Online work order processing method and equipment

Abstract

The embodiment of the invention provides an online work order processing method and equipment, and the method comprises the steps: monitoring a customer service work order system in a bypass mode through a preset work order large language model, and obtaining the conversation content generated by a target work order which is currently processed by the customer service work order system; wherein the work order large language model is generated by adopting a work order data set with a label to carry out preference learning on a general large language model; performing classification judgment on the dialogue content by adopting a thinking chain mode through a work order large language model, and determining whether the dialogue content accords with a category needing work order upgrading processing or not; and when the dialogue content accords with the category of the work order needing to be upgraded, automatically triggering upgrading processing of the customer service work order system for the target work order. According to the embodiment of the invention, whether the customer service work order needs to be upgraded or not can be automatically and accurately judged through the large language model, the problem that the work order is delayed due to the fact that the work order is not accurately or timely upgraded is avoided, manual control is not needed, rules do not need to be maintained, the accuracy is improved, and the cost is reduced.

Description

Technical Field

[0001] The embodiments of the present disclosure relate to the field of computer and network communication technology, and more particularly to an online work order processing method and device. Background Art

[0002] In some network services, service stability is becoming increasingly important to service providers. Service providers usually arrange customer service personnel to respond to issues raised by customers, promptly identify online service risks, promote timely resolution of online problems, effectively control negative impacts, improve service stability and ultimately improve customer satisfaction with the service.

[0003] The problems reported by customers are usually open questions. In a customer service ticket, the problems reported by customers may involve multiple aspects such as product usage, business processes, troubleshooting, etc. Considering customer service efficiency and cost, different customer service personnel are usually assigned to handle different problems. When the current customer service personnel cannot solve the problem, the ticket needs to be upgraded, that is, the customer service personnel who can solve the problem will be added to the conversation. In the existing technology, whether the ticket is upgraded is usually controlled manually, or according to preset rules.

[0004] However, manual control is inefficient and may delay problem resolution; whereas rule-based control requires rule maintenance and is prone to misjudgment. Summary of the invention

[0005] The embodiments of the present disclosure provide an online work order processing method and device to solve the above technical problems.

[0006] In a first aspect, an embodiment of the present disclosure provides an online work order processing method, comprising:

[0007] The customer service work order system is monitored in a bypass mode through a preset work order large language model to obtain the conversation content generated by the target work order currently being processed by the customer service work order system; wherein the work order large language model is generated by preference learning of a general large language model using a labeled work order data set;

[0008] The work order large language model is used to classify and judge the conversation content in a chain of thought manner to determine whether the conversation content meets the category that requires work order upgrade processing;

[0009] When the content of the conversation meets the category of the work order that needs to be upgraded, the customer service work order system is automatically triggered to upgrade the target work order.

[0010] In a second aspect, an embodiment of the present disclosure provides an online work order processing device, including:

[0011] An acquisition unit is used to monitor the customer service work order system in a bypass mode through a preset work order large language model to obtain the conversation content generated by the target work order currently being processed by the customer service work order system; wherein the work order large language model is generated by preference learning of a general large language model using a work order data set with labels;

[0012] A judgment unit, configured to classify and judge the conversation content by using the work order large language model in a thought chain manner, and determine whether the conversation content meets the category requiring work order upgrade processing;

[0013] The upgrade unit is used to automatically trigger the customer service work order system to upgrade the target work order when the content of the conversation meets the category of the work order that needs to be upgraded.

[0014] In a third aspect, an embodiment of the present disclosure provides an electronic device, including: a processor and a memory;

[0015] The memory stores computer-executable instructions;

[0016] The processor executes the computer-executable instructions stored in the memory, so that the at least one processor executes the online work order processing method as described in the first aspect and various possible designs of the first aspect.

[0017] In a fourth aspect, an embodiment of the present disclosure provides a computer-readable storage medium, in which computer execution instructions are stored. When a processor executes the computer execution instructions, the online work order processing method described in the first aspect and various possible designs of the first aspect is implemented.

[0018] In a fifth aspect, an embodiment of the present disclosure provides a computer program product, including a computer program, which, when executed by a processor, implements the online work order processing method as described in the first aspect and various possible designs of the first aspect.

[0019] The online work order processing method and device provided in the disclosed embodiment monitors the customer service work order system in a bypass manner through a preset work order large language model to obtain the conversation content generated by the target work order currently being processed by the customer service work order system; wherein the work order large language model is generated by preference learning of a general large language model using a labeled work order data set; the work order large language model classifies and judges the conversation content using a thinking chain method to determine whether the conversation content meets the category of work order upgrade processing required; when the conversation content meets the category of work order upgrade processing required, the customer service work order system is automatically triggered to upgrade the target work order. In the disclosed embodiment, the large language model can be used to automatically and accurately determine whether a customer service work order needs to be upgraded, thereby avoiding delays in the processing of work orders due to failure to accurately or timely upgrade the work order, and no manual control or maintenance rules are required, thereby improving accuracy and reducing costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0021] Figure 1 A schematic diagram of a scenario of an online work order processing method provided by an embodiment of the present disclosure;

[0022] Figure 2 A schematic diagram of a flow chart of an online work order processing method provided by an embodiment of the present disclosure;

[0023] Figure 3 A schematic diagram of a flow chart of an online work order processing method provided by another embodiment of the present disclosure;

[0024] Figure 4 A structural block diagram of an online work order processing device provided in an embodiment of the present disclosure;

[0025] Figure 5 A schematic diagram of the hardware structure of an online work order processing device provided in one embodiment of the present disclosure. DETAILED DESCRIPTION

[0026] In order to make the purpose, technical solution and advantages of the embodiments of the present disclosure clearer, the technical solution in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are part of the embodiments of the present disclosure, not all of the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present disclosure.

