Business scenario identification methods, devices, equipment and storage media

Abstract

This invention relates to the field of scene recognition technology, and discloses a business scene recognition method, apparatus, device, and storage medium. The method includes: upon receiving a question inquiry request from a target user, acquiring the user's question and background information; determining a set of questions to be tested based on the user's background information and the channel through which the user inquired, and determining a set of scene-related information based on a scene information score table; obtaining a set of valid background scores based on the set of questions to be tested and the set of scene-related information; and determining the target business scene corresponding to the user's question based on the set of valid background scores. This invention obtains a set of valid background scores based on the set of questions to be tested and the set of scene-related information, and determines the target business scene corresponding to the target user's question based on the set of valid background scores, thereby solving the technical problem in existing technologies where manual processing of user inquiries cannot quickly locate the scene to which the question belongs, resulting in the inability to provide personalized services to customers.

Description

Technical Field

[0001] This invention relates to the field of scene recognition technology, and in particular to a business scene recognition method, apparatus, device, and storage medium. Background Technology

[0002] Business scenario identification refers to recognizing the business scenario a user is in based on user data. Taking the financial sector as an example, a user's business scenario could be a conventional one such as payment, transfer, or quick loan, or it could be a scenario like fault repair or service consultation. For enterprise-level services, user questions may involve multiple different business scenarios. If the user's question is not accurately categorized into the corresponding scenario, it may lead to inaccurate answers or delays in processing. Currently, similar products or technologies typically require manual processing of a large number of similar or repetitive questions. This method cannot quickly locate the scenario to which a user's question pertains, thus failing to provide personalized services.

[0003] The above content is only used to help understand the technical solution of the present invention and does not represent an admission that the above content is prior art. Summary of the Invention

[0004] The main objective of this invention is to provide a business scenario identification method, apparatus, device, and storage medium, which aims to solve the technical problem that existing technologies cannot quickly locate the scenario to which the question belongs by manually processing user questions, thus making it impossible to provide personalized services to customers.

[0005] To achieve the above objectives, the present invention provides a business scenario identification method, the business scenario identification method comprising:

[0006] Upon receiving a question inquiry request from a target user, obtain the user's question and user background information;

[0007] Based on the user background information and the channels through which the user asked questions, a set of questions to be tested is determined, and based on the scene information score table, a set of scene-related information is determined;

[0008] A set of valid background scores is obtained based on the set of questions to be tested and the set of scene-related information.

[0009] The target business scenario corresponding to the user's question is determined based on the set of valid background scores.

[0010] Optionally, before the step of obtaining the user's question and background information when receiving a question inquiry request from the target user, the method further includes:

[0011] The operation configuration platform determines the service channels, associated users, and keyword sets corresponding to all business scenarios.

[0012] A scenario information score table is constructed based on the service channels, the associated users, and the keyword set.

[0013] Optionally, after the step of constructing a scenario information score table based on the service channel, the associated user, and the keyword set, the method further includes:

[0014] Obtain the weight information of each score table object in the scene information score table in the corresponding business scene;

[0015] The step of determining the set of scene-related information based on the scene information score table includes:

[0016] According to the business scenario, the group chat information, user information, keywords and the weight information are classified and merged from the scenario information score table to obtain a scenario-related information set.

[0017] Optionally, after the step of determining the target business scenario corresponding to the user question based on the set of valid background scores, the method further includes:

[0018] Determine the target knowledge base corresponding to the target business scenario;

[0019] The answer to the user's question is obtained from the target knowledge base.

[0020] The number of incorrect answers in the question's answer set is obtained, and the number of incorrect responses is determined based on the number of incorrect answers, so that the evaluation operation can be performed based on the number of incorrect responses.

[0021] Optionally, after the step of obtaining the number of incorrect answers in the question's answers and determining the number of incorrect responses based on the number of incorrect answers, so as to perform the evaluation operation based on the number of incorrect responses, the method further includes:

[0022] When the number of incorrect responses does not exceed a preset threshold for the number of incorrect responses, high-frequency questions are generated by high-frequency word matching and the high-frequency questions are output to the target user terminal.

