Work order generation method and device, program product and storage medium

Abstract

The invention discloses a work order generation method and device, a program product and a storage medium, and the method comprises the steps: generating initial session analysis information based on a session prompt word and a session between a first object and a second object, the session prompt word being used for indicating a rule for generating the initial session analysis information; generating a work order cue word based on the initial session analysis information and work order demand information, the work order demand information being used for indicating a demand of a second object for generating a work order, and the work order cue word being used for indicating a rule for generating the work order; and generating a target work order based on the work order cue word and the initial session analysis information, the target work order being used for recording a session between the first object and the second object. Through the method and the device, the problem that the work order cannot be dynamically generated in related technologies is solved, and the effect of dynamically generating the work order is further achieved.

Description

Technical Field

[0001] This application relates to the field of computer technology, and more specifically, to a method and apparatus for generating work orders, a program product, and a storage medium. Background Technology

[0002] In customer service technology, the interaction between customers and customer service representatives mainly relies on human agents and robots based on discriminative intent recognition models. Currently, most of the interaction conversations are recorded manually or natural language processing technology is used to extract key information from the conversation content. Finally, the manual agent edits the corresponding work order based on the extracted key information so that other customer service can be provided or a structured customer service work analysis report can be generated.

[0003] Although the aforementioned technologies have achieved automated recording of session content and standardized work order formats to some extent, they still ultimately rely on manual generation of work orders and lack the ability to adjust the work order format according to real-time changes in session content, resulting in a technical problem of not being able to dynamically generate work orders. Summary of the Invention

[0004] This application provides a method and apparatus for generating work orders, a program product, and a storage medium, to at least solve the problem of the inability to dynamically generate work orders in related technologies.

[0005] According to one aspect of the embodiments of this application, a method for generating a work order is provided, comprising: generating initial session analysis information based on a session prompt word and a session between a first object and a second object, wherein the session prompt word is used to indicate the rules for generating the initial session analysis information; generating a work order prompt word based on the initial session analysis information and work order request information, wherein the work order request information is used to indicate the second object's request for generating a work order, and the work order prompt word is used to indicate the rules for generating a work order; and generating a target work order based on the work order prompt word and the initial session analysis information, wherein the target work order is used to record the session between the first object and the second object.

[0006] In an exemplary embodiment, generating initial session analysis information based on a session prompt and a session between a first object and a second object includes: extracting first session information of the first object and second session information of the second object from the session, wherein the first session information includes request information and response information sent by the first object to the second object, and the second session information includes feedback information sent by the second object to the first object; and generating the initial session analysis information based on the session prompt, the first session information, and the second session information.

[0007] In an exemplary embodiment, generating the initial conversation analysis information based on the conversation prompt words, the first conversation information, and the second conversation information includes: extracting a first feature from the first conversation information based on the conversation prompt words, wherein the first feature includes at least one of the following: a first demand feature issued by the first object to the second object, a first question feature issued by the first object to the second object, a first identity feature of the first object, a first result feature issued by the first object to the second object, and an emotional feature of the first object; extracting a second feature from the second conversation information based on the conversation prompt words, wherein the second feature includes: a first strategy feature for resolving the first problem fed back by the second object to the first object; and generating the initial conversation analysis information based on the first feature, the second feature, the first conversation information, and the second conversation information.

[0008] In one exemplary embodiment, generating a work order prompt based on the initial session analysis information and work order requirement information includes: determining a work order type from the work order requirement information and generating a first work order prompt based on the work order type, wherein the first work order prompt is used to represent the format of the generated work order template; generating a second work order prompt based on the first work order prompt and the initial session analysis information, wherein the second work order prompt is used to represent the work order content that meets the work order type determined from the initial session analysis information; and determining the first work order prompt and the second work order prompt as the work order prompt.

[0009] In an exemplary embodiment, generating a second work order prompt based on the first work order prompt and the initial session analysis information includes: determining multiple target work order fields included in the target work order based on the first work order prompt; determining attribute information of the multiple target work order fields based on the initial session analysis information to obtain multiple basic work order prompts; and determining the second work order prompt from the multiple basic work order prompts, wherein the second work order prompt is a prompt that satisfies a preset rule among the multiple basic work order prompts.

[0010] In an exemplary embodiment, generating a target work order based on the aforementioned work order prompt and the aforementioned initial session analysis information includes: modifying a preset initial work order template based on the aforementioned first work order prompt to generate a target work order template, wherein the format of the target work order template satisfies the format of the aforementioned work order template; generating target work order content based on the aforementioned second work order prompt and the aforementioned initial session analysis information; and filling the target work order content into the aforementioned target work order template to obtain the aforementioned target work order.

[0011] In an exemplary embodiment, generating target work order content based on the second work order prompt word and the initial session analysis information includes: determining session analysis information matching the second work order prompt word from the initial session analysis information to obtain the target work order content.

[0012] According to another aspect of the embodiments of this application, a work order generation apparatus is also provided, comprising: a first generation module, configured to generate initial session analysis information based on a session prompt word and a session between a first object and a second object, wherein the session prompt word is used to indicate the rules for generating the initial session analysis information; a second generation module, configured to generate a work order prompt word based on the initial session analysis information and work order requirement information, wherein the work order requirement information is used to indicate the second object's requirement to generate a work order, and the work order prompt word is used to indicate the rules for generating a work order; and a third generation module, configured to generate a target work order based on the work order prompt word and the initial session analysis information, wherein the target work order is used to record the session between the first object and the second object.