[0027] The problems reported by customers are usually open questions. In a customer service ticket, the problems reported by customers may involve multiple aspects such as product use, business processes, troubleshooting, etc. Considering customer service efficiency and cost, different customer service personnel are usually assigned to handle different problems. For example, customer service personnel can be divided into different levels. For example, the first-level customer service personnel can handle product consultation issues or other simple issues, and the second-level customer service personnel can handle complex issues such as product failures, or other methods of dividing levels can be used. When customers encounter different problems, they need to add the corresponding customer service personnel to the conversation. In the existing technology, the customer service end is usually manually controlled to join the conversation, or the customer service end is controlled to join the conversation according to preset rules.

[0028] However, manual control is inefficient and may delay problem resolution. Different people have different judgment criteria, requiring a lot of training. Rule-based control requires rule maintenance and is prone to misjudgment. For example, keyword-based rules cannot understand the semantics of the conversation context and will confuse the correct semantics of certain keywords, leading to misjudgment.

[0029] In order to solve the above technical problems, the embodiments of the present disclosure provide an online work order processing method, which can automatically and accurately determine whether a customer service work order needs to be upgraded through a large language model, thereby avoiding delays in work order processing due to failure to accurately or timely upgrade the work order. There is no need for manual control or maintenance rules, which improves accuracy and reduces costs.

[0030] The application scenarios of the online work order processing method of the disclosed embodiment are as follows: Figure 1 As shown, the customer service work order system is monitored in a bypass manner through the preset work order large language model to obtain the conversation content generated by the target work order currently being processed by the customer service work order system, that is, the conversation content between the client and the first customer service end; the conversation content is classified and judged by the work order large language model in a thinking chain manner to determine whether the conversation content meets the category of work order upgrade processing; when the conversation content meets the category of work order upgrade processing, the customer service work order system is automatically triggered to upgrade the target work order, that is, the second customer service end is added to the conversation, and the work order person in charge of the second customer service end is introduced into the communication group of the target work order to jointly participate in the processing of the work order.

[0031] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this disclosure are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0032] The online work order processing method disclosed in the present invention will be described in detail below in conjunction with specific embodiments.

[0033] refer to Figure 2 , Figure 2 A flowchart of an online work order processing method provided by an embodiment of the present disclosure. The method of this embodiment can be applied in a terminal device or a server, and the online work order processing method includes:

[0034] S201. Monitor the customer service work order system in a bypass manner through a preset work order large language model to obtain the conversation content generated by the target work order currently being processed by the customer service work order system; wherein the work order large language model is generated by preference learning a general large language model using a labeled work order data set.

[0035] In this embodiment, the customer service work order is a form that records the conversations between the customer and the customer service, such as consultation and complaints. The customer service work order system can receive, distribute, and circulate the customer service work order, and can also conduct conversations between the client and the customer service end or record the content of the conversation. In the customer service work order system, the customer service end can be divided into different levels. For example, the first-level customer service end can be a customer service end that handles product consultation issues or other simple issues, or a customer service end that plays the role of receiving customers. The second-level customer service end can be a customer service end that handles complex issues such as product failures. Of course, there can also be other levels of customer service ends, or other methods of dividing levels can be used. After the customer service work order system receives a customer service work order, it can first specify a certain level of customer service end to handle the customer service work order. If the customer service end encounters a problem that cannot be handled, it is necessary to upgrade the customer service work order to add a higher-level customer service end to the conversation of the customer service work order.

[0036] In this embodiment, in order to realize the automatic upgrade of customer service work orders, a work order large language model can be pre-trained. The work order large language model is generated by preference learning of a general large language model using a labeled work order data set, so that it has the ability to classify and judge according to the conversation content of the work order, and determine whether the conversation content meets the category that requires work order upgrade processing, so as to upgrade the work order. The work order large language model can be any language model that can process natural language tasks, such as the LLM large language model, etc., which is not limited in this embodiment.

[0037] The work order big prediction model can monitor the customer service work order system in a bypass manner, that is, it can monitor the conversation content generated by the target work order currently being processed by the customer service work order system, obtain the conversation content from the target work order and input it into the work order big prediction model for processing without affecting the normal operation of the customer service work order system.

[0038] Optionally, in addition to obtaining the conversation content generated by the target work order, the work order big prediction model can also obtain other information such as the metadata of the target work order (such as title, problem source, work order form content, etc.).

[0039] S202: using the work order large language model and adopting a chain of thought approach to classify and judge the conversation content, and determine whether the conversation content meets the category that requires work order upgrade processing.

[0040] In this embodiment, after obtaining the conversation content of the target work order, the conversation content can be classified and judged through the work order large language model, wherein the work order large language model has the ability to classify and judge the conversation content after the above-mentioned training process, that is, to judge which one or more preset conversation content categories the conversation content specifically belongs to, some categories are categories that require work order upgrade processing, and other categories are categories that do not require work order upgrade processing. Then, the work order large language model classifies and judges the conversation content, and can determine whether the conversation content meets the categories that require work order upgrade processing.

[0041] Optionally, the above categories may be topic categories of the conversation content, wherein the topic categories that require work order escalation processing are topic categories that cannot be processed by low-level client terminals, while the topic categories that do not require work order escalation processing are topic categories that can be processed by low-level client terminals. For example, the topic categories may specifically include but are not limited to one or more of the following:

[0042] 1) Large-scale failures, such as problems with multiple customers, multiple regions, or multiple products;

[0043] 2) Customer services are damaged, such as unavailability, jitter, timeout, etc., or there are potential risks during customer service peak periods;

[0044] 3) The customer has negative emotions, such as customer relationship risk or the customer is ready to stop using the product;

[0045] 4) Key customer failures, such as usage problems encountered by some key customers.