[0023] Optionally, after the step of obtaining the number of incorrect answers in the question's answers and determining the number of incorrect responses based on the number of incorrect answers, so as to perform the evaluation operation based on the number of incorrect responses, the method further includes:

[0024] When the number of incorrect responses exceeds a preset threshold, a decision is made based on the preset urgency level of the scenario to determine whether to transfer the call to a human agent.

[0025] If so, create a target group chat and associate the target group chat with the user's question;

[0026] Once the association is complete, the manual operation is switched through the target group chat.

[0027] Optionally, the step of determining the target business scenario corresponding to the user question based on the set of valid background scores includes:

[0028] The business scenario for obtaining the target element corresponding to the extreme value of the score in the set of valid background scores;

[0029] The business scenario corresponding to the extreme value of the score is determined as the target business scenario.

[0030] Furthermore, to achieve the above objectives, the present invention also proposes a business scenario identification device, the device comprising:

[0031] The information acquisition module is used to acquire the user's question and background information when it receives a question inquiry request sent by the target user.

[0032] The set determination module is used to determine the set of questions to be tested based on the user background information and the user's questioning channels, and to determine the set of scene-related information based on the scene information score table;

[0033] The set determination module is further configured to obtain a set of valid background scores based on the set of questions to be tested and the set of scene association information.

[0034] The business scenario determination module is used to determine the target business scenario corresponding to the user question based on the set of valid background scores.

[0035] In addition, to achieve the above objectives, the present invention also proposes a business scenario recognition device, the device comprising: a memory, a processor, and a business scenario recognition program stored in the memory and executable on the processor, the business scenario recognition program being configured to implement the steps of the business scenario recognition method described above.

[0036] In addition, to achieve the above objectives, the present invention also proposes a storage medium storing a business scenario identification program, wherein the business scenario identification program, when executed by a processor, implements the steps of the business scenario identification method described above.

[0037] This invention discloses a method for obtaining the user's question and background information upon receiving a question inquiry request from a target user; determining a set of questions to be tested based on the user's background information and the channel through which the user inquired; determining a set of scenario-related information based on a scenario information score table; obtaining a set of valid background scores based on the set of questions to be tested and the set of scenario-related information; and determining the target business scenario corresponding to the user's question based on the set of valid background scores. Because this invention obtains a set of valid background scores based on the set of questions to be tested and the set of scenario-related information, and determines the target business scenario corresponding to the target user's question based on the set of valid background scores, it solves the technical problem in existing technologies where manual processing of user inquiries cannot quickly locate the scenario to which the question belongs, thus preventing the provision of personalized services to customers. Attached Figure Description

[0038] Figure 1 This is a schematic diagram of the business scenario recognition device for the hardware operating environment involved in the embodiments of the present invention;

[0039] Figure 2 This is a flowchart illustrating the first embodiment of the business scenario identification method of the present invention;

[0040] Figure 3 This is a flowchart illustrating the second embodiment of the business scenario identification method of the present invention;

[0041] Figure 4 This is a schematic diagram of the user support process in the second embodiment of the business scenario identification method of the present invention;

[0042] Figure 5 This is a structural block diagram of the first embodiment of the business scenario recognition device of the present invention.

[0043] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0044] It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the invention.

[0045] Reference Figure 1 , Figure 1 This is a schematic diagram of the business scenario identification device structure in the hardware operating environment involved in the embodiments of the present invention.

[0046] like Figure 1As shown, the service scenario identification device may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. The communication bus 1002 is used to enable communication between these components. The user interface 1003 may include a display screen or an input unit such as a keyboard; optionally, the user interface 1003 may also include a standard wired interface or a wireless interface. The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a Wireless-Fidelity (Wi-Fi) interface). The memory 1005 may be high-speed random access memory (RAM) or stable non-volatile memory (NVM), such as a disk storage device. Optionally, the memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0047] Those skilled in the art will understand that Figure 1 The structure shown does not constitute a limitation on the business scenario recognition device. It may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0048] like Figure 1 As shown, the memory 1005, which serves as a storage medium, may include an operating system, a network communication module, a user interface module, and a business scenario identification program.

