Method and device for generating target service, program product and storage medium

Abstract

The embodiment of the invention provides a method and device for generating a target service, a program product and a storage medium, and the method comprises the steps: obtaining first text data and target voice data generated by a target object through a target account in a session interaction process in the session interaction process with the target account; extracting N emotional features of the target object from the target text data and the target voice data; based on the N emotional characteristics, determining an operation level of the target object for executing the target operation for the session interaction; and generating a target service based on the operation level and the target parameters, the target parameters including a first parameter and a second parameter, the first parameter including an object feature of the target object, the second parameter including a dynamic operation feature executed by the target object through the target account in a historical time period, and the target service being a service corresponding to the target operation. The problem that the service cannot be generated in a targeted mode in the related technology is solved, and the effect of generating the service in a targeted mode is achieved.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of computer technology, and in particular, to a method and device for generating a target service, a program product, and a storage medium. Background Art

[0002] In the process of handling business, there are cases where problems arise due to unexpected problems, leading to customer complaints. In this case, customers usually show negative emotions when communicating with customer service. In the communication with customers, the customer service of the traditional consumer finance industry does not judge the customer's emotions and the level of susceptibility to litigation. In order to reduce subsequent customer complaints, they generally adopt a unified soothing speech, which is unable to generate targeted services to reduce subsequent customer complaints. Summary of the invention

[0003] The embodiments of the present application provide a method and apparatus, a program product, and a storage medium for generating a target service, so as to at least solve the problem that services cannot be generated in a targeted manner in the related art.

[0004] According to one embodiment of the present application, a method for generating a target service is provided, comprising: in a process of conducting a conversation interaction with a target account, obtaining first text data and target voice data generated by a target object through the target account during the conversation interaction; extracting N emotional features of the target object from the target text data and the target voice data, wherein the target text data is determined by the following steps: converting the target voice data into second text data, merging the first text data and the second text data to obtain the target text data, and N is a natural number greater than 1; determining an operation level of the target object performing a target operation for the conversation interaction based on the N emotional features; generating a target service based on the operation level and target parameters, wherein the target parameters include a first parameter and a second parameter, the first parameter includes the object features of the target object, and the second parameter includes the dynamic operation features performed by the target object through the target account within a historical time period, and the target service is a service corresponding to the target operation.

[0005] In an exemplary embodiment, during a conversation interaction with a target account, first text data and target voice data generated by the target object through the target account during the conversation interaction are obtained, including: obtaining text data and voice data output by the target object through the target account; based on text keywords, obtaining data within a preset text range from the text data to obtain the first text data, wherein the text keywords are keywords determined based on a conversation topic of the conversation interaction; based on voice keywords, obtaining data within a preset time range from the voice data to obtain the target voice data, wherein the voice keywords are keywords determined based on a conversation topic of the conversation interaction.

[0006] In an exemplary embodiment, N emotion features of the target object are extracted from the target text data and the target voice data, including: extracting target text features from the target text data based on text emotion features to obtain a first emotion feature, wherein the text emotion feature is used to represent the text emotion expression of the target object; extracting target voice features from the target voice data based on acoustic features to obtain a second emotion feature, wherein the acoustic feature is used to represent the voice emotion expression of the target object; and determining the first emotion feature and the second emotion feature as the emotion features.

[0007] In an exemplary embodiment, determining the operation level of the target operation performed by the target object for the conversation interaction based on the N emotion features includes: performing a classification operation on the N emotion features to obtain M emotion feature groups, wherein the M is a natural number greater than or equal to 1 and less than or equal to the N, wherein each emotion feature group includes at least one emotion feature, and each emotion feature group has a different emotion level; determining the emotion value of each emotion feature group according to preset weights of the emotion levels; and determining the operation level based on the M emotion values.

[0008] In an exemplary embodiment, a target service is generated based on the operation level and the target parameters, including: based on the operation level and the first parameter, obtaining a first service rule from a service rule library to generate a first target service, wherein the service rule library includes service rules for reducing the operation level, the service rules include the first service rule, and the first target service is used to provide the target object with session interaction data for reducing the operation level; based on the operation level and the second parameter, obtaining a second service rule from the service rule library to generate a second target service, wherein the service rules include the second service rule, and the second target service is used to provide the target object with policy data for reducing the operation level; and determining the first target service and the second target service as the target services.