[0046] 5) Other issues are the types of topics that do not require customer service ticket escalation. If they do not belong to the above topic categories, they can be classified as other issues.

[0047] Among them, the first four topic categories are topic categories that cannot be handled by low-level client terminals, that is, topic categories that require work order upgrade processing, while other topic categories except the first four topic categories can be handled by low-level client terminals.

[0048] When the work order large language model classifies and judges the content of the conversation, the input of the work order large language model is the input instruction Prompt, which is an instruction used to instruct or guide the model to produce a specific output. In this embodiment, an input instruction can be constructed to instruct the work order large language model to determine the topic category of the conversation content according to preset classification rules, where the preset classification rules can be any feasible classification rules as long as it can determine which preset topic category the conversation content belongs to.

[0049] In order to improve accuracy and explainability, the Chain of Thought (CoT) method can be used, where the Chain of Thought refers to a complete thinking process (including content analysis and classification basis) before the language model makes a category judgment, and then gives a specific classification category. When implementing the Chain of Thought method, instructions for outputting the analysis process and classification basis can be added to the input instructions, so that the work order language model can output the specific analysis process and classification basis.

[0050] In a specific implementation, an input instruction of the work order large language model can be constructed according to the conversation content, wherein the input instruction is used to instruct the work order large language model to determine the topic category of the conversation content according to a preset classification rule, and output the analysis process and classification basis;

[0051] The work order large language model outputs the topic category of the conversation content, as well as the analysis process and classification basis according to the input instruction; determines whether the topic category of the conversation content belongs to a preset topic category that requires work order upgrade processing, and if so, determines whether the conversation content meets the category that requires work order upgrade processing.

[0052] For example, one or more of the following constraints can be added to the input instructions of the work order language model (system prompt):

[0053] 1) For uncertain tones, interrogative sentences, consultations, hypothetical / expected / imaginary problems, prevention / avoidance / worry about problems, problems that have been solved / have solutions, business processes, progress synchronization, conclusion synchronization, testing / stress testing and other environments, and those that do not clearly indicate that problems or failures have occurred, please classify them as "other issues";

[0054] 2) Do not make assumptions or inferences, but make judgments strictly according to the given rules;

[0055] 3) Please give your thoughts and make judgments based on your thoughts;

[0056] 4) The classification type must belong to the type given in the judgment rule.

[0057] Through the above constraints, the work order language model can output the topic category of the current conversation content, as well as the analysis process and classification basis, improve the accuracy and explainability of the topic category classification, and increase credibility. In order to enable the work order language model to output the topic category of the conversation content, as well as the analysis process and classification basis, it can be trained in advance.

[0058] Of course, the work order language model's use of a chain of thought approach to classify and judge conversation content is not limited to the above-mentioned method, and can also be implemented in any other feasible way, such as fully annotating the training data during the training process, wherein the annotated labels include the topic category of the conversation content, as well as the analysis process and classification basis, thereby enabling the work order language model to output the topic category of the conversation content, as well as the analysis process and classification basis.

[0059] S203: When the content of the conversation meets the category of the work order that needs to be upgraded, the customer service work order system is automatically triggered to upgrade the target work order.

[0060] In this embodiment, when it is determined that the conversation content meets the category of work order that needs to be upgraded, the work order upgrade process may be required, and the customer service work order system may be automatically triggered to upgrade the target work order. Specifically, a trigger instruction for work order upgrade may be sent to the customer service work order system, and the customer service work order system may upgrade the target work order according to the trigger instruction. The upgrade process also involves adding a higher-level customer service end to the conversation, so that the higher-level customer service end can investigate and solve the problem together.

[0061] Optionally, when it is determined that the conversation content meets the category of work order that needs to be upgraded, the conversation content is analyzed through the work order large language model to determine the troubleshooting direction involved in the target work order, and the customer service work order system is automatically triggered to introduce the work order person in charge corresponding to the troubleshooting direction into the communication group for the target work order to continue processing.

[0062] In this embodiment, the conversation content can be analyzed through the work order large language model to determine the troubleshooting direction involved in the target work order. The troubleshooting direction is the content that needs to be investigated when finding the abnormal cause of the problem phenomenon to be solved. For example, the problem of message sending failure may be caused by network reasons or resource problems (CPU, storage devices, etc.). Therefore, it is necessary to troubleshoot the network and resources, that is, different troubleshooting directions, and different troubleshooting directions may involve the participation of different client terminals. Therefore, when the work order is upgraded, the person in charge of the troubleshooting direction can be pulled into the conversation, that is, the person in charge of the work order corresponding to the troubleshooting direction can be introduced into the communication group of the target work order to jointly participate in the processing of the work order.

[0063] Of course, when upgrading a work order, you are not limited to introducing only the work order person in charge corresponding to the troubleshooting direction, you can also introduce other persons in charge, and there is no restriction here.

[0064] The online work order processing method provided in this embodiment uses a preset work order large language model to monitor the customer service work order system in a bypass manner to obtain the conversation content generated by the target work order currently being processed by the customer service work order system; wherein the work order large language model uses a labeled work order data set to perform preference learning on a general large language model; the work order large language model uses a chain of thought method to classify and judge the conversation content to determine whether the conversation content meets the category of work order upgrade processing required; when the conversation content meets the category of work order upgrade processing required, the customer service work order system is automatically triggered to upgrade the target work order. In this embodiment, the large language model can be used to automatically and accurately determine whether a customer service work order needs to be upgraded, thereby avoiding delays in work order processing due to failure to accurately or timely upgrade the work order, and no manual control or maintenance rules are required, thereby improving accuracy and reducing costs.