[0049] exist Figure 1 In the business scenario recognition device shown, the network interface 1004 is mainly used for data communication with the network server; the user interface 1003 is mainly used for data interaction with the user; the processor 1001 and the memory 1005 in the business scenario recognition device of the present invention can be set in the business scenario recognition device, and the business scenario recognition device calls the business scenario recognition program stored in the memory 1005 through the processor 1001 and executes the business scenario recognition method provided in the embodiment of the present invention.

[0050] This invention provides a business scenario identification method, referring to... Figure 2 , Figure 2 This is a flowchart illustrating the first embodiment of the business scenario identification method of the present invention.

[0051] In this embodiment, the business scenario identification method includes the following steps:

[0052] Step S10: Upon receiving a question inquiry request from a target user, obtain the user's question and background information.

[0053] It should be noted that the execution subject of the method in this embodiment can be a chatbot with data processing capabilities that identifies the business scenario to which the user's question belongs, or other business scenario identification devices that include the chatbot and can achieve the same or similar functions. Here, the business scenario identification methods provided in this embodiment and the following embodiments will be specifically described using a business scenario identification device (hereinafter referred to as the device).

[0054] It should be understood that the aforementioned inquiry request can be triggered by a target user when asking a question. The target user can be any user who asks a question to the device; this embodiment does not impose any restrictions on this.

[0055] It is understood that the aforementioned target question is the question raised by the target user to the device. In this embodiment, there are no restrictions on the specific way in which the target user raises the question to the device.

[0056] It should be noted that the aforementioned user background information can be the target user's personal information (such as user name, user ID) or browsing information, etc., and this embodiment does not impose any restrictions on this. In this embodiment, each user has a corresponding user ID, so background information associated with the user, such as work order information and browsing history, can be queried through the user ID.

[0057] In practice, target users can send questions to the chatbot privately via the client or call the chatbot in a group chat to trigger a question request. After receiving the question request, the device (or chatbot) can parse the request message body, including the message ID, message type, message sending time, target user's user identifier (such as user ID), target user's group ID, and message content. The message body content is then structured and inserted into the backend database. At the same time, the target user's background information, including past work order information and browsing history, can be queried through the target user's user ID. At this point, the user's question and background information can be obtained.

[0058] Step S20: Determine the set of questions to be tested based on the user background information and the user's questioning channels, and determine the set of scene-related information based on the scene information score table.

[0059] It should be understood that the aforementioned user question channels can be group chats where target users can ask questions. In this embodiment, multiple group chats can be set up for users to ask questions in each scenario.

[0060] Understandably, the aforementioned set of questions to be tested can be a collection of background information provided by the user and keywords contained in the questions. In practical applications, the device can filter the background information provided by the user and the keywords contained in the questions to obtain the keywords to be tested, and integrate the user's personal information, group chat information or private chat information of the question, historical work order information, operation records, and keywords to be tested, and mark them as the set of questions to be tested.

[0061] It should be noted that the above scenario information score table can be a table that stores information such as service channels, associated users, and keyword sets corresponding to each business scenario.

[0062] It should be understood that the aforementioned set of scenario-related information can be obtained by classifying and merging the scenario information score table according to the scenario, including group chat information, user information, keywords, and their corresponding weights.

[0063] Furthermore, in order to pre-build a scenario information score table to improve data processing efficiency, before step S10, the method further includes: determining the service channels, associated users, and keyword sets corresponding to all business scenarios through the operation configuration platform; and building a scenario information score table based on the service channels, associated users, and keyword sets.

[0064] It is understandable that the aforementioned operation configuration platform can be a platform used to configure group chats or user-related information in various scenarios.

