatically identify faulty objects (such as "toilet") and fault phenomena (such as "water stains on the floor"), and cross-validate or complement the image recognition results with the text analysis results to further improve the accuracy and richness of work order information.

[0055] Figure 4 This is a schematic diagram illustrating the work order field mapping processing of structured semantic information according to an embodiment of this application.

[0056] In one embodiment, the structured semantic information is mapped to work order fields to obtain an initial maintenance work order, which includes: mapping the structured semantic information to preset work order information based on preset work order mapping rules to obtain an initial maintenance work order, wherein the preset work order information includes at least one of work order type, work order content, and urgency level.

[0057] For example, such as Figure 2b , Figure 4 As shown, the entity information extracted by the AI ​​model can be returned to the application service module in a structured data format, such as JSON (JavaScript Object Notation). The application service module then fills this structured semantic information into the corresponding fields of the work order according to preset work order mapping rules, thus obtaining the initial maintenance work order.

[0058] For example, the text input by the user or the content after voice-to-text conversion is: "The sink in our bathroom is leaking. Please send a repairman over to fix it as soon as possible." After the user terminal receives the text data, it sends the text data to the server. After the server's application module receives the text data, it calls the AI ​​big model to perform natural language processing on the input text data, including text string encoding of the input text data, and performing intent recognition and entity information extraction based on preset instructions and text strings.

[0059] Then, the AI ​​large model with Figure 4The output is in JSON format, where intent recognition yields an intent field, entities, and several category branches for the entities: problem location (e.g., bathroom), fault object (e.g., washbasin), fault phenomenon (e.g., leak), and urgency level (e.g., high). Finally, according to preset work order mapping rules, this structured semantic information is filled into the corresponding fields of the work order to obtain the initial repair work order.

[0060] For example, during the natural language processing of the input text data using the AI ​​model, the application service module initiates a query request to the user information database based on the current user's login status or unique identifier (User ID). This retrieves the user's basic information, such as: community name, building number, room number, and default contact number. The application service module then merges the parsed initial repair work order with the retrieved basic user information to generate a new work order. Figure 3 The information shown is a complete and structured target maintenance work order.

[0061] In another example, the system can also choose to allow the user to confirm whether to generate and submit a work order. Specifically, the system can choose to return a summary of the generated work order (e.g., "Do you want to report a 'leaky bathroom sink' for 'Building X, Unit X, Room XXX, XX Community'? Urgency: Urgent") to the user's terminal for final confirmation. After user confirmation, the work order is officially created and stored in the work order database, entering the subsequent dispatch process. This step can also be configured to be confirmation-free to further improve efficiency.

[0062] Traditional work order systems require users to manually select or input each of the preset form fields (such as community, building, problem category, urgency level, etc.), which is cumbersome, time-consuming, and reduces the efficiency of work order generation.

[0063] In the embodiments of this application, by leveraging the deep semantic understanding capabilities of the AI ​​big data model, multi-dimensional and interconnected structured semantic information (such as location, object, phenomenon, and urgency) can be accurately extracted from unstructured spoken descriptions. Its accuracy and information dimensionality far surpass traditional keyword matching techniques. Furthermore, the embodiments of this application can automatically associate and integrate the dynamic repair information parsed by the AI ​​big data model with static user information in the background database, generating complete structured data in one step, thus achieving a high degree of intelligence and automation in the data processing workflow.

[0064] The AI ​​model used in the proposed solution can be either a general-purpose public cloud service model or a dedicated model fine-tuned on a corpus for a specific domain (such as property maintenance) to improve recognition accuracy and processing efficiency in specific scenarios. This model can be deployed in the cloud or privately.

[0065] In one embodiment, intent recognition and entity information extraction are performed on the text string to obtain structured semantic information, including: judging the validity information of the text string; if the validity information of the text string is insufficient, converting it into a dialogue mode and outputting a clarifying question to the user; and based on the user's response to the clarifying question, performing intent recognition and entity information extraction to obtain structured semantic information.

[0066] For example, when the information input by the user is vague or insufficient (e.g., the user only says "it's broken"), the AI ​​model in this embodiment may not directly generate a work order, but instead switch to a dialogue mode and proactively ask the user clarifying questions (e.g., "What exactly is broken?" or "Can you tell me which room it's in?"). After one or more rounds of dialogue, the user is guided to provide sufficient information before the work order generation step is executed.

[0067] In traditional solutions, differences in user understanding and expression often lead to inaccurate or incomplete information when describing problems. This increases the workload for back-end staff in verifying and supplementing information, impacting subsequent order dispatch and processing efficiency. Furthermore, traditional solutions lack the ability to understand users' natural language intent, failing to automatically extract key information from users' colloquial and unstructured descriptions. This results in a rigid system that is not user-friendly, especially for users unfamiliar with complex operations.