[0009] In an exemplary embodiment, after generating the target service based on the above-mentioned operation level and target parameters, the above-mentioned method also includes: pushing the above-mentioned target service to the above-mentioned target account according to the object characteristics of the above-mentioned target object; collecting feedback information of the above-mentioned target object on the above-mentioned target service; and updating the service rule library based on the above-mentioned feedback information, wherein the above-mentioned service rule library includes service rules for lowering the above-mentioned operation level.

[0010] According to one embodiment of the present application, a device for generating a target service includes: a first acquisition module, which is used to acquire first text data and target voice data generated by a target object through the target account during the conversation interaction with a target account; a first extraction module, which is used to extract N emotional features of the target object from the target text data and the target voice data, wherein the target text data is determined by the following steps: converting the target voice data into second text data, merging the first text data and the second text data to obtain the target text data, and N is a natural number greater than 1; a first determination module, which is used to determine the operation level of the target object performing a target operation for the conversation interaction based on the N emotional features; a first generation module, which is used to generate a target service based on the operation level and target parameters, wherein the target parameters include a first parameter and a second parameter, the first parameter includes the object characteristics of the target object, and the second parameter includes the dynamic operation characteristics performed by the target object through the target account in a historical time period, and the target service is a service corresponding to the target operation.

[0011] In an exemplary embodiment, the above-mentioned first acquisition module includes: a first acquisition sub-module, used to acquire the text data and voice data output by the above-mentioned target object through the above-mentioned target account; a second acquisition sub-module, used to acquire data within a preset text range from the above-mentioned text data based on text keywords, and obtain the above-mentioned first text data, wherein the above-mentioned text keywords are keywords determined based on the conversation topic of the above-mentioned conversation interaction; a third acquisition sub-module, used to acquire data within a preset time range from the above-mentioned voice data based on voice keywords, and obtain the above-mentioned target voice data, wherein the above-mentioned voice keywords are keywords determined based on the conversation topic of the above-mentioned conversation interaction.

[0012] In an exemplary embodiment, the above-mentioned first extraction module includes: a first extraction submodule, used to extract target text features from the above-mentioned target text data based on text emotion features to obtain a first emotion feature, wherein the above-mentioned text emotion feature is used to represent the text emotion expression of the above-mentioned target object; a second extraction submodule, used to extract target speech features from the above-mentioned target speech data based on acoustic features to obtain a second emotion feature, wherein the above-mentioned acoustic features are used to represent the speech emotion expression of the above-mentioned target object; a first determination submodule, used to determine the above-mentioned first emotion feature and the above-mentioned second emotion feature as the above-mentioned emotion features.

[0013] In an exemplary embodiment, the first determination module includes: a first execution submodule, used to perform classification operations on the N emotion features to obtain M emotion feature groups, wherein the M is a natural number greater than or equal to 1 and less than or equal to the N, wherein each of the emotion feature groups includes at least one emotion feature, and each of the emotion feature groups has a different emotion level; a second determination submodule, used to determine the emotion value of each of the emotion feature groups according to the preset weight of the emotion level; and a third determination submodule, used to determine the operation level based on the M emotion values.

[0014] In an exemplary embodiment, the first generation module includes: a fourth acquisition submodule, used to acquire the first service rule from the service rule library based on the operation level and the first parameter, and generate a first target service, wherein the service rule library includes service rules for reducing the operation level, the service rules include the first service rule, and the first target service is used to provide the target object with session interaction data for reducing the operation level; a fifth acquisition submodule, used to acquire the second service rule from the service rule library based on the operation level and the second parameter, and generate a second target service, wherein the service rules include the second service rule, and the second target service is used to provide the target object with policy data for reducing the operation level; a fourth determination submodule, used to determine the first target service and the second target service as the target services.

[0015] In an exemplary embodiment, the above-mentioned device also includes: a first push module, which is used to generate the target service based on the above-mentioned operation level and target parameters, and then push the above-mentioned target service to the above-mentioned target account according to the object characteristics of the above-mentioned target object; a first collection module, which is used to collect feedback information of the above-mentioned target object on the above-mentioned target service; and a first update module, which is used to update the service rule library based on the above-mentioned feedback information, wherein the above-mentioned service rule library includes service rules for lowering the above-mentioned operation level.