[0065] In its implementation, this embodiment provides an operation configuration platform categorized by scenario. For a given question-and-answer service scenario, the platform can configure information for each group chat within that scenario. By adding a group chat ID, the device can automatically query detailed information about associated group chats (i.e., the aforementioned service channels), including group name, group members, and individual member information. Furthermore, by configuring the user ID of associated users through the operation configuration platform, the device can automatically query detailed user information. Finally, the platform integrates the service channels, associated users, and keyword sets corresponding to each scenario, and constructs a scenario information score table based on the service channels, associated users, and keyword sets corresponding to each business scenario. If a scenario-related information set is subsequently needed, it can be directly obtained from the scenario information score table. That is, the scenario information score table categorizes and merges group chat information, user information, keywords, and their corresponding weights according to scenario to obtain a combined scenario-related information, thereby improving data processing efficiency.

[0066] Furthermore, after the step of constructing a scenario information score table based on the service channel, the associated user, and the keyword set, the method further includes: obtaining the weight information of each score table object in the scenario information score table in the corresponding business scenario; the step of determining the scenario-related information set based on the scenario information score table includes: classifying and merging group chat information, user information, keywords, and the weight information from the scenario information score table according to the business scenario to obtain the scenario-related information set.

[0067] It should be noted that the above weight information can be the weight information of the score table object in the score table corresponding to each business scenario in that scenario.

[0068] In its implementation, this embodiment adds the service channels, associated users, and keyword sets corresponding to each business scenario to a scenario information score table. Weights are assigned based on their importance through an operational configuration platform to calculate the optimal solution among all possible matching solutions for the problem to be tested. Specifically, the scenario classification set is denoted as S = {s1, s2, ..., sn}, where each scenario si corresponds to a score table Fi = {fi1, fi2, ..., fim}, where fi represents the weight of the i-th score table object in that scenario. This embodiment quickly locates the business scenario among all possible matching solutions for the problem to be tested, allowing subsequent access to the corresponding knowledge base for answers. This ensures users receive more targeted and professional solutions, improving the intelligence and accuracy of the device.

[0069] Step S30: Obtain a set of valid background scores based on the set of questions to be tested and the set of scene-related information.

[0070] It should be understood that the above set of valid background scores can be a set of business scenarios that may correspond to user questions.

[0071] In this specific implementation, the intersection of the set of questions to be tested and the set of scene-related information can be taken as the effective background set. Specifically, if the set of questions to be tested is A and the set of scene-related information is B, then the intersection A∩B can be determined as the effective background score set.

[0072] Step S40: Determine the target business scenario corresponding to the user question based on the set of valid background scores.

[0073] It is understood that the aforementioned target business scenario can be the business scenario corresponding to the user's problem. This embodiment can determine the business scenario corresponding to the user's problem based on the scores of each element in the valid background score set.

[0074] Furthermore, in order to accurately obtain the business scenario corresponding to the user's question, step S40 specifically includes: obtaining the business scenario of the target element corresponding to the extreme value of the score in the set of valid background scores; and determining the business scenario corresponding to the extreme value of the score as the target business scenario.

[0075] It should be noted that the target element mentioned above can be the element with the highest score in the set of valid background scores.

[0076] In the specific implementation, for each element x∈(A∩B) (i.e., the set of valid background scores), if its weight is w(x), then the sum of the weights of each element can obtain the contribution value of that element in the set of valid background scores, that is: score(x)=Σw(x), x∈(A∩B). Assuming the set of valid background scores is C, after the elements in it are sorted by size, we can obtain an ordered list from largest to smallest: c1, c2, ..., cn. At this time, the business scenario corresponding to the highest value c1 can be selected as the target business scenario. In addition, if the scores are the same and the target business scenario cannot be determined, the scenario can be pushed to the user for the user to choose from. At this time, the user can choose the corresponding business scenario according to their own needs or the full scenario.