[0013] According to another aspect of the embodiments of this application, a computer-readable storage medium is also provided, wherein a computer program is stored therein, wherein the computer program is configured to perform the steps in any of the above method embodiments when executed by a processor.

[0014] According to another aspect of the embodiments of this application, a computer program product or computer program is provided, the computer program product or computer program including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, causing the computer device to perform the steps in any of the method embodiments described above.

[0015] According to another aspect of the embodiments of this application, an electronic device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to perform the steps of any of the above method embodiments through the computer program.

[0016] This application utilizes conversation prompts to guide the analysis of the conversation between the first and second parties, generating initial conversation analysis information. This ensures that the generated initial conversation analysis information is more accurate and better meets user needs. Furthermore, based on the initial conversation analysis information and work order requirement information, work order prompts are dynamically generated. Finally, based on the work order prompts and initial conversation analysis information, a target work order is generated. This makes the work order no longer a static template, but rather allows for dynamic adjustment of the work order format and content requirements based on the current conversation analysis information and work order requirement information, using work order prompts. Therefore, it solves the problem of not being able to dynamically generate work orders in related technologies, achieving the effect of dynamically generating work orders. Attached Figure Description

[0017] Figure 1 This is a schematic diagram illustrating an application scenario of a work order generation method according to an embodiment of this application;

[0018] Figure 2 This is a flowchart illustrating an optional work order generation method according to an embodiment of this application.

[0019] Figure 3 This is a schematic diagram of an optional process for generating initial session analysis information according to an embodiment of this application;

[0020] Figure 4 This is a schematic diagram of an optional process for generating a target work order according to an embodiment of this application;

[0021] Figure 5 This is a schematic diagram illustrating an optional method for generating a target work order according to an embodiment of this application;

[0022] Figure 6 This is a flowchart illustrating another optional method for generating work orders according to an embodiment of this application. Figure 1 ;

[0023] Figure 7 This is a flowchart illustrating another optional method for generating work orders according to an embodiment of this application. Figure 2 ;

[0024] Figure 8 This is a flowchart illustrating another optional method for generating work orders according to an embodiment of this application. Figure 3 ;

[0025] Figure 9 This is a structural block diagram of an optional work order generation device according to an embodiment of this application. Detailed Implementation

[0026] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0027] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0028] According to one aspect of the embodiments of this application, a method for generating work orders is provided. Optionally, in this embodiment, the above-described work order generation method may be applied, but is not limited to, to applications such as... Figure 1 The hardware environment shown includes terminal device 102 and server 104. Server 104 can be connected to terminal device 102 via a network and can be used to provide services (e.g., application services, etc.) to terminal device 102 or clients installed on terminal device 102. A database can be set up on server 104 or independently of server 104 to provide data storage services for server 104.

[0029] The aforementioned network may include, but is not limited to, at least one of the following: wired network and wireless network. The aforementioned wired network may include, but is not limited to, at least one of the following: wide area network (WAN), metropolitan area network (MAN), and local area network (LAN). The aforementioned wireless network may include, but is not limited to, at least one of the following: Wireless Fidelity (WIFI) and Bluetooth. Terminal device 102 may be, but is not limited to, a personal computer (PC), mobile phone, tablet computer, etc. Server 104 may be, but is not limited to, a cloud server, server cluster, or other server types.

[0030] The work order generation method of this application embodiment can be executed by server 104, by terminal device 102, or by both server 104 and terminal device 102. Alternatively, the work order generation method of this application embodiment can be executed by a client installed on the terminal device 102.

[0031] Taking the work order generation method in this embodiment executed by terminal device 102 as an example, Figure 2 This is a flowchart illustrating an optional work order generation method according to an embodiment of this application, such as... Figure 2 As shown, the process of this method may include the following steps:

[0032] Step S202: Based on the conversation prompt and the conversation between the first object and the second object, generate initial conversation analysis information, wherein the conversation prompt is used to indicate the rules for generating the initial conversation analysis information.

[0033] Optionally, the first object can be a customer, and the second object can be a human customer service representative or a machine customer service representative. Conversation prompts are a set of predefined instructions or rules based on the conversation scenario, used to guide the large model in effectively extracting and analyzing information when processing conversations between the first object (e.g., a customer) and the second object (e.g., a customer service representative). Conversation prompts include, but are not limited to, the output format rules for initial conversation analysis information and the generation requirements for initial conversation analysis information. For example, in a customer service consultation conversation scenario, conversation prompts are used to indicate that the generated initial conversation analysis information should include: the type of customer's problem, the solution provided by the customer service representative, and the customer's immediate reaction to the solution; in a debt collection outbound call conversation scenario, conversation prompts are used to indicate that the generated initial conversation analysis information should include: the customer's overdue details, the customer's attitude and willingness to repay, and the customer service representative's collection script.

[0034] Optionally, the initial conversation analysis information is a conversation summary output by the large model based on conversation prompts. The initial conversation analysis information includes, but is not limited to, the customer's specific questions, emotional state, service requests, customer service response strategies, and any key conversation information related to business.

[0035] Optionally, such as Figure 3 As shown, Figure 3 This is a schematic diagram of an optional process for generating initial session analysis information according to an embodiment of this application, including the following steps:

[0036] Step S302, Begin.

[0037] Step S304: The call is connected to a live agent.

[0038] Step S306: The human agent enters the summary entry page through the customer relationship management system and clicks the smart summary button.