[0068] In view of this, this application introduces a dialogue mechanism. When a user inputs insufficiently valid text such as "broken," the system can proactively ask the user clarifying questions. By initiating a multi-turn dialogue mode, the system guides the user to supplement key information by asking targeted clarifying questions. This overcomes the limitations of traditional technologies that cannot process or generate erroneous work orders due to incomplete information, reduces the error rate of work order information extraction caused by incomplete or ambiguous information, improves the reliability of automated processes, and ensures the efficiency and accuracy of work order generation.

[0069] Figure 5 This is a schematic diagram of a work order generation system provided in an embodiment of this application.

[0070] Embodiments of this application provide a work order generation system 500, the system 500 comprising: The acquisition module 510 is used to acquire natural language maintenance instructions input by the user.

[0071] The extraction module 520 is used to perform intent recognition and entity information extraction on natural language maintenance instructions to obtain structured semantic information.

[0072] The processing module 530 is used to perform work order field mapping processing on the structured semantic information to obtain the initial maintenance work order.

[0073] The identification module 540 is used to retrieve user profile information based on user identification information.

[0074] The generation module 550 is used to generate a target maintenance work order based on the initial maintenance work order and user profile information.

[0075] In some embodiments, the extraction module 520 is further configured to: in response to a natural language maintenance instruction including a voice instruction, perform speech recognition on the voice instruction to obtain a text string; and perform intent recognition and entity information extraction on the text string to obtain structured semantic information.

[0076] In some embodiments, the extraction module 520 is further configured to: perform intent recognition on the natural language maintenance instruction to obtain intent recognition results; and, if the intent recognition results indicate that the user needs to create a maintenance work order, extract entity information from the natural language maintenance instruction to obtain structured semantic information.

[0077] In some embodiments, the extraction module 520 is further configured to: extract entity information from the text string based on a preset instruction to obtain entity information extraction results for a preset entity category, wherein the preset entity category includes at least one of problem location, fault object, fault phenomenon, and urgency level; and obtain structured semantic information based on the entity information extraction results.

[0078] In some embodiments, the processing module 530 is further configured to: map structured semantic information to preset work order information based on preset work order mapping rules to obtain an initial maintenance work order, wherein the preset work order information includes at least one of work order type, work order content, and urgency level.

[0079] In some embodiments, the extraction module 520 is further configured to: perform intent recognition and entity information extraction on text instructions and / or voice instructions to obtain text recognition results; perform entity information intent recognition and entity information extraction on image instructions to obtain image recognition results; and perform cross-validation or information complementation on text recognition results and image recognition results to obtain structured semantic information.

[0080] In some embodiments, the extraction module 520 is further configured to: determine the validity information of the text string; if the validity information of the text string is insufficient, convert it to a dialogue mode and output a clarifying question to the user; and based on the user's response to the clarifying question, perform intent recognition and entity information extraction to obtain structured semantic information.

[0081] It is understandable that for a detailed description of the work order generation system 500, please refer to the description of the work order generation method 100 above.

[0082] This application provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the method in any of the above embodiments.

[0083] This application provides a computer program product that includes instructions that, when executed by a processor of a computer device, enable the computer device to perform the steps of the method described in any of the above embodiments.

[0084] Figure 6 A block diagram of an electronic device provided in an embodiment of this application.

[0085] This application provides an electronic device, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method in any of the above embodiments.

[0086] like Figure 6 As shown, for ease of understanding, an embodiment of this application illustrates a specific electronic device 600.

[0087] Electronic device 600 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic device 600 may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0088] like Figure 6As shown, device 600 includes a computing unit 601, which can perform various appropriate actions and processes based on a computer program stored in read-only memory (ROM) 602 or a computer program loaded from storage unit 608 into random access memory (RAM) 603. RAM 603 may also store various programs and data required for the operation of electronic device 600. The computing unit 601, ROM 602, and RAM 603 are interconnected via bus 604. Input / output (I / O) interface 605 is also connected to bus 604.

[0089] Multiple components in electronic device 600 are connected to I / O interface 605. These components include: input unit 606, such as a keyboard or mouse; output unit 607, such as various types of displays or speakers; storage unit 608, such as a disk or optical disk; and communication unit 609, such as a network interface card (NIC), modem, or wireless transceiver. Communication unit 609 allows electronic device 600 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0090] The computing unit 601 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 601 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 601 performs the various methods described above. For example, in some embodiments, any one or more of the methods described above can be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 608. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 600 via ROM 602 and / or communication unit 609. When the computer program is loaded into RAM 603 and executed by the computing unit 601, one or more steps of any one or more of the methods described above can be performed. Alternatively, in other embodiments, the computing unit 601 can be configured to perform any one or more of the methods described above by any other suitable means (e.g., by means of firmware).