[0077] This embodiment discloses a method for obtaining the user's question and background information when receiving a question inquiry request from a target user; determining a set of questions to be tested based on the user's background information and the channel through which the user inquires; determining a set of scenario-related information based on a scenario information score table; obtaining a set of valid background scores based on the set of questions to be tested and the set of scenario-related information; and determining the target business scenario corresponding to the user's question based on the set of valid background scores. Because this embodiment obtains the set of valid background scores based on the set of questions to be tested and the set of scenario-related information, and determines the target business scenario corresponding to the target user's question based on the set of valid background scores, it solves the technical problem in existing technologies where manual processing of user inquiries cannot quickly locate the scenario to which the question belongs, thus preventing the provision of personalized services to customers.

[0078] refer to Figure 3 , Figure 3 This is a flowchart illustrating the second embodiment of the business scenario identification method of the present invention.

[0079] Based on the first embodiment described above, in order to evaluate the device's performance, in this embodiment, after step S40, the method further includes:

[0080] Step S50: Determine the target knowledge base corresponding to the target business scenario.

[0081] It should be noted that the aforementioned target knowledge base is the knowledge base used to answer user questions corresponding to the target business scenario.

[0082] Step S60: Obtain the answer to the user's question through the target knowledge base.

[0083] It should be understood that the answers to the above questions can be the answers to user questions stored in the target knowledge base.

[0084] Step S70: Obtain the number of incorrect answers in the question's answer set, and determine the number of incorrect responses based on the number of incorrect answers, so as to perform an evaluation operation based on the number of incorrect responses.

[0085] It is understood that the aforementioned number of incorrect answers can be the number of incorrect answers corresponding to user questions determined by the target knowledge base. Correspondingly, the aforementioned number of incorrect responses can be the number of times incorrect answers to user questions determined by the target knowledge base are found. Specifically, in this embodiment, the number of incorrect answers can be determined as the number of incorrect responses.

[0086] In its implementation, the device can answer user questions based on a pre-built and regularly updated target knowledge base, while simultaneously recording the number of incorrect responses as a basis for evaluating device performance. The target knowledge base can be optimized based on specific user questions, and operators can configure guidance scripts, knowledge base questions and corresponding answers, and related questions for their respective scenarios through the front-end page to provide the optimal answers. Additionally, the device can also record the number of incorrect responses as a basis for system performance evaluation.

[0087] Furthermore, in order to improve the question-answering hit rate, after step S70, the method further includes: when the number of incorrect responses does not exceed a preset threshold for the number of incorrect responses, generating high-frequency questions through high-frequency word matching, and outputting the high-frequency questions to the target user terminal.

[0088] It should be understood that the aforementioned preset threshold for the number of incorrect responses can be a pre-set maximum number of incorrect responses allowed for a user's question.

[0089] It is understandable that the aforementioned target user terminal refers to the terminal corresponding to the target user.

[0090] In its implementation, when the number of incorrect responses does not exceed a preset threshold, this embodiment not only prompts users with high-frequency questions but also proactively displays potentially helpful and feasible additional scripts and related content. High-frequency questions are generated by extracting keywords from all user inquiries using a word frequency-inverse word frequency algorithm and matching high-frequency words. The first round of questions can be answered directly from the knowledge base, while user information, question content, robot response content, and response round number are inserted into the backend database. If the first round does not match, the second round of questions can be queried and pushed. This embodiment can perform automated multi-round inquiries when the number of incorrect responses does not exceed a preset threshold. Based on historical data, the algorithm extracts and matches keywords from all inquiries and generates a list of high-frequency questions, pushing real-time high-frequency questions, related questions, and guiding scripts to help customers better understand their needs and reduce manual intervention, thereby optimizing the customer service experience and reducing the burden on human agents.

[0091] Furthermore, in order to transfer the user to a human agent for processing when the number of incorrect responses exceeds a preset threshold, so as to ensure service quality, after step S70, the method further includes: when the number of incorrect responses exceeds the preset threshold, determining whether to transfer the user to a human agent based on a preset scenario urgency level; if so, creating a target group chat and associating the target group chat with the user's question; and when the association is completed, switching to a human agent through the target group chat.