[0065] Based on any of the above embodiments, in order to facilitate the client who newly joins the conversation after the work order is upgraded to quickly understand the situation of the target work order, such as the problem phenomenon to be solved, the direction of investigation, the progress of investigation, the conclusion of the investigation, the customers affected by the problem phenomenon, etc., the content of the target work order can be summarized. Figure 3 As shown, the specific details can be as follows:

[0066] S301, according to the structured data in the conversation content and the association relationship between the conversation contents based on the work order large language model, extract all topics in the conversation content, and summarize the troubleshooting process of the target work order based on each topic;

[0067] S302: Analyze the meta information of the target work order and / or the conversation content through the work order large language model to obtain the problem phenomenon and troubleshooting conclusion of the target work order;

[0068] S303: Generate a content summary of the target work order based on the problem phenomenon, troubleshooting process and troubleshooting conclusion of the target work order, and display the content summary of the target work order in the communication group of the target work order in the customer service work order system.

[0069] In this embodiment, the work order language model can be trained in advance, or another model can be trained to have the ability to summarize the content of the work order, especially to summarize one or more information of the work order, such as the phenomenon of the problem to be solved, the direction of investigation, the progress of investigation, the conclusion of investigation, and the customers affected by the phenomenon of the problem to be solved. The investigation direction is the content that needs to be investigated when finding the abnormal cause of the problem to be solved, and different investigation directions may involve the participation of different customer services; the investigation progress can be used to show whether the investigation of the investigation direction has been completed, or the progress of completion, etc.; the investigation conclusion is used to show whether there is a problem in the investigation direction; the customers affected by the phenomenon of the problem to be solved are used to show which specific customers are affected by the problem phenomenon, which can reflect the scope of impact or the severity of the problem, and can also be used for targeted repairs.

[0070] The following is an example of the work order language model's ability to summarize content. The work order language model can extract all topics in the conversation content based on the structured data in the conversation content and the association between the conversation content determined by the work order language model, where the structured data is the conversation content with a clear association. For example, if a conversation content refers to the previous conversation content, it means that the two conversation contents have a clear association; and summarize the troubleshooting process of the target work order based on each topic, where the troubleshooting process is clearly recorded in the conversation content. For example, the troubleshooting process for the CPU can specifically include the troubleshooting time, troubleshooting content, troubleshooting progress, troubleshooting conclusions, etc., and the work order language model only needs to identify and organize according to the topic category.

[0071] In addition, the meta-information of the target work order may also include some information, such as the source of the problem, problem phenomenon, troubleshooting conclusion, etc., and the conversation content may also include some problem sources, problem phenomena, troubleshooting conclusions, etc. Therefore, the meta-information and / or conversation content of the target work order can be analyzed through the work order large language model to obtain the problem phenomenon and troubleshooting conclusion of the target work order.

[0072] Finally, a content summary of the target work order can be generated based on the problem phenomenon, troubleshooting process and troubleshooting conclusion of the target work order, and the content summary of the target work order can be displayed in the communication group of the target work order in the customer service work order system, or it can be directly sent to the customer service end that has newly joined the communication group, so as to quickly understand the specific situation and progress of the target work order.

[0073] In addition, based on the above embodiment, the affected customers of the target work order can also be determined by the work order language model, which can be specifically as follows:

[0074] Performing character string matching on the preset customer names included in the preset customer set and the conversation content, and determining customer names that hit the preset customer set from all the conversation contents as a candidate set;

[0075] Determine whether there is a superset relationship between the customer names in the candidate set, and if so, delete the customer names as subsets from the candidate set;

[0076] The work order large language model is used to determine whether the customer in the candidate set is affected by the target work order based on a few-sample learning method, and the customer name of the affected customer is returned as the affected customer of the target work order.

[0077] In this embodiment, a preset customer set can be obtained first, wherein the preset customer set may include multiple preset customers, or may also include all customers, and customer names can be retrieved in the conversation content according to the preset customer set to determine the customer names included in the conversation content to form a candidate set, wherein the customer name retrieval process can be implemented by any retrieval method without the participation of the work order large language model, so as to reduce the pressure on the work order large language model and reduce the number of customers that need more attention. The work order large language model only needs to determine the target customers affected by the phenomenon to be solved from the candidate set according to the conversation content.

[0078] In this embodiment, a string matching method can be used to retrieve customer names in the conversation content, that is, for any customer name included in the preset customer set, it is determined whether the customer name appears in the conversation content. If it appears, it is determined to be a matching customer, and all matching customers constitute a set of matching customers, that is, a candidate set; however, considering that if a customer name (recorded as the first customer name) is a subset of another customer name (recorded as the second customer name), for example, the first customer name is "AB" and the second customer name is "ABC", if the second customer name appears in the conversation content, then the first customer name must also appear in the conversation content, and usually the first customer name is the abbreviation of the second customer name, in order to avoid duplication, in this embodiment, it can be determined whether there is a subset in the candidate set in which the first customer name is the second customer name. If so, the full name is retained and the abbreviation is deleted, that is, the second customer name is retained and the first customer name is deleted, and the remaining customers in the set of matching customers after the deletion are determined as candidate customers, and then the work order large language model determines the target customers affected by the problem to be solved from the candidate customers according to the conversation content, thereby reducing the waste of resources of repeated customers, and enabling the work order large language model to learn through a few samples (Few-Shot Learning) method can quickly identify the target customers affected by the problem to be solved, reducing the amount of data to be processed and improving processing efficiency.