[0016] According to another embodiment of the present application, a computer program product is provided, including a computer program, wherein the computer program is configured for a processor to execute the steps of any of the above method embodiments.

[0017] According to another embodiment of the present application, a computer-readable storage medium is provided, in which a computer program is stored, wherein the computer program is configured to enable a processor to execute the steps in any of the above method embodiments.

[0018] According to another embodiment of the present application, an electronic device is provided, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor is configured to execute the computer program to perform the steps in any one of the method embodiments.

[0019] Through this application, in the process of conducting a conversation interaction with a target account, the target text data and target voice data generated by the target object through the target account during the conversation interaction are obtained. And N emotional features of the target object are extracted from the target text data and the target voice data, and the operation level of the target object performing the target operation for the conversation interaction is determined based on the N emotional features; based on the operation level and the object characteristics of the target object, and the dynamic operation characteristics performed by the target object through the target account in the historical time period, the target service is generated. Since the present application can generate the target service according to the emotional state and historical behavior characteristics of the target object during the conversation interaction, it solves the problem of the inability to generate services in a targeted manner in the related technology, and achieves the effect of generating services in a targeted manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a hardware structure block diagram of a mobile terminal of a method for generating a target service according to an embodiment of the present application;

[0021] Figure 2 is a flow chart of a method for generating a target service according to an embodiment of the present application;

[0022] Figure 3 It is a flow chart of a method for generating a target service according to a specific embodiment of the present application;

[0023] Figure 4 It is a structural block diagram of a device for generating a target service according to an embodiment of the present application. DETAILED DESCRIPTION

[0024] The embodiments of the present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0025] It should be noted that the terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0026] The method embodiments provided in the embodiments of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Taking running on a mobile terminal as an example, Figure 1 1 is a hardware structure block diagram of a mobile terminal of a method for generating a target service according to an embodiment of the present application. Figure 1 As shown, the mobile terminal may include one or more ( Figure 1 Only one is shown in the figure) a processor 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data, wherein the mobile terminal may also include a transmission device 106 and an input / output device 108 for communication functions. It can be understood by those skilled in the art that Figure 1 The structure shown is for illustration only and does not limit the structure of the mobile terminal. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.

[0027] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as a computer program corresponding to a method for generating a target service in an embodiment of the present application. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, implementing the above method. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the mobile terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0028] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the mobile terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0029] In this embodiment, a method for generating a target service is provided. Figure 2 is a flow chart of a method for generating a target service according to an embodiment of the present application. Figure 2 As shown, the process includes the following steps:

[0030] Step S202, in the process of performing a conversation interaction with a target account, obtaining first text data and target voice data generated by the target object through the target account in the process of the conversation interaction;

[0031] Step S204, extracting N emotion features of the target object from the target text data and the target voice data, wherein the target text data is determined by the following steps: converting the target voice data into second text data, merging the first text data and the second text data to obtain the target text data, and N is a natural number greater than 1;

[0032] Step S206, determining the operation level of the target object performing the target operation for the conversation interaction based on the N emotion features;

[0033] Optionally, the target operation may be a complaint operation, and the operation level may be a complaint level.

[0034] Step S208: Generate a target service based on the above operation level and target parameters, wherein the above target parameters include a first parameter and a second parameter, the above first parameter includes the object characteristics of the above target object, the above second parameter includes the dynamic operation characteristics performed by the above target object through the above target account within a historical time period, and the above target service is a service corresponding to the above target operation.

[0035] Optionally, the object characteristics include, but are not limited to, the target object's gender, age, occupation, interests, and geographic location. Dynamic operational characteristics include, but are not limited to: the target object's recent purchase behavior, the target object's feedback and complaint history, and the target object's browsing history. For example, the target object's recently purchased products or services, which can be used to predict the subsequent services or products that the customer may need. The target object's past feedback and complaint records, which help identify the target object's concerns and potential areas for service improvement.