[0092] It should be noted that the urgency level of the above-mentioned preset scenarios can be the urgency level corresponding to each pre-set business scenario.

[0093] It should be understood that being transferred to a live agent means having your issues handled by a human customer service representative.

[0094] It is understandable that in this embodiment, when the number of incorrect responses exceeds the preset threshold, the user will be transferred to a human agent or a work order will be entered for further follow-up. In this case, even if a situation cannot be resolved, it can be promptly transferred to an experienced human agent with professional knowledge.

[0095] It should be understood that when the number of incorrect responses exceeds a preset threshold, this embodiment can automatically determine whether to transfer the issue to a human agent or enter a work order based on the pre-set urgency and importance of the scenario, as well as the available operational support manpower. For transferring to a human agent, the robot can create a chat room that supports users and user groups (i.e., the aforementioned target chat room) to associate user questions, providing guided responses within the group while offering automated features such as switching to human support, timed escalation reminders, timeout escalation, and automated group closure. For entering work orders, the robot can automatically generate work orders based on user information, question time, and question content, pushing them to back-end operations personnel. Operations then respond asynchronously through back-end batch processing, ensuring that complex user issues are resolved accurately and efficiently. This approach also reduces customer service workload and allows for rapid response to customer needs, largely satisfying the balance between user requirements and corporate efficiency.

[0096] It should be noted that this embodiment can automatically collect all user interaction data through background tracking throughout the question-and-answer process, and analyze and process this data to generate and update metrics such as processing efficiency and quality in real time. These metrics may include response time rate, knowledge base hit rate, effectiveness rate, resolution time, satisfaction rate, and independent resolution rate by frontline staff, thereby outputting a unified service operation view for the service provider.

[0097] In the specific implementation, refer to Figure 4 , Figure 4 This is a schematic diagram of the user support process in the second embodiment of the business scenario identification method of the present invention. Figure 4 As shown, in this embodiment, after a user asks a question, the device can automatically locate the business scenario corresponding to the user's question based on the user's background / information source (such as user name, user ID, group chat information where the question was asked, historical work order records, browsing history, etc.). If the user's business scenario cannot be determined, scenarios can be pushed for the user to select, allowing the user to choose a specific business scenario according to their needs to determine the accurate scenario knowledge base. In addition, this embodiment can also provide a backup of the entire knowledge base. For the accurate scenario knowledge base, it can be determined whether the knowledge base matches the answer to the user's question. If it does not match, it is determined whether it is urgent. If it is urgent, it is transferred to human processing in real time; if it is not urgent, a work order record is entered. In addition, if it does not match, detailed question and answer records can be made, and high-frequency questions can be generated. Then, the quality of question and answer (including reports, number of questions and answers, missing cases, high-frequency questions, user classification, etc.) can be determined, and high-frequency questions, guiding words, and related questions in this scenario can be pushed to the user. For the entire knowledge base, this embodiment can also provide user support based on the above method.

[0098] This embodiment obtains the answers to user questions from a target knowledge base corresponding to the target business scenario, and determines the number of incorrect responses based on the number of incorrect answers. This number is used for evaluation to optimize the equipment. Simultaneously, when the number of incorrect responses does not exceed a preset threshold, high-frequency questions are generated through high-frequency word matching and output to the target user. When the number of incorrect responses exceeds the preset threshold, a decision is made based on the urgency of the scenario to determine whether to transfer the user to a human agent. If so, a target group chat is created and associated with the user question. Once the association is complete, a human agent is selected through the target group chat, thereby reducing customer service pressure and enabling rapid response to user needs.

[0099] Furthermore, this embodiment of the invention also proposes a storage medium storing a business scenario identification program, which, when executed by a processor, implements the steps of the business scenario identification method described above.

[0100] Reference Figure 5 , Figure 5 This is a structural block diagram of the first embodiment of the business scenario recognition device of the present invention.