[0079] Furthermore, when generating the content summary of the target work order, the content summary of the target work order is generated according to the problem phenomenon, troubleshooting process, troubleshooting conclusion and affected customers of the target work order, so that the content summary is more complete and rich.

[0080] On the basis of any of the above embodiments, before the work order large language model is used to classify and judge the conversation content in a chain of thought manner to determine whether the conversation content meets the category that requires work order upgrade processing, it can also be determined whether the target work order has been upgraded. If it is determined that the target work order has not been upgraded, the work order large language model can be used to classify and judge the conversation content in a chain of thought manner; otherwise, if it is determined that the target work order has been upgraded, there is no need to classify and judge the conversation content in a chain of thought manner through the work order large language model. At this time, due to the progress of the conversation, the target work order may have added a new troubleshooting direction, and the customer service end corresponding to the new troubleshooting direction needs to be added to the conversation. Whether the target work order has added a new troubleshooting direction can be determined by the work order large language model, that is, the work order large language model is used to analyze the conversation content to determine whether the target work order has added a new troubleshooting direction. If it is determined that the target work order has added a new troubleshooting direction, the customer service work order system is automatically triggered to introduce the work order person in charge corresponding to the new troubleshooting direction into the communication group for the target work order to continue processing. The above scheme can realize that before the target work order is upgraded, the work order large language model can be used to judge the work order upgrade multiple times, and there is no need to judge the work order upgrade again after the target work order is upgraded. The timing of the work order upgrade can be accurately determined, and unnecessary calculation of the work order large language model can be avoided to waste time and computing resources.

[0081] Corresponding to the online work order processing method in the above embodiment, Figure 4 This is a structural block diagram of an online work order processing device provided by an embodiment of the present disclosure. For ease of explanation, only the parts related to the embodiment of the present disclosure are shown. Figure 4 The online work order processing device 400 includes: an acquisition unit 401, a judgment unit 402, and an upgrade unit 403.

[0082] The acquisition unit 401 is used to monitor the customer service work order system in a bypass mode through a preset work order large language model to obtain the conversation content generated by the target work order currently being processed by the customer service work order system; wherein the work order large language model is generated by preference learning of a general large language model using a work order data set with labels;

[0083] A judgment unit 402 is used to classify and judge the conversation content by using the work order large language model in a thought chain manner to determine whether the conversation content meets the category that requires work order upgrade processing;

[0084] The upgrade unit 403 is used to automatically trigger the customer service work order system to upgrade the target work order when the conversation content meets the category of the work order that needs to be upgraded.

[0085] In one or more embodiments of the present disclosure, the device further includes a summarizing unit 404, configured to:

[0086] According to the structured data in the conversation content and the correlation between the conversation contents based on the work order large language model, all topics in the conversation content are extracted, and the troubleshooting process of the target work order is summarized based on each topic;

[0087] Analyze the meta information of the target work order and / or the conversation content through the work order large language model to obtain the problem phenomenon and troubleshooting conclusion of the target work order;

[0088] A content summary of the target work order is generated based on the problem phenomenon, troubleshooting process and troubleshooting conclusion of the target work order, and the content summary of the target work order is displayed in the communication group of the target work order in the customer service work order system.

[0089] In one or more embodiments of the present disclosure, the summarizing unit 403 is further configured to:

[0090] Performing character string matching on the preset customer names included in the preset customer set and the conversation content, and determining customer names that hit the preset customer set from all the conversation contents as a candidate set;

[0091] Determine whether there is a superset relationship between the customer names in the candidate set, and if so, delete the customer names as subsets from the candidate set;

[0092] The work order large language model is used to determine whether the customer in the candidate set is affected by the target work order based on a few-sample learning method, and the customer name of the affected customer is returned as the affected customer of the target work order.

[0093] In one or more embodiments of the present disclosure, when the summarizing unit 403 generates a content summary of the target work order according to the problem phenomenon, troubleshooting process, and troubleshooting conclusion of the target work order, it is used to:

[0094] A content summary of the target work order is generated based on the problem phenomenon, troubleshooting process, troubleshooting conclusion and affected customers of the target work order.

[0095] In one or more embodiments of the present disclosure, when the conversation content meets the category of the work order requiring upgrade processing, the upgrade unit 403 automatically triggers the customer service work order system to upgrade the target work order, and is used to:

[0096] When the content of the conversation meets the category of work orders that need to be upgraded, the work order large language model is used to analyze the content of the conversation, determine the troubleshooting direction involved in the target work order, and automatically trigger the customer service work order system to introduce the work order person in charge corresponding to the troubleshooting direction into the communication group for the target work order to continue processing.

[0097] In one or more embodiments of the present disclosure, when the judgment unit 402 classifies and judges the conversation content by using the work order large language model in a thought chain manner to determine whether the conversation content meets the category requiring work order escalation processing, it is used to:

[0098] An input instruction for constructing the work order large language model according to the conversation content, wherein the input instruction is used to instruct the work order large language model to determine the topic category of the conversation content according to a preset classification rule, and output an analysis process and a classification basis;

[0099] Outputting the topic category of the conversation content, the analysis process and the classification basis according to the input instruction through the work order large language model;

[0100] Determine whether the topic category of the conversation content belongs to a preset topic category that requires work order escalation processing. If so, determine whether the conversation content meets the category that requires work order escalation processing.

[0101] In one or more embodiments of the present disclosure, when the judgment unit 402 classifies and judges the conversation content by using the work order large language model in a thought chain manner to determine whether the conversation content meets the category requiring work order escalation processing, it is used to:

[0102] Determine whether the target work order has been upgraded. If it is determined that the target work order has not been upgraded, classify and judge the conversation content using the work order large language model in a thinking chain manner to determine whether the conversation content meets the category that requires work order upgrade processing.