[0091] It should be noted that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this application, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which programs can be printed, because programs can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0092] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0093] In the description of this application, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this application, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0094] In the description of this application, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", "axial", "radial", "circumferential", etc., indicating the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application.

[0095] Furthermore, the terms "first," "second," etc., used in the embodiments of this application are for descriptive purposes only and should not be construed as indicating or implying relative importance, or implicitly specifying the number of technical features indicated in this embodiment. Therefore, features defined with terms such as "first" and "second" in the embodiments of this application can explicitly or implicitly indicate that the embodiment includes at least one of those features. In the description of this application, the word "multiple" means at least two or more, such as two, three, four, etc., unless otherwise explicitly and specifically defined in the embodiments.

[0096] In this application, unless otherwise explicitly specified or limited in the embodiments, the terms "installation," "connection," "joining," and "fixing" appearing in the embodiments should be interpreted broadly. For example, a connection can be a fixed connection, a detachable connection, or an integral part; it can also be a mechanical connection, an electrical connection, etc. Of course, it can also be a direct connection, or an indirect connection through an intermediate medium, or it can be the internal communication between two components, or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific implementation.

[0097] In this application, unless otherwise expressly specified and limited, "above" or "below" the second feature can mean that the first feature is in direct contact with the second feature, or that the first feature is in indirect contact with the second feature through an intermediate medium. Furthermore, "above," "on top of," and "over" the second feature can mean that the first feature is directly above or diagonally above the second feature, or simply that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature can mean that the first feature is directly below or diagonally below the second feature, or simply that the first feature is at a lower horizontal level than the second feature.

Claims

1. A method for generating work orders, characterized in that, The method includes: Obtain natural language repair instructions input by the user; The natural language maintenance instructions are subjected to intent recognition and entity information extraction to obtain structured semantic information; The structured semantic information is mapped to work order fields to obtain an initial maintenance work order; Based on user identification information, user profile information can be retrieved. Based on the initial repair work order and the user profile information, a target repair work order is generated.

2. The method according to claim 1, characterized in that, The process of performing intent recognition and entity information extraction on the natural language maintenance instructions to obtain structured semantic information includes: In response to the natural language maintenance instruction including a voice instruction, the voice instruction is subjected to speech recognition to obtain a text string; The text string is subjected to intent recognition and entity information extraction to obtain the structured semantic information.

3. The method according to claim 1 or 2, characterized in that, The process of performing intent recognition and entity information extraction on the natural language maintenance instructions to obtain structured semantic information includes: The natural language maintenance command is subjected to intent recognition to obtain the intent recognition result; When the intent recognition result indicates that the user needs to create a repair work order, entity information is extracted from the natural language repair instruction to obtain the structured semantic information.

4. The method according to claim 2, characterized in that, The process of performing intent recognition and entity information extraction on the text string to obtain the structured semantic information includes: Based on preset instructions, entity information is extracted from the text string to obtain entity information extraction results of preset entity categories, wherein the preset entity categories include at least one of problem location, fault object, fault phenomenon, and urgency level; Based on the entity information extraction results, the structured semantic information is obtained.

5. The method according to claim 1, characterized in that, The process of mapping the structured semantic information to work order fields to obtain the initial maintenance work order includes: Based on preset work order mapping rules, the structured semantic information is mapped to preset work order information to obtain the initial maintenance work order, wherein the preset work order information includes at least one of work order type, work order content, and urgency level.

6. The method according to claim 1, characterized in that, The natural language maintenance instructions include text instructions and / or voice instructions, and also include image instructions; the process of performing intent recognition and entity information extraction on the natural language maintenance instructions to obtain structured semantic information includes: The text instructions and / or voice instructions are subjected to intent recognition and entity information extraction to obtain text recognition results; The image command is subjected to entity information intent recognition and entity information extraction to obtain the image recognition result; The text recognition results and the image recognition results are cross-validated or complementary to obtain the structured semantic information.

7. The method according to claim 2, characterized in that, The process of performing intent recognition and entity information extraction on the text string to obtain the structured semantic information includes: The validity information of the text string is judged. If the validity information of the text string is insufficient, the process is switched to dialogue mode and a clarification question is output to the user. Based on the user's response to the clarification question, intent recognition and entity information extraction are performed to obtain the structured semantic information.

8. A work order generation system, characterized in that, The system includes: The acquisition module is used to acquire natural language maintenance instructions input by the user; The extraction module is used to perform intent recognition and entity information extraction on the natural language maintenance instructions to obtain structured semantic information; The processing module is used to perform work order field mapping processing on the structured semantic information to obtain an initial maintenance work order; The identification module is used to retrieve user profile information based on user identification information; The generation module is used to generate a target maintenance work order based on the initial maintenance work order and the user profile information.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1-7.

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

Images