[0101] like Figure 5 As shown, the business scenario identification device proposed in this embodiment of the invention includes:

[0102] The information acquisition module 501 is used to acquire the user's question and user background information when it receives a question inquiry request sent by the target user.

[0103] The set determination module 502 is used to determine the set of questions to be tested based on the user background information and the user's questioning channels, and to determine the set of scene-related information based on the scene information score table;

[0104] The set determination module 502 is further configured to obtain a set of valid background scores based on the set of questions to be tested and the set of scene association information.

[0105] The business scenario determination module 503 is used to determine the target business scenario corresponding to the user question based on the set of valid background scores.

[0106] Furthermore, the information acquisition module 501 is also used to determine the service channels, associated users and keyword sets corresponding to all business scenarios through the operation configuration platform; and to construct a scenario information score table based on the service channels, the associated users and the keyword sets.

[0107] Furthermore, the information acquisition module 501 is also used to acquire the weight information of each score table object in the scene information score table in the corresponding business scene; and to classify and merge the group chat information, user information, keywords and the weight information from the scene information score table according to the business scene to obtain a scene-related information set.

[0108] Furthermore, the business scenario determination module 503 is also used to obtain the business scenario of the target element corresponding to the extreme value of the score in the set of valid background scores; and to determine the business scenario corresponding to the extreme value of the score as the target business scenario.

[0109] This embodiment of the business scenario identification device discloses that, upon receiving a question inquiry request from a target user, it acquires the target user's question and background information; determines a set of questions to be tested based on the user's background information and the channel through which the user inquires; and determines a set of scenario-related information based on a scenario information score table; obtains a set of valid background scores based on the set of questions to be tested and the set of scenario-related information; and determines the target business scenario corresponding to the user's question based on the set of valid background scores. Because this embodiment obtains a set of valid background scores based on the set of questions to be tested and the set of scenario-related information, and determines the target business scenario corresponding to the target user's question based on the set of valid background scores, it solves the technical problem in existing technologies where manual processing of user inquiries cannot quickly locate the scenario to which the question belongs, thus preventing the provision of personalized services to customers.

[0110] Based on the first embodiment of the business scenario recognition device of the present invention, a second embodiment of the business scenario recognition device of the present invention is proposed.

[0111] In this embodiment, the business scenario determination module 503 is further configured to determine the target knowledge base corresponding to the target business scenario; obtain the question answer corresponding to the user question through the target knowledge base; obtain the number of incorrect answers in the question answer, and determine the number of incorrect responses based on the number of incorrect answers, so as to perform an evaluation operation based on the number of incorrect responses.

[0112] Furthermore, the business scenario determination module 503 is also used to generate high-frequency questions by high-frequency word matching when the number of incorrect responses does not exceed a preset threshold for the number of incorrect responses, and output the high-frequency questions to the target user terminal.

[0113] Furthermore, the business scenario determination module 503 is also used to determine whether to transfer to a human agent based on the preset scenario urgency level when the number of incorrect responses exceeds a preset threshold number of incorrect responses; if so, a target group chat is created and associated with the user's question; when the association is completed, the human agent is replaced through the target group chat.

[0114] This embodiment obtains the answers to user questions from a target knowledge base corresponding to the target business scenario, and determines the number of incorrect responses based on the number of incorrect answers. This number is used for evaluation to optimize the equipment. Simultaneously, when the number of incorrect responses does not exceed a preset threshold, high-frequency questions are generated through high-frequency word matching and output to the target user. When the number of incorrect responses exceeds the preset threshold, a decision is made based on the urgency of the scenario to determine whether to transfer the user to a human agent. If so, a target group chat is created and associated with the user question. Once the association is complete, a human agent is selected through the target group chat, thereby reducing customer service pressure and enabling rapid response to user needs.

[0115] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0116] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0117] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of 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 the present invention, 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 / random access memory, 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 the present invention.