[0036] Through the above steps, in the process of conversation interaction with the target account, the target text data and target voice data generated by the target object through the target account during the conversation interaction are obtained. And N emotional features of the target object are extracted from the target text data and the target voice data, and the operation level of the target object performing the target operation for the conversation interaction is determined based on the N emotional features; based on the operation level and the object characteristics of the target object, and the dynamic operation characteristics performed by the target object through the target account in the historical time period, the target service is generated. Since the present application can generate the target service according to the emotional state and historical behavior characteristics of the target object during the conversation interaction, it solves the problem of the inability to generate services in a targeted manner in the related technology, and achieves the effect of generating services in a targeted manner.

[0037] In an exemplary embodiment, during a conversation interaction with a target account, first text data and target voice data generated by the target object through the target account during the conversation interaction are obtained, including: obtaining text data and voice data output by the target object through the target account; based on text keywords, obtaining data within a preset text range from the text data to obtain the first text data, wherein the text keywords are keywords determined based on a conversation topic of the conversation interaction; based on voice keywords, obtaining data within a preset time range from the voice data to obtain the target voice data, wherein the voice keywords are keywords determined based on a conversation topic of the conversation interaction.

[0038] Optionally, the conversation topic is bank account services, and keywords include but are not limited to account opening, transfer, account management, deposit, withdrawal, account balance, and bank fees. The conversation topic is insurance product consultation, and keywords include but are not limited to insurance, premium, insured amount, claim, and insurance terms. The conversation topic is investment consultation, and keywords include but are not limited to stocks, funds, bonds, investment returns, risk assessment, asset allocation, and market analysis. The conversation topic is loan services, and keywords include but are not limited to housing loans, personal loans, commercial loans, interest rates, loan terms, repayment plans, and loan conditions. The conversation topic is credit card services, and keywords include but are not limited to credit card applications, credit limit increases, bill inquiries, points redemption, annual fees, and overdue fees.

[0039] Optionally, for example, when the target object consults with the manual customer service about the claim status of a certain insurance product through the target account, the keywords include but are not limited to insurance, premium, insured amount, and claim amount. Using text analysis tools and speech recognition technology, target text data and target speech data containing premium, insured amount, and claim amount are extracted from the text data and speech data in the consultation process.

[0040] This embodiment can extract valuable data from a large amount of conversation data by acquiring data within a preset range based on keywords, thereby improving data utilization and effectively locating specific needs and problems of the target object.

[0041] In an exemplary embodiment, N emotion features of the target object are extracted from the target text data and the target voice data, including: extracting target text features from the target text data based on text emotion features to obtain a first emotion feature, wherein the text emotion feature is used to represent the text emotion expression of the target object; extracting target voice features from the target voice data based on acoustic features to obtain a second emotion feature, wherein the acoustic feature is used to represent the voice emotion expression of the target object; and determining the first emotion feature and the second emotion feature as the emotion features.

[0042] Optionally, text sentiment features include, but are not limited to, word choice, sentiment orientation, sentiment complexity, sentiment context, word choice, sentence structure, and rhetoric. Acoustic features include, but are not limited to, pitch variation, speech rate and rhythm, volume and intensity, intonation patterns, physical properties of sound, energy, and resonance.

[0043] Optionally, before extracting emotional features, the target text data and target voice data may be preprocessed to improve extraction efficiency and accuracy, including but not limited to cleaning the target text data, including removing stop words and punctuation marks, performing stem extraction or word form restoration, etc., to prepare the data for sentiment analysis. The target voice data may be preprocessed by noise reduction, segmentation, etc., to improve the accuracy of voice recognition.

[0044] Optionally, for example, the target object is having a conversation with a customer service representative, expressing dissatisfaction with the financial product recently purchased. Target text data: I am very disappointed. The description on your website is different from the actual financial product! I request an immediate refund. Based on the recognition of the above text sentiment features, the following first sentiment features can be extracted: Negative sentiment intensity: high, sentiment type: disappointment and anger, urgency: high. After that, the target object had a voice communication with the customer service representative. The target voice data has the characteristics of high fundamental frequency, strong sound, and high zero crossing rate. Based on the recognition of the above acoustic features, the following second sentiment feature can be extracted: anger.

[0045] This embodiment achieves the purpose of improving the accuracy of emotion recognition by combining multimodal emotion recognition with text and voice.