[0039] Step S308: The customer relationship management system calls the application programming interface (API) of the smart work order.

[0040] Step S310: Use the API of the smart work order to obtain the audio text of this session in the Sales as a Service system.

[0041] Step S312: Encapsulate the session prompt words using the API of the smart work order.

[0042] Step S314: The Sales as a Service system calls the large model.

[0043] In step S316, the Sales-as-a-Service system generates an intelligent summary (i.e., initial conversation analysis information) based on conversation prompts using a large model and stores the intelligent summary.

[0044] Step S318: Render the page in the customer relationship management system and display the smart summary to the human agent.

[0045] Step S320: The human agent can modify the content in the smart summary in the customer relationship management system to obtain the target smart summary, and then click submit.

[0046] Step S322: The Sales-as-a-Service system obtains the target intelligent summary submitted by the human agent.

[0047] Step S324: Compare the smart summary and the target smart summary, generate a comparison record, and modify the conversation prompt words based on the comparison record.

[0048] Step S326, End.

[0049] Step S204: Based on the initial session analysis information and work order request information, generate a work order prompt, wherein the work order request information is used to indicate the second object's request to generate a work order, and the work order prompt is used to indicate the rules for generating a work order.

[0050] Optionally, the work order requirement information is used to indicate the work order fields included in the generated work order. For example, in a scenario where a customer inquires about the loan application process, a customer service representative may need to create a work order to record detailed questions and forward them to a professional team for further explanation or modification of contract terms. The work order requirement information may include customer information, a specific description of the customer's needs (such as the terms that need to be explained, the repayment date that needs to be adjusted), etc.

[0051] Optionally, work order prompts are instructions or rules used by the large model when generating work orders, guiding the specific format and content of the work order. Specifically, work order prompts can be rules for filling in work order fields, definitions of work order types, etc. For example, in a scenario where a customer inquires about a loan application process, the work order prompts could include: a work order type prompt: "Loan Inquiry Work Order," indicating that the work order is a record about a loan inquiry; a field filling prompt: "Must include customer identification and inquiry details," indicating that the customer's identification information and the specific content of the inquiry must be included when generating the work order; and a priority prompt: "Mark as Urgent," indicating that the priority of the work order should be set to urgent when generating the work order to quickly respond to the customer's needs.

[0052] Step S206: Based on the above-mentioned work order prompt words and the above-mentioned initial session analysis information, a target work order is generated, wherein the target work order is used to record the session between the first object and the second object.

[0053] Optionally, in addition to recording the session between the first object and the second object, the target work order can also record the follow-up plan for the first object. The follow-up plan is generated based on the demand information of the first object in the initial session analysis information. The follow-up plan is used to instruct the second object on the follow-up strategy generated after handling the customer request during the session.

[0054] Optionally, in a customer service consultation scenario, a customer asks a customer service representative for detailed information about their loan repayment plan and expresses interest in a particular loan product. The customer service representative, through conversation with the customer, utilizes large-scale model analysis capabilities to generate initial conversation analysis information, including "customer is interested in repayment plans" and "interested in loan product type A". The work order prompts include "Record customer ID and account information"; "Clearly state the specific repayment plan details inquired about by the customer"; and "Note the type of product the customer expressed interest in". The initial conversation analysis information includes "Customer ID: 123456789"; "Account status: outstanding loans, etc."; "Customer inquiry: repayment date, repayment method, etc."; "Customer interest: product type A, etc."; and "Customer service response content". The generated target work order contains the following information: Work order title: "Customer Repayment Plan Inquiry and Product Consultation"; Customer information: Customer ID 123456789, account has outstanding loans; Problem description: The customer inquired about the repayment plan for their loan, including the specific repayment date and payment method, and expressed interest in product type A; Processing guidelines: Explain the customer's repayment plan in detail and prepare detailed information on product type A for the customer's reference; Priority: Medium; Follow-up plan: Arrange for a financial advisor to call the customer back to explain the repayment plan and introduce product type A.

[0055] Optionally, such as Figure 4 As shown, Figure 4This is a schematic diagram of an optional process for generating a target work order according to an embodiment of this application, including the following steps:

[0056] Step S402, Begin.

[0057] Step S404: The human agent clicks the upgrade work order button.

[0058] Step S406: The human agent enters the work order information entry page through the customer relationship management system and clicks the "Intelligently Generate Work Order" button.

[0059] Step S408: The customer relationship management system calls the API of the smart work order.

[0060] Step S410: Use the API of the smart work order to obtain the smart summary of this session in the Sales as a Service system.

[0061] Step S412: Encapsulate the work order prompt text using the smart work order API.

[0062] Step S414: The Sales as a Service system calls the large model.

[0063] In step S416, the Sales-as-a-Service system generates intelligent work orders (i.e., target work orders) based on work order prompts using a large model and stores the intelligent work orders.

[0064] Step S418: Render the page in the customer relationship management system and display the smart work order to the human agent.

[0065] Step S420: The human agent can modify the content of the smart work order in the customer relationship management system to obtain the target smart work order, and then click submit.

[0066] Step S422: The Sales-as-a-Service system obtains the target smart work order submitted by a human agent.

[0067] Step S424: Compare the smart summary and the target smart work order, generate a comparison record, and modify the work order prompt words based on the comparison record, such as... Figure 5 As shown, Figure 5 It describes the comparison information between the intelligent work order and the final submitted work order.

[0068] Step S426, End.