[0103] In one or more embodiments of the present disclosure, the determining unit 402 is further configured to:

[0104] If it is determined that the target work order has been upgraded, the conversation content is analyzed by the work order large language model to determine whether a new troubleshooting direction has been added to the target work order;

[0105] If it is determined that a new troubleshooting direction is added to the target work order, the customer service work order system is automatically triggered to introduce the work order person in charge corresponding to the new troubleshooting direction into the communication group of the target work order to continue processing.

[0106] The device provided in this embodiment can be used to execute the technical solution of the above method embodiment. Its implementation principle and technical effect are similar, and this embodiment will not be repeated here.

[0107] In order to implement the above embodiment, the embodiment of the present disclosure also provides an electronic device.

[0108] refer to Figure 5 , which shows a schematic diagram of the structure of an electronic device 500 suitable for implementing the embodiment of the present disclosure, and the electronic device 500 may be a terminal device or a server. The terminal device may include but is not limited to mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, personal digital assistants (PDAs), tablet computers, portable multimedia players (PMPs), vehicle terminals (such as vehicle navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 5 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0109] like Figure 5 As shown, the electronic device 500 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 501, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 502 or a program loaded from a storage device 508 to a random access memory (RAM) 503. Various programs and data required for the operation of the electronic device 500 are also stored in the RAM 503. The processing device 501, the ROM 502, and the RAM 503 are connected to each other via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0110] Typically, the following devices may be connected to the I / O interface 505: input devices 506 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; output devices 507 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 508 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 509. The communication devices 509 may allow the electronic device 500 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 5The electronic device 500 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0111] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device 509, or installed from a storage device 508, or installed from a ROM 502. When the computer program is executed by the processing device 501, the above-mentioned functions defined in the method of the embodiment of the present disclosure are executed.

[0112] It should be noted that the computer-readable medium disclosed above may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, device or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0113] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0114] The computer-readable medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device executes the method shown in the above embodiment.

[0115] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as an independent software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0116] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0117] The units involved in the embodiments described in the present disclosure may be implemented by software or hardware. The name of a unit does not limit the unit itself in some cases. For example, the first acquisition unit may also be described as a "unit for acquiring at least two Internet Protocol addresses".

[0118] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0119] In a first aspect, according to one or more embodiments of the present disclosure, there is provided an online work order processing method, comprising:

[0120] An acquisition unit is used to monitor the customer service work order system in a bypass mode through a preset work order large language model to obtain the conversation content generated by the target work order currently being processed by the customer service work order system; wherein the work order large language model is generated by preference learning of a general large language model using a work order data set with labels;

[0121] A judgment unit, configured to classify and judge the conversation content by using the work order large language model in a thought chain manner, and determine whether the conversation content meets the category requiring work order upgrade processing;

[0122] The upgrade unit is used to automatically trigger the customer service work order system to upgrade the target work order when the content of the conversation meets the category of the work order that needs to be upgraded.

[0123] According to one or more embodiments of the present disclosure, the method further includes:

[0124] According to the structured data in the conversation content and the correlation between the conversation contents based on the work order large language model, all topics in the conversation content are extracted, and the troubleshooting process of the target work order is summarized based on each topic;

[0125] Analyze the meta information of the target work order and / or the conversation content through the work order large language model to obtain the problem phenomenon and troubleshooting conclusion of the target work order;

[0126] A content summary of the target work order is generated based on the problem phenomenon, troubleshooting process and troubleshooting conclusion of the target work order, and the content summary of the target work order is displayed in the communication group of the target work order in the customer service work order system.

[0127] According to one or more embodiments of the present disclosure, the method further includes:

[0128] Performing character string matching on the preset customer names included in the preset customer set and the conversation content, and determining customer names that hit the preset customer set from all the conversation contents as a candidate set;

[0129] Determine whether there is a superset relationship between the customer names in the candidate set, and if so, delete the customer names as subsets from the candidate set;

[0130] The work order large language model is used to determine whether the customer in the candidate set is affected by the target work order based on a few-sample learning method, and the customer name of the affected customer is returned as the affected customer of the target work order.

[0131] According to one or more embodiments of the present disclosure, generating a content summary of the target work order according to the problem phenomenon, troubleshooting process and troubleshooting conclusion of the target work order includes:

[0132] A content summary of the target work order is generated based on the problem phenomenon, troubleshooting process, troubleshooting conclusion and affected customers of the target work order.

[0133] According to one or more embodiments of the present disclosure, when the conversation content meets the category of the work order requiring escalation processing, automatically triggering the customer service work order system to perform escalation processing on the target work order includes:

[0134] When the content of the conversation meets the category of work orders that need to be upgraded, the work order large language model is used to analyze the content of the conversation, determine the troubleshooting direction involved in the target work order, and automatically trigger the customer service work order system to introduce the work order person in charge corresponding to the troubleshooting direction into the communication group for the target work order to continue processing.

[0135] According to one or more embodiments of the present disclosure, the process of classifying and judging the conversation content by using the work order large language model in a thought chain manner to determine whether the conversation content meets the category requiring work order upgrade processing includes:

[0136] An input instruction for constructing the work order large language model according to the conversation content, wherein the input instruction is used to instruct the work order large language model to determine the topic category of the conversation content according to a preset classification rule, and output an analysis process and a classification basis;

[0137] Outputting the topic category of the conversation content, the analysis process and the classification basis according to the input instruction through the work order large language model;

[0138] Determine whether the topic category of the conversation content belongs to a preset topic category that requires work order escalation processing. If so, determine whether the conversation content meets the category that requires work order escalation processing.