[0046] In an exemplary embodiment, determining the operation level of the target operation performed by the target object for the conversation interaction based on the N emotion features includes: performing a classification operation on the N emotion features to obtain M emotion feature groups, wherein the M is a natural number greater than or equal to 1 and less than or equal to the N, wherein each emotion feature group includes at least one emotion feature, and each emotion feature group has a different emotion level; determining the emotion value of each emotion feature group according to preset weights of the emotion levels; and determining the operation level based on the M emotion values.

[0047] Optionally, the extracted N emotional features are classified to obtain M emotional feature groups. For example, N=5, M=3, and possible emotional feature groups include an anger group, an anxiety group, and a satisfaction group. An emotional level is assigned to each emotional feature group, such as an anger level of 3, an anxiety level of 2, and a satisfaction level of 1. Weights are assigned to each emotional level according to business requirements. For example, the highest weight of anger is 0.5, the second is 0.3, and the lowest is 0.2. According to the emotional level and preset weight of each emotional feature group, the emotional value of each emotional feature group is calculated. For example, if the emotional level of the anger group is 3 and the weight is 0.5, the emotional value is 1.5. Determine the operation level based on the emotional value: summarize or weighted average the M emotional values ​​or determine the final operation level based on the threshold judgment. For example, if the emotional value of the anxiety group is 0.6 and the emotional value of the satisfaction group is 0.2, the final operation level may be determined based on the emotional value of the anger group, 1.5, because it has the highest weight, and the final operation level is anger-high level.

[0048] This embodiment achieves the purpose of effectively optimizing customer service processes, improving customer experience, and reducing the probability of customer complaints by using a method for determining operation levels based on emotional characteristics.

[0049] In an exemplary embodiment, a target service is generated based on the operation level and the target parameters, including: based on the operation level and the first parameter, obtaining a first service rule from a service rule library to generate a first target service, wherein the service rule library includes service rules for reducing the operation level, the service rules include the first service rule, and the first target service is used to provide the target object with session interaction data for reducing the operation level; based on the operation level and the second parameter, obtaining a second service rule from the service rule library to generate a second target service, wherein the service rules include the second service rule, and the second target service is used to provide the target object with policy data for reducing the operation level; and determining the first target service and the second target service as the target services.

[0050] Optionally, a first target service can be generated based on the operation level, target text data, target voice data, first parameter and second parameter. The specific reasons why the target object produces the above-mentioned emotions can be accurately located through the target text data and target voice data. Combined with the first parameter and the second parameter, the probability of the target object performing the target operation can be effectively reduced.

[0051] Optionally, after determining that the target object's operation level for performing the target operation is anger-high level by analyzing the target text data and the target voice data, the first service rule is obtained from the service rule library based on the first parameter: the target object's gender, age, occupation, and interest. The first service rule is obtained from the service rule library based on the second parameter: the target object's recent purchase behavior, the target object's feedback and complaint history, and the target object's browsing history. For example, a bank hopes to reduce customer complaints, especially complaints against its investment products, by providing personalized services. The bank has an intelligent customer service platform that can provide customized services based on the customer's characteristics (gender, age, occupation, and interest). For example, the first parameter of the target object: gender-female, age-40 years old, occupation-financial analyst, interest-investment, reading financial news, golf, the second parameter of the target object: recently frequently browse a certain type of financial product, and analyze the first parameter: as a female customer of a financial analyst, she may have an in-depth understanding of complex financial products and expect to receive professional investment advice and high-quality customer service. A 40-year-old customer may be in the wealth accumulation stage and have high expectations for asset appreciation and investment returns. The client's interest in investment and financial news indicates that she may be very concerned about market trends and investment opportunities. Service rules for occupation: Provide her with instant market analysis and investment advice. Given the client's interest in investment, the bank can provide customized financial news summaries and market updates. Service rules for age: Provide asset allocation advice and wealth management services for clients in the wealth accumulation stage. Combining the above service rules, the bank generates the following first-target service: Provide a dedicated investment advisor to provide professional market analysis and investment advice to financial analysts. Regularly send customized financial news summaries and market updates to help customers keep abreast of market changes. Second-target service: Provide personalized asset allocation advice and wealth management services to meet the needs of customers in the wealth accumulation stage. Push preferential services on financial products.