[0118] The above are merely preferred embodiments of the present invention and do not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A business scenario identification method, characterized in that, The business scenario identification method includes: Upon receiving a question inquiry request from a target user, obtain the user's question and user background information; Based on the user background information and the channels through which the user asked questions, a set of questions to be tested is determined, and based on the scene information score table, a set of scene-related information is determined; A set of valid background scores is obtained based on the set of questions to be tested and the set of scene-related information. The target business scenario corresponding to the user's question is determined based on the set of valid background scores. Before the step of obtaining the user's question and background information when receiving a question inquiry request from the target user, the method further includes: The operation configuration platform determines the service channels, associated users, and keyword sets corresponding to all business scenarios. A scenario information score table is constructed based on the service channels, the associated users, and the keyword set. Obtain the weight information of each score table object in the scene information score table in the corresponding business scene; The step of determining the set of scene-related information based on the scene information score table includes: According to the business scenario, the group chat information, user information, keywords and the weight information are classified and merged from the scenario information score table to obtain a scenario-related information set.

2. The business scenario identification method as described in claim 1, characterized in that, After the step of determining the target business scenario corresponding to the user question based on the set of valid background scores, the method further includes: Determine the target knowledge base corresponding to the target business scenario; The answer to the user's question is obtained from the target knowledge base. The number of incorrect answers in the question's answer set is obtained, and the number of incorrect responses is determined based on the number of incorrect answers, so that the evaluation operation can be performed based on the number of incorrect responses.

3. The business scenario identification method as described in claim 2, characterized in that, After the steps of obtaining the number of incorrect answers in the question's answers and determining the number of incorrect responses based on the number of incorrect answers, and then using the number of incorrect responses for evaluation, the method further includes: When the number of incorrect responses does not exceed a preset threshold for the number of incorrect responses, high-frequency questions are generated by high-frequency word matching and the high-frequency questions are output to the target user terminal.

4. The business scenario identification method as described in claim 2, characterized in that, After the steps of obtaining the number of incorrect answers in the question's answers and determining the number of incorrect responses based on the number of incorrect answers, and then using the number of incorrect responses for evaluation, the method further includes: When the number of incorrect responses exceeds a preset threshold, a decision is made based on the preset urgency level of the scenario to determine whether to transfer the call to a human agent. If so, create a target group chat and associate the target group chat with the user's question; Once the association is complete, the manual operation is switched through the target group chat.

5. The business scenario identification method as described in claim 1, characterized in that, The step of determining the target business scenario corresponding to the user question based on the set of valid background scores includes: The business scenario for obtaining the target element corresponding to the extreme value of the score in the set of valid background scores; The business scenario corresponding to the extreme value of the score is determined as the target business scenario.

6. A business scenario recognition device, characterized in that, The device includes: The information acquisition module is used to acquire the user's question and background information when it receives a question inquiry request sent by the target user. The set determination module is used to determine the set of questions to be tested based on the user background information and the user's questioning channels, and to determine the set of scene-related information based on the scene information score table; The set determination module is further configured to obtain a set of valid background scores based on the set of questions to be tested and the set of scene association information. The business scenario determination module is used to determine the target business scenario corresponding to the user question based on the set of valid background scores. The information acquisition module is further configured to determine the service channels, associated users, and keyword sets corresponding to all business scenarios through the operation configuration platform; construct a scenario information score table based on the service channels, associated users, and keyword sets; obtain the weight information of each score table object in the scenario information score table in the corresponding business scenario; and classify and merge the group chat information, user information, keywords, and weight information from the scenario information score table according to the business scenario to obtain a scenario-related information set.

7. A business scenario recognition device, characterized in that, The device includes: a memory, a processor, and a service scenario identification program stored in the memory and executable on the processor, the service scenario identification program being configured to implement the steps of the service scenario identification method as described in any one of claims 1 to 5.

8. A storage medium, characterized in that, The storage medium stores a business scenario identification program, which, when executed by a processor, implements the steps of the business scenario identification method as described in any one of claims 1 to 5.

Images