[0069] The work order generation method in this embodiment can be applied to scenarios such as customer service consultation, debt collection outbound calls, and usage follow-up in the consumer finance service field. Specifically, in the consumer finance service field, customer service representatives of consumer finance companies provide various services to customers through online channels, such as customer service, sales, debt collection, and follow-up. After the customer and customer service have finished communicating, the corresponding work order is usually edited manually based on the communication content between the customer and customer service. This is so that other customer services can be provided based on the work order or a structured customer service work analysis report can be generated. For example, based on the work order for debt collection outbound calls, suggestions for subsequent debt collection strategies and tags can be generated.

[0070] Although the aforementioned technologies have achieved automated recording of session content and standardized work order formats to some extent, they still ultimately rely on manual generation of work orders and lack the ability to adjust the work order format according to real-time changes in session content. Specifically, this manifests in the following aspects: the work order templates are too fixed and cannot flexibly cope with changing business environments and customer needs; the content in the work orders is not concise enough.

[0071] To address the inability to at least partially resolve the aforementioned technical problems, this embodiment utilizes conversation prompts to guide the analysis of the conversation between the first and second objects, generating initial conversation analysis information. This ensures that the generated initial conversation analysis information is more accurate and better meets user needs. Furthermore, based on the initial conversation analysis information and work order requirement information, work order prompts are dynamically generated. Finally, based on the work order prompts and initial conversation analysis information, a target work order is generated. This transforms the work order from a static template into one that can dynamically adjust its format and content requirements based on the current conversation analysis information and work order requirement information using work order prompts. Therefore, it solves the problem of not being able to dynamically generate work orders in related technologies, achieving the effect of dynamically generating work orders.

[0072] In an exemplary embodiment, generating initial session analysis information based on a session prompt and a session between a first object and a second object includes: extracting first session information of the first object and second session information of the second object from the session, wherein the first session information includes request information and response information sent by the first object to the second object, and the second session information includes feedback information sent by the second object to the first object; and generating the initial session analysis information based on the session prompt, the first session information, and the second session information.

[0073] Optionally, the demand information is used to indicate the specific questions, requests, or issues that need to be addressed raised by the first object; the response information is used to indicate the feedback from the first object to the information or suggestions provided to the second object. For example, demand information: The customer asks, "How do I apply for early repayment?" Response information: The customer confirms: Yes, I intend to use this option.

[0074] Optionally, feedback information is used to instruct the second object to provide answers, suggestions, guidance, or emotional support to the needs expressed by the first object.

[0075] Optionally, the initial conversation analysis information is a summary of conversation content and key information generated by the large model based on conversation prompts, the first conversation information of the first object, and the second conversation information of the second object, satisfying the conversation prompts. This information typically includes the customer's specific needs, the customer service's response, the business content involved in the conversation, and possible emotional states or customer attitudes. For example, in a product consultation scenario, conversation prompts are used to indicate that the generated initial conversation analysis information should include: the type of customer's question, the solution provided by the customer service, and the customer's immediate reaction to the solution. The output format rules for the generated initial conversation analysis information are: according to the content to be included in the initial conversation analysis information, the conversation information sent by the conversation objects is output item by item; the first conversation information includes the customer's inquiry about loan interest rates and confirmation of whether early repayment is possible, and the second conversation information includes the customer service's explanation of the current interest rate and confirmation of early repayment. Initial conversation analysis information: The customer inquired about the loan interest rate and expressed interest in early repayment; the customer service provided detailed information on the loan interest rate and explained the early repayment process and possible fees.

[0076] In this embodiment, conversation prompts guide the initial extraction of the needs and feedback of the first object and the response of the second object from a large number of dialogues, thus obtaining the first conversation information and the second conversation information, avoiding the omission of conversation information. At the same time, conversation prompts further guide the generation of initial conversation analysis information, effectively removing redundant and repetitive conversation information in the first and second conversation information, and improving the accuracy of generating initial conversation analysis information.

[0077] In an exemplary embodiment, generating the initial conversation analysis information based on the conversation prompt words, the first conversation information, and the second conversation information includes: extracting a first feature from the first conversation information based on the conversation prompt words, wherein the first feature includes at least one of the following: a first demand feature issued by the first object to the second object, a first question feature issued by the first object to the second object, a first identity feature of the first object, a first result feature issued by the first object to the second object, and an emotional feature of the first object; extracting a second feature from the second conversation information based on the conversation prompt words, wherein the second feature includes: a first strategy feature for resolving the first problem fed back by the second object to the first object; and generating the initial conversation analysis information based on the first feature, the second feature, the first conversation information, and the second conversation information.

[0078] Optionally, extracting a first feature from the first conversation information and a second feature from the second conversation information based on conversation prompt words includes: automatically labeling and classifying key elements in the conversation, such as needs, questions, feedback and emotional states, by searching for specific words and identifying sentiment tendencies in the first and second conversation information, to obtain the first and second features.

[0079] Optionally, the first feature is used to indicate the key attributes extracted from the first conversation information of the first object based on the conversation prompt words, reflecting the needs, problems, identity or emotional state of the first object, specifically including at least one of the following: first need feature, first problem feature, first identity feature, first result feature, and emotional feature.

[0080] Optionally, the second feature is used to indicate key attributes extracted from the second conversation information of the second object based on conversation prompts, reflecting how the second object responds to the questions, needs, or emotions of the first object, specifically including at least one of the following: a first strategy feature, a second need feature issued by the second object to the first object, a second question feature asked by the second object to the first object, and an emotional feature of the second object.