[0139] According to one or more embodiments of the present disclosure, the process of classifying and judging the conversation content by using the work order large language model in a thought chain manner to determine whether the conversation content meets the category requiring work order upgrade processing includes:

[0140] Determine whether the target work order has been upgraded. If it is determined that the target work order has not been upgraded, classify and judge the conversation content using the work order large language model in a thinking chain manner to determine whether the conversation content meets the category that requires work order upgrade processing.

[0141] According to one or more embodiments of the present disclosure, the method further includes:

[0142] If it is determined that the target work order has been upgraded, the conversation content is analyzed by the work order large language model to determine whether a new troubleshooting direction has been added to the target work order;

[0143] If it is determined that a new troubleshooting direction is added to the target work order, the customer service work order system is automatically triggered to introduce the work order person in charge corresponding to the new troubleshooting direction into the communication group of the target work order to continue processing.

[0144] In a second aspect, according to one or more embodiments of the present disclosure, there is provided an online work order processing device, including:

[0145] An acquisition unit is used to monitor the customer service work order system in a bypass mode through a preset work order large language model to obtain the conversation content generated by the target work order currently being processed by the customer service work order system; wherein the work order large language model is generated by preference learning of a general large language model using a work order data set with labels;

[0146] A judgment unit, configured to classify and judge the conversation content by using the work order large language model in a thought chain manner, and determine whether the conversation content meets the category requiring work order upgrade processing;

[0147] The upgrade unit is used to automatically trigger the customer service work order system to upgrade the target work order when the content of the conversation meets the category of the work order that needs to be upgraded.

[0148] According to one or more embodiments of the present disclosure, the device further includes a summarizing unit, configured to:

[0149] According to the structured data in the conversation content and the correlation between the conversation contents based on the work order large language model, all topics in the conversation content are extracted, and the troubleshooting process of the target work order is summarized based on each topic;

[0150] Analyze the meta information of the target work order and / or the conversation content through the work order large language model to obtain the problem phenomenon and troubleshooting conclusion of the target work order;

[0151] A content summary of the target work order is generated based on the problem phenomenon, troubleshooting process and troubleshooting conclusion of the target work order, and the content summary of the target work order is displayed in the communication group of the target work order in the customer service work order system.

[0152] According to one or more embodiments of the present disclosure, the summarizing unit is further configured to:

[0153] Performing character string matching on the preset customer names included in the preset customer set and the conversation content, and determining customer names that hit the preset customer set from all the conversation contents as a candidate set;

[0154] Determine whether there is a superset relationship between the customer names in the candidate set, and if so, delete the customer names as subsets from the candidate set;

[0155] The work order large language model is used to determine whether the customer in the candidate set is affected by the target work order based on a few-sample learning method, and the customer name of the affected customer is returned as the affected customer of the target work order.

[0156] According to one or more embodiments of the present disclosure, when the summary unit generates a content summary of the target work order according to the problem phenomenon, troubleshooting process and troubleshooting conclusion of the target work order, it is used to:

[0157] A content summary of the target work order is generated based on the problem phenomenon, troubleshooting process, troubleshooting conclusion and affected customers of the target work order.

[0158] According to one or more embodiments of the present disclosure, when the conversation content meets the category of the work order requiring upgrade processing, the upgrade unit automatically triggers the customer service work order system to upgrade the target work order, and is used to:

[0159] When the content of the conversation meets the category of work orders that need to be upgraded, the work order large language model is used to analyze the content of the conversation, determine the troubleshooting direction involved in the target work order, and automatically trigger the customer service work order system to introduce the work order person in charge corresponding to the troubleshooting direction into the communication group for the target work order to continue processing.

[0160] According to one or more embodiments of the present disclosure, when the judgment unit classifies and judges the conversation content by using the work order large language model in a thought chain manner to determine whether the conversation content meets the category requiring work order escalation processing, it is used to:

[0161] An input instruction for constructing the work order large language model according to the conversation content, wherein the input instruction is used to instruct the work order large language model to determine the topic category of the conversation content according to a preset classification rule, and output an analysis process and a classification basis;

[0162] Outputting the topic category of the conversation content, the analysis process and the classification basis according to the input instruction through the work order large language model;

[0163] Determine whether the topic category of the conversation content belongs to a preset topic category that requires work order escalation processing. If so, determine whether the conversation content meets the category that requires work order escalation processing.

[0164] According to one or more embodiments of the present disclosure, when the judgment unit classifies and judges the conversation content by using the work order large language model in a thought chain manner to determine whether the conversation content meets the category requiring work order escalation processing, it is used to:

[0165] Determine whether the target work order has been upgraded. If it is determined that the target work order has not been upgraded, classify and judge the conversation content using the work order large language model in a thinking chain manner to determine whether the conversation content meets the category that requires work order upgrade processing.

[0166] According to one or more embodiments of the present disclosure, the judging unit is further configured to:

[0167] If it is determined that the target work order has been upgraded, the conversation content is analyzed by the work order large language model to determine whether a new troubleshooting direction has been added to the target work order;

[0168] If it is determined that a new troubleshooting direction is added to the target work order, the customer service work order system is automatically triggered to introduce the work order person in charge corresponding to the new troubleshooting direction into the communication group of the target work order to continue processing.

[0169] In a third aspect, according to one or more embodiments of the present disclosure, there is provided an electronic device, comprising: at least one processor and a memory;

[0170] The memory stores computer-executable instructions;

[0171] The at least one processor executes the computer-executable instructions stored in the memory, so that the at least one processor executes the online work order processing method as described in the first aspect and various possible designs of the first aspect.