[0052] This embodiment can provide services closely related to customers based on the first parameter and the second parameter, meet customers' professional needs and personalized services, and achieve the purpose of effectively reducing customer complaints and improving service effectiveness and customer stickiness.

[0053] In an exemplary embodiment, after generating the target service based on the above-mentioned operation level and target parameters, the above-mentioned method also includes: pushing the above-mentioned target service to the above-mentioned target account according to the object characteristics of the above-mentioned target object; collecting feedback information of the above-mentioned target object on the above-mentioned target service; and updating the service rule library based on the above-mentioned feedback information, wherein the above-mentioned service rule library includes service rules for lowering the above-mentioned operation level.

[0054] Optionally, the feedback information includes, but is not limited to, proactive feedback information, emotion recognition feedback information, and behavior tracking feedback information. Proactive feedback: After the service is pushed, the target object is proactively asked about the service satisfaction and improvement suggestions through intelligent customer service or questionnaires, rather than relying solely on passive complaints or opinion collection. Emotion recognition feedback: Using sentiment analysis technology, by analyzing the target object's language and behavior during the use of the service (such as through voice recognition and chat record analysis), its emotional feedback on the service is automatically identified, even if the user does not express it directly. Behavior tracking feedback: Analyze the target object's behavioral changes through the target account's usage data, such as whether there is an increase in account activity, whether a second purchase has been made, etc., to indirectly evaluate the effectiveness and satisfaction of the service.

[0055] Optionally, the bank pushes these services to users in the form of personalized notifications through the application interface, such as through mobile banking application push, email or SMS. At the same time, the most suitable push method and time point can be selected according to the behavior habits and preferences of the target object. For example, if the target object usually uses the service at night, it may be more effective to push service information at night.

[0056] Optionally, the service rule base contains service rules for reducing the operation level, which are designed based on customer information, service history and behavior patterns. Based on customer feedback, the bank updates the service rule base. For example, if customers give poor feedback on certain services, the bank will analyze the reasons and adjust the service rules to improve service quality. It can provide personalized services based on customer characteristics, reduce customer complaint rates, and improve service quality and customer satisfaction.

[0057] The present invention will be described below in conjunction with specific embodiments:

[0058] Figure 3 This is a flowchart of a method for generating a target service in a specific embodiment of the present application, comprising the following steps:

[0059] S302, the customer consults the intelligent customer service about a financial product in the form of text and voice;

[0060] S304, using text analysis tools and speech recognition technology to extract target text data and target speech data containing keywords such as premium, insured amount, and claim amount from the text data and speech data during the consultation process;

[0061] S306, based on the recognition of text-based emotional features, the following first emotional features are extracted: negative emotional intensity: low, emotional type: anxious, and urgency: high. The target speech data has the features of high fundamental frequency, strong sound, and high zero-crossing rate. Based on the recognition of the above acoustic features, the following second emotional features are extracted: urgency;

[0062] S308, classify the extracted 4 emotional features to obtain 2 emotional feature groups, including an anger group and an anxiety group. Assign an emotional level to each emotional feature group, the anger level is 1, and the anxiety level is 3. According to the emotional level and preset weight of each emotional feature group, calculate the emotional value of each emotional feature group, the emotional value of the anxiety group is 0.6, and the emotional value of the anger group is 0.2. Determine the operation level based on the emotional value: summarize or weighted average the two emotional values ​​or determine the final operation level based on the threshold judgment. For example, the emotional value of the anxiety group is 0.6, which is greater than the preset threshold and the emotional value of the anger group is less than the preset threshold. The final operation level is anxiety-high level;

[0063] S310, based on the operational level anxiety-high level prediction that the customer is likely to complain, go to S312, otherwise go to S316;

[0064] S312, add customer complaint label;

[0065] S314, intelligent customer service transfers to manual customer service, displays the Yisu label and provides target services, where the target services are determined based on the customer's object characteristics and dynamic operation characteristics. Object characteristics: gender - female, age - 40 years old, occupation - financial analyst, interests - investment, reading financial news, golf, dynamic operation parameters: frequently browse the rise and fall of a certain type of financial product recently, analyze the above characteristics to generate the following first target service: provide professional market analysis and investment advice. Regularly send customized financial news summaries and market dynamics updates to help customers keep abreast of market changes. Second target service: push preferential services for a certain financial product;

[0066] S316, the intelligent customer service continues to provide services and provides the following target services: regularly sending customized financial news summaries and market dynamic updates to help customers keep abreast of market changes. Second target service: push preferential services for financial products;

[0067] S318, end.