[0081] Optionally, the initial session analysis information is a session summary that integrates the first feature, the second feature, and the first and second session information, used to indicate key information of the session and the interaction between the first and second objects.

[0082] Optionally, generating the initial conversation analysis information based on the first feature, the second feature, the first conversation information, and the second conversation information includes: using the semantic understanding and context awareness capabilities of a large language model to analyze and summarize specific conversation information—the first conversation information and the second conversation information—based on the first feature and the second feature, to obtain the initial conversation analysis information.

[0083] Optionally, generating the initial conversation analysis information based on the first feature, the second feature, the first conversation information, and the second conversation information includes: generating summary-style initial conversation analysis information using a large language model based on key elements in the categorized conversation, such as needs, questions, feedback, and emotional states.

[0084] Optionally, in a collection outbound call scenario, the first target is the customer who has defaulted on payment, and the second target is the collection agent. Conversation prompts indicate that the generated initial conversation analysis information should include: the customer's reason for default, the customer's willingness to repay, the customer's communication attitude, and the agent's collection strategy. The output format rules for the generated initial conversation analysis information are: output the conversation information sent by the conversation target item by item according to the content to be included in the initial conversation analysis information. The first demand feature extracted from the first conversation information is: explaining the reason for the default. The first problem feature is: inquiring whether it is possible to adjust the repayment plan; the first outcome feature is: the customer expresses a willingness to resolve the default issue as soon as possible, but needs time; the emotional feature is: the customer shows anxiety and urgency in the conversation, with a certain degree of unease in their tone. The first strategy feature extracted from the second conversation information is: providing options for deferred repayment and methods for reducing late fees. The large model generates initial conversation analysis information based on the conversation prompts, the first conversation information (the customer's description of their family situation and financial difficulties), and the second conversation information (the agent's expression of empathy and solutions). The initial conversation analysis information includes, but is not limited to: the conversation involves the customer's family's urgent medical expenses leading to repayment difficulties. The customer expressed a strong willingness to repay, but sought adjustments to their repayment plan due to financial pressure. The customer service representative offered a payment extension and late fee reduction to alleviate the customer's current financial burden, while maintaining a positive and friendly communication style, which helped build customer trust and facilitate a resolution.

[0085] In this embodiment, feature extraction is used to transform dialogue content into structured information. Based on the extracted features, initial conversation analysis information is generated, ensuring that the generated initial conversation analysis information has strong logic and further improving the accuracy of the generated initial conversation analysis information.

[0086] In one exemplary embodiment, generating a work order prompt based on the initial session analysis information and work order requirement information includes: determining a work order type from the work order requirement information and generating a first work order prompt based on the work order type, wherein the first work order prompt is used to represent the format of the generated work order template; generating a second work order prompt based on the first work order prompt and the initial session analysis information, wherein the second work order prompt is used to represent the work order content that meets the work order type determined from the initial session analysis information; and determining the first work order prompt and the second work order prompt as the work order prompt.

[0087] Optionally, the work order requirement information refers to the specific requirements and details for creating the work order. Different application scenarios correspond to different work order requirement information. For example, the work order requirement information for a loan consultation work order includes, but is not limited to, basic customer information, loan type, consultation question, and consultation suggestions. For example, the work order requirement information for an overdue collection work order includes, but is not limited to, basic customer information, reason for overdue payment, extension application, and operational suggestions.

[0088] Optionally, different work order types correspond to different processing procedures, responsible departments, or response times. Work order types include, but are not limited to, repayment inquiry work orders, loan application work orders, overdue collection work orders, complaints and suggestions, technical fault report work orders, credit assessment appeal work orders, account information change work orders, and product inquiry work orders.

[0089] Optionally, the first work order prompt is based on a specific work order type and is used to instruct the work order template to follow a specific format or structure.

[0090] Optionally, the second work order prompt is an instruction to fill in various details in the work order template based on the specific content in the initial session analysis information. For example, the customer problem field should be filled with a specific description of the service problem.

[0091] This embodiment generates a first work order prompt and a second work order prompt, ensuring that the generated work orders follow unified professional standards, while also dynamically creating different work orders based on different application scenarios, thus improving the flexibility of work order generation.

[0092] In an exemplary embodiment, generating a second work order prompt based on the first work order prompt and the initial session analysis information includes: determining multiple target work order fields included in the target work order based on the first work order prompt; determining attribute information of the multiple target work order fields based on the initial session analysis information to obtain multiple basic work order prompts; and determining the second work order prompt from the multiple basic work order prompts, wherein the second work order prompt is a prompt that satisfies a preset rule among the multiple basic work order prompts.

[0093] Optionally, the target work order fields are the various information modules that need to be filled in when creating a specific type of work order. They constitute the basic framework of the work order, ensuring the complete collection of all necessary information. Attribute information is used to indicate the type and format of the target work order fields. For example, the "Customer Problem" field may require a detailed problem description, while the "Solution" field may require recording the preliminary solutions proposed.

[0094] Optionally, the basic work order prompt word is a fill suggestion or instruction generated for each target work order field based on the first work order prompt word and the initial session analysis message.

[0095] Optionally, the second work order prompt is a set of prompts selected from multiple basic work order prompts that meet preset rules. These preset rules include, but are not limited to, accuracy rules (selecting prompts that precisely reflect the problem of the first target, rather than vague descriptions); sensitivity rules (avoiding ambiguous terms); compliance rules (ensuring prompts adhere to company decisions and laws and regulations); and service standard rules (reflecting high service standards).