[0172] In a fourth aspect, according to one or more embodiments of the present disclosure, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer execution instructions. When a processor executes the computer execution instructions, the online work order processing method described in the first aspect and various possible designs of the first aspect is implemented.

[0173] In a fifth aspect, according to one or more embodiments of the present disclosure, a computer program product is provided, including a computer program, wherein when the computer program is executed by a processor, the online work order processing method described in the first aspect and various possible designs of the first aspect is implemented.

[0174] The above description is only a preferred embodiment of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present disclosure is not limited to the technical solutions formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in the present disclosure (but not limited to) by each other to form a technical solution.

[0175] In addition, although each operation is described in a specific order, this should not be understood as requiring these operations to be performed in the specific order shown or in a sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although some specific implementation details are included in the above discussion, these should not be interpreted as limiting the scope of the present disclosure. Some features described in the context of a separate embodiment can also be implemented in a single embodiment in combination. On the contrary, the various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination mode.

[0176] Although the subject matter has been described in language specific to structural features and / or methodological logical actions, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. On the contrary, the specific features and actions described above are merely example forms of implementing the claims.

Claims

1. An online work order processing method, characterized in that: include: The customer service work order system is monitored in a bypass mode through a preset work order large language model to obtain the conversation content generated by the target work order currently being processed by the customer service work order system; wherein the work order large language model is generated by preference learning of a general large language model using a labeled work order data set; The work order large language model is used to classify and judge the conversation content in a chain of thought manner to determine whether the conversation content meets the category that requires work order upgrade processing; When the content of the conversation meets the category of the work order that needs to be upgraded, the customer service work order system is automatically triggered to upgrade the target work order.

2. The method according to claim 1, characterized in that The method further comprises: According to the structured data in the conversation content and the correlation between the conversation contents based on the work order large language model, all topics in the conversation content are extracted, and the troubleshooting process of the target work order is summarized based on each topic; Analyze the meta information of the target work order and / or the conversation content through the work order large language model to obtain the problem phenomenon and troubleshooting conclusion of the target work order; A content summary of the target work order is generated based on the problem phenomenon, troubleshooting process and troubleshooting conclusion of the target work order, and the content summary of the target work order is displayed in the communication group of the target work order in the customer service work order system.

3. The method according to claim 2, characterized in that The method further comprises: Performing character string matching on the preset customer names included in the preset customer set and the conversation content, and determining customer names that hit the preset customer set from all the conversation contents as a candidate set; Determine whether there is a superset relationship between the customer names in the candidate set, and if so, delete the customer names as subsets from the candidate set; The work order large language model is used to determine whether the customer in the candidate set is affected by the target work order based on a few-sample learning method, and the customer name of the affected customer is returned as the affected customer of the target work order.

4. The method according to claim 3, characterized in that Generating a content summary of the target work order according to the problem phenomenon, troubleshooting process and troubleshooting conclusion of the target work order includes: A content summary of the target work order is generated based on the problem phenomenon, troubleshooting process, troubleshooting conclusion and affected customers of the target work order.

5. The method according to claim 1, characterized in that When the content of the conversation meets the category of the work order that needs to be upgraded, the customer service work order system is automatically triggered to upgrade the target work order, including: When the content of the conversation meets the category of work orders that need to be upgraded, the work order large language model is used to analyze the content of the conversation, determine the troubleshooting direction involved in the target work order, and automatically trigger the customer service work order system to introduce the work order person in charge corresponding to the troubleshooting direction into the communication group for the target work order to continue processing.

6. The method according to claim 1, characterized in that The process of classifying and judging the conversation content by using the work order large language model in a thought chain manner to determine whether the conversation content meets the category requiring work order upgrade processing includes: An input instruction for constructing the work order large language model according to the conversation content, wherein the input instruction is used to instruct the work order large language model to determine the topic category of the conversation content according to a preset classification rule, and output an analysis process and a classification basis; Outputting the topic category of the conversation content, the analysis process and the classification basis according to the input instruction through the work order large language model; Determine whether the topic category of the conversation content belongs to a preset topic category that requires work order escalation processing. If so, determine whether the conversation content meets the category that requires work order escalation processing.

7. The method according to claim 1, characterized in that The process of classifying and judging the conversation content by using the work order large language model in a thought chain manner to determine whether the conversation content meets the category requiring work order upgrade processing includes: Determine whether the target work order has been upgraded. If it is determined that the target work order has not been upgraded, classify and judge the conversation content using the work order large language model in a thinking chain manner to determine whether the conversation content meets the category that requires work order upgrade processing.

8. The method according to claim 7, characterized in that The method further comprises: If it is determined that the target work order has been upgraded, the conversation content is analyzed by the work order large language model to determine whether a new troubleshooting direction has been added to the target work order; If it is determined that a new troubleshooting direction is added to the target work order, the customer service work order system is automatically triggered to introduce the work order person in charge corresponding to the new troubleshooting direction into the communication group of the target work order to continue processing.

9. An online work order processing device, characterized in that: include: An acquisition unit is used to monitor the customer service work order system in a bypass mode through a preset work order large language model to obtain the conversation content generated by the target work order currently being processed by the customer service work order system; wherein the work order large language model is generated by preference learning of a general large language model using a work order data set with labels; A judgment unit, configured to classify and judge the conversation content by using the work order large language model in a thought chain manner, and determine whether the conversation content meets the category requiring work order upgrade processing; The upgrade unit is used to automatically trigger the customer service work order system to upgrade the target work order when the content of the conversation meets the category of the work order that needs to be upgraded.

10. An electronic device, characterized in that: include: Processor and memory; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory, so that the processor performs the method according to any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and when the processor executes the computer-executable instructions, the method according to any one of claims 1 to 8 is implemented.

12. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Images