[0096] In an exemplary embodiment, generating a target work order based on the aforementioned work order prompt and the aforementioned initial session analysis information includes: modifying a preset initial work order template based on the aforementioned first work order prompt to generate a target work order template, wherein the format of the target work order template satisfies the format of the aforementioned work order template; generating target work order content based on the aforementioned second work order prompt and the aforementioned initial session analysis information; and filling the target work order content into the aforementioned target work order template to obtain the aforementioned target work order.

[0097] Optionally, the initial work order template is a preset table or document format that includes a series of fixed fields, such as session time, identity information of the first and second objects, business type, and whether it is urgent, as shown in Table 1.

[0098] Table 1:

[0099] Field Name Field type Required / Optional Customer ID text Required Session time time Required Business type Drop-down list (Loans / Repayments / Inquiries) Required Problem Description text Required Is it urgent? Select (Yes / No) Required

[0100] Optionally, the target work order template is a modified work order template based on the initial work order template and using the prompt words of the first work order. It not only retains the integrity of the original initial work order template, but also adjusts the field order, prompt information, etc. according to the prompt words to better adapt to the creation needs of specific work orders, as shown in Table 2. Table 2 is the target work order template modified from the initial work order template in Table 1.

[0101] Table 2:

[0102]

[0103]

[0104] Optionally, the target work order content is a set of specific work order information filled in based on the second work order prompt and the initial session analysis information, including a personalized description and key data for the first object.

[0105] This embodiment reduces the steps of manual input and editing by automatically generating target work orders, thereby improving the efficiency of work order generation.

[0106] In an exemplary embodiment, generating target work order content based on the second work order prompt word and the initial session analysis information includes: determining session analysis information matching the second work order prompt word from the initial session analysis information to obtain the target work order content.

[0107] The method for generating work orders in the embodiments of this application will be explained below with reference to optional examples.

[0108] Figure 6 This is a flowchart illustrating another optional method for generating work orders according to an embodiment of this application. Figure 1 ,like Figure 6 As shown, in a customer service consultation scenario, the process for generating this work order may include the following steps:

[0109] Step S602: The customer initiates an inquiry request.

[0110] Step S604 involves intent recognition and dialogue flow control to either enable the customer service robot to respond or allow a human agent to respond. During the human agent response process, a large-scale model is used to analyze question-and-answer prompts and implement intelligent assistance functions. This provides real-time scripts and communication strategies via a floating screen, improving customer service efficiency and quality. Specifically, during communication with customers, the customer service representative encodes the customer's question and uses Retrieval-Augmented Generation (RAG) to retrieve relevant business knowledge from the company's knowledge base. This information, along with the customer's request, is then fed into the large-scale model to generate a suggested response, which is displayed as a prompt to the customer service representative on the agent assistant interface.

[0111] Step S606: After responding, record the conversation between the customer and customer service.

[0112] Step S608, Large Model Analysis.

[0113] In step S610, the large model analyzes the dialogue content based on conversation prompts to obtain an intelligent summary; it also analyzes the intelligent summary based on work order prompts to obtain an intelligent work order.

[0114] In step S612, the human agent modifies the content in the intelligent summary to obtain the target intelligent summary. The large model compares the intelligent summary and the target intelligent summary to generate a comparison record. Based on the comparison record, the conversation prompt words are optimized. The human agent modifies the content in the intelligent work order to obtain the target intelligent work order. The large model compares the intelligent work order and the target intelligent work order to generate a comparison record. Based on the comparison record, the work order prompt words are optimized. At the same time, the subsequent execution results of the work order are recorded. Based on the subsequent execution results, the conversation prompt words and work order prompt words are further optimized.

[0115] Figure 7This is a flowchart illustrating another optional method for generating work orders according to an embodiment of this application. Figure 2 ,like Figure 7 As shown, in a debt collection outbound call scenario, the process for generating this work order may include the following steps:

[0116] Step S702: Customer service makes outbound calls to customers to collect debts.

[0117] Step S704, customer response, wherein, during the response process, the big model generates collection strategy reminders based on question and answer prompts, the customer's current tag information in the user center, and the current user profile.

[0118] Step S706: After responding, record the conversation between the customer and customer service.

[0119] Step S708, Large Model Analysis.

[0120] Step S710: The large model analyzes the dialogue content based on conversation prompts to obtain an intelligent summary; it analyzes the intelligent summary based on work order prompts to obtain an intelligent work order; the intelligent work order includes key information such as the customer's reason for delinquency, willingness to repay, and attitude.

[0121] In step S712, the human agent modifies the content in the smart summary to obtain the target smart summary. The large model compares the smart summary and the target smart summary to generate a comparison record, and optimizes the conversation prompts based on the comparison record. The human agent also modifies the content in the smart work order to obtain the target smart work order. The large model compares the smart work order and the target smart work order to generate a comparison record, optimizes the work order prompts based on the comparison record, and records the subsequent execution results of the work order. Based on the subsequent execution results, the conversation prompts and work order prompts are further optimized. The current tag information in the user center is improved based on the smart work order and smart summary.

[0122] Figure 8 This is a flowchart illustrating another optional method for generating work orders according to an embodiment of this application. Figure 3 ,like Figure 8 As shown, in a usage follow-up scenario, the process of generating this work order may include the following steps:

[0123] Step S802: Customer service makes a follow-up call to the customer to inquire about the purpose of the service.

[0124] Step S804, customer response.

[0125] Step S806: After responding, record the conversation between the customer and customer service.

[0126] Step S808, Large Model Analysis.

[0127] In step S810, the large model analyzes the dialogue content based on conversation prompts to obtain an intelligent summary; it analyzes the intelligent summary based on work order prompts to obtain an intelligent work order; the intelligent work order includes key information such as the customer's loan purpose.

[0128] In step S812, the human agent modifies the content in the intelligent summary to obtain the target intelligent summary. The large model compares the intelligent summary and the target intelligent summary to generate a comparison record. Based on the comparison record, the conversation prompt words are optimized. The human agent modifies the content in the intelligent work order to obtain the target intelligent work order. The large model compares the intelligent work order and the target intelligent work order to generate a comparison record. Based on the comparison record, the work order prompt words are optimized. At the same time, the subsequent execution results of the work order are recorded. Based on the subsequent execution results, the conversation prompt words and work order prompt words are further optimized.

[0129] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0130] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as read-only memory (ROM) / random access memory (RAM), magnetic disk, optical disk), and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0131] According to another aspect of the embodiments of this application, a work order generation apparatus is also provided. This work order generation apparatus can be used to implement the work order generation method provided in the above embodiments, and details already described will not be repeated. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the apparatus described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0132] Figure 9This is a structural block diagram of an optional work order generation device according to an embodiment of this application, such as... Figure 9 As shown, the device for generating the work order includes:

[0133] The first generation module 902 is used to generate initial session analysis information based on the session prompt words and the session between the first object and the second object, wherein the session prompt words are used to indicate the rules for generating the initial session analysis information.

[0134] The second generation module 904 is used to generate a work order prompt based on the initial session analysis information and the work order requirement information, wherein the work order requirement information is used to indicate the second object's requirement to generate a work order, and the work order prompt is used to indicate the rules for generating a work order.

[0135] The third generation module 906 is used to generate a target work order based on the above-mentioned work order prompt words and the above-mentioned initial session analysis information, wherein the target work order is used to record the session between the above-mentioned first object and the above-mentioned second object.

[0136] It should be noted that the first generation module 902 in this embodiment can be used to execute the above step S202, the second generation module 904 in this embodiment can be used to execute the above step S204, and the third generation module 906 in this embodiment can be used to execute the above step S206.

[0137] In an exemplary embodiment, the first generation module 902 includes: a first extraction submodule, configured to extract first session information of the first object and second session information of the second object from the session, wherein the first session information includes request information and response information sent by the first object to the second object, and the second session information includes feedback information sent by the second object to the first object; and a first generation submodule, configured to generate the initial session analysis information based on the session prompt words, the first session information, and the second session information.

[0138] In an exemplary embodiment, the first generation submodule includes: a first extraction unit, configured to extract a first feature from the first conversation information based on the conversation prompt words, wherein the first feature includes at least one of the following: a first demand feature issued by the first object to the second object, a first question feature issued by the first object to the second object, a first identity feature of the first object, a first result feature issued by the first object to the second object, and an emotion feature of the first object; a first extraction unit, configured to extract a second feature from the second conversation information based on the conversation prompt words, wherein the second feature includes: a first strategy feature for resolving the first problem fed back by the second object to the first object; and a first generation unit, configured to generate the initial conversation analysis information based on the first feature, the second feature, the first conversation information, and the second conversation information.

[0139] In an exemplary embodiment, the second generation module 904 includes: a second generation submodule, configured to determine the work order type from the work order requirement information and generate a first work order prompt word based on the work order type, wherein the first work order prompt word is used to represent the format of the generated work order template; a third generation submodule, configured to generate a second work order prompt word based on the first work order prompt word and the initial session analysis information, wherein the second work order prompt word is used to represent the work order content that meets the work order type determined from the initial session analysis information; and a first determination submodule, configured to determine the first work order prompt word and the second work order prompt word as the work order prompt word.

[0140] In an exemplary embodiment, the third generation submodule includes: a first determining unit, configured to determine multiple target work order fields included in the target work order based on the first work order prompt word; a second determining unit, configured to determine attribute information of the multiple target work order fields based on the initial session analysis information, thereby obtaining multiple basic work order prompt words; and a third determining unit, configured to determine the second work order prompt word from the multiple basic work order prompt words, wherein the second work order prompt word is a prompt word that satisfies a preset rule among the multiple basic work order prompt words.

[0141] In an exemplary embodiment, the third generation module 906 includes: a fourth generation submodule, configured to modify a preset initial work order template based on the first work order prompt word to generate a target work order template, wherein the format of the target work order template satisfies the format of the work order template; a fifth generation submodule, configured to generate target work order content based on the second work order prompt word and the initial session analysis information; and a first filling submodule, configured to fill the target work order content into the target work order template to obtain the target work order.

[0142] In an exemplary embodiment, the fifth generation submodule includes: a fourth determining unit, configured to determine, from the initial session analysis information, session analysis information matching the second work order prompt word, and obtain the target work order content.

[0143] It should be noted that the above modules can be implemented by software or hardware. For the latter, they can be implemented in the following ways, but are not limited to: all the above modules are located in the same processor; or, the above modules are located in different processors in any combination.

[0144] According to another aspect of the embodiments of this application, a computer-readable storage medium is provided, the computer-readable storage medium including a stored program, wherein the program executes the steps in any of the above method embodiments when it is run.

[0145] In one exemplary embodiment, the aforementioned computer-readable storage medium may include, but is not limited to, various media capable of storing computer programs, such as USB flash drives, ROMs, RAMs, portable hard drives, magnetic disks, or optical disks.

[0146] According to another aspect of the embodiments of this application, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor is configured to perform the steps of any of the method embodiments described above via the computer program. In an exemplary embodiment, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0147] Specific examples in this embodiment can be found in the examples described in the above embodiments and exemplary implementations, and will not be repeated here.

[0148] According to another aspect of the embodiments of this application, a computer program product is also provided, which includes a computer program / instructions containing program code for performing the methods shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network via a communication component, and / or installed from a removable medium. When the computer program is executed by a central processing unit, it performs various functions provided in the embodiments of this application. The sequence numbers of the embodiments of this application above are merely descriptive and do not represent the superiority or inferiority of the embodiments.

[0149] Obviously, those skilled in the art should understand that the modules or steps of this application described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. They can be implemented using computer-executable program code, and thus can be stored in a storage device for execution by a computing device. In some cases, the steps shown or described can be performed in a different order than those described herein, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, this application is not limited to any particular combination of hardware and software.

[0150] The collection, storage, use, processing, transmission, provision, and disclosure of financial data or user data involved in the technical solution of this application all comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0151] It should be noted that in the embodiments of this application, certain software, components, models and other existing solutions in the industry may be mentioned. These should be regarded as exemplary and are only intended to illustrate the feasibility of implementing the technical solution of this application. However, it does not mean that the applicant has used or necessarily used the solution.

[0152] The above are merely preferred embodiments of this application and are not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the principles of this application should be included within the protection scope of this application.

Claims

1. A method for generating work orders, characterized in that, include: Based on the conversation prompt and the conversation between the first object and the second object, initial conversation analysis information is generated, wherein the conversation prompt is used to indicate the rules for generating the initial conversation analysis information; Based on the initial session analysis information and the work order request information, a work order prompt is generated, wherein the work order request information is used to indicate the second object's request to generate a work order, and the work order prompt is used to indicate the rules for generating a work order; Based on the work order prompt and the initial session analysis information, a target work order is generated, wherein the target work order is used to record the session between the first object and the second object.

2. The method according to claim 1, characterized in that, Based on the conversation prompts and the conversation between the first and second objects, initial conversation analysis information is generated, including: Extract first session information of the first object and second session information of the second object from the session, wherein the first session information includes request information and response information sent by the first object to the second object, and the second session information includes feedback information sent by the second object to the first object; The initial session analysis information is generated based on the session prompt words, the first session information, and the second session information.

3. The method according to claim 2, characterized in that, Based on the conversation prompt words, the first conversation information, and the second conversation information, the initial conversation analysis information is generated, including: Based on the conversation prompt words, a first feature is extracted from the first conversation information, wherein the first feature includes at least one of the following: a first demand feature issued by the first object to the second object, a first question feature issued by the first object to the second object, a first identity feature of the first object, a first result feature issued by the first object to the second object, and an emotional feature of the first object. Based on the conversation prompt words, a second feature is extracted from the second conversation information, wherein the second feature includes: a first strategy feature for solving the first problem fed back by the second object to the first object; The initial session analysis information is generated based on the first feature, the second feature, the first session information, and the second session information.

4. The method according to claim 1, characterized in that, Based on the initial session analysis information and work order request information, a work order prompt message is generated, including: The work order type is determined from the work order requirement information, and a first work order prompt word is generated based on the work order type, wherein the first work order prompt word is used to indicate the format of the generated work order template; A second work order prompt is generated based on the first work order prompt and the initial session analysis information, wherein the second work order prompt is used to indicate work order content that meets the work order type determined from the initial session analysis information; The first work order prompt word and the second work order prompt word are determined as the work order prompt words.

5. The method according to claim 4, characterized in that, Generate a second work order prompt based on the first work order prompt and the initial session analysis information, including: Based on the first work order prompt, determine the multiple target work order fields included in the target work order; Based on the initial session analysis information, the attribute information of multiple target work order fields is determined, and multiple basic work order prompt words are obtained; The second work order prompt word is determined from a plurality of the basic work order prompt words, wherein the second work order prompt word is a prompt word that satisfies a preset rule among the plurality of basic work order prompt words.

6. The method according to claim 5, characterized in that, Based on the work order prompt and the initial session analysis information, a target work order is generated, including: Based on the first work order prompt, modify the preset initial work order template to generate a target work order template, wherein the format of the target work order template satisfies the format of the work order template; Based on the second work order prompt and the initial session analysis information, the target work order content is generated; The target work order content is filled into the target work order template to obtain the target work order.

7. The method according to claim 6, characterized in that, Based on the second work order prompt and the initial session analysis information, the target work order content is generated, including: The target work order content is obtained by determining the conversation analysis information that matches the second work order prompt word from the initial conversation analysis information.

8. A work order generation device, characterized in that, include: The first generation module is used to generate initial session analysis information based on the session prompt words and the session between the first object and the second object, wherein the session prompt words are used to indicate the rules for generating the initial session analysis information; The second generation module is used to generate a work order prompt based on the initial session analysis information and the work order request information, wherein the work order request information is used to indicate the second object's request to generate a work order, and the work order prompt is used to indicate the rules for generating a work order. The third generation module is used to generate a target work order based on the work order prompt words and the initial session analysis information, wherein the target work order is used to record the session between the first object and the second object.

9. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein the computer program, when executed by a processor, implements the steps of the method according to any one of claims 1 to 7.

Images