[0068] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods of each embodiment of the present application.

[0069] In the present embodiment, a device for generating a target service is also provided, and the device is used to implement the above-mentioned embodiment and preferred implementation mode, and the descriptions that have been made are not repeated. As used below, the term "module" can implement a combination of software and / or hardware of a predetermined function. Although the device described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware is also possible and conceived.

[0070] Figure 4 is a structural block diagram of a device for generating a target service according to an embodiment of the present application, such as Figure 4 As shown, the device comprises:

[0071] A first acquisition module 402 is used to acquire, during a conversation interaction with a target account, first text data and target voice data generated by a target object through the target account during the conversation interaction;

[0072] A first extraction module 404 is used to extract N emotion features of the target object from the target text data and the target voice data, wherein the target text data is determined by the following steps: converting the target voice data into second text data, merging the first text data and the second text data to obtain the target text data, and N is a natural number greater than 1;

[0073] A first determination module 406 is used to determine the operation level of the target object performing the target operation for the conversation interaction based on the N emotion features;

[0074] The first generation module 408 is used to generate a target service based on the above-mentioned operation level and target parameters, wherein the above-mentioned target parameters include a first parameter and a second parameter, the above-mentioned first parameter includes the object characteristics of the above-mentioned target object, and the above-mentioned second parameter includes the dynamic operation characteristics performed by the above-mentioned target object through the above-mentioned target account within a historical time period, and the above-mentioned target service is a service corresponding to the above-mentioned target operation.

[0075] In an exemplary embodiment, the above-mentioned first acquisition module includes: a first acquisition sub-module, used to acquire the text data and voice data output by the above-mentioned target object through the above-mentioned target account; a second acquisition sub-module, used to acquire data within a preset text range from the above-mentioned text data based on text keywords, and obtain the above-mentioned first text data, wherein the above-mentioned text keywords are keywords determined based on the conversation topic of the above-mentioned conversation interaction; a third acquisition sub-module, used to acquire data within a preset time range from the above-mentioned voice data based on voice keywords, and obtain the above-mentioned target voice data, wherein the above-mentioned voice keywords are keywords determined based on the conversation topic of the above-mentioned conversation interaction.

[0076] In an exemplary embodiment, the above-mentioned first extraction module includes: a first extraction submodule, used to extract target text features from the above-mentioned target text data based on text emotion features to obtain a first emotion feature, wherein the above-mentioned text emotion feature is used to represent the text emotion expression of the above-mentioned target object; a second extraction submodule, used to extract target speech features from the above-mentioned target speech data based on acoustic features to obtain a second emotion feature, wherein the above-mentioned acoustic features are used to represent the speech emotion expression of the above-mentioned target object; a first determination submodule, used to determine the above-mentioned first emotion feature and the above-mentioned second emotion feature as the above-mentioned emotion features.

[0077] In an exemplary embodiment, the first determination module includes: a first execution submodule, used to perform classification operations on the N emotion features to obtain M emotion feature groups, wherein the M is a natural number greater than or equal to 1 and less than or equal to the N, wherein each of the emotion feature groups includes at least one emotion feature, and each of the emotion feature groups has a different emotion level; a second determination submodule, used to determine the emotion value of each of the emotion feature groups according to the preset weight of the emotion level; and a third determination submodule, used to determine the operation level based on the M emotion values.

[0078] In an exemplary embodiment, the first generation module includes: a fourth acquisition submodule, used to acquire the first service rule from the service rule library based on the operation level and the first parameter, and generate a first target service, wherein the service rule library includes service rules for reducing the operation level, the service rules include the first service rule, and the first target service is used to provide the target object with session interaction data for reducing the operation level; a fifth acquisition submodule, used to acquire the second service rule from the service rule library based on the operation level and the second parameter, and generate a second target service, wherein the service rules include the second service rule, and the second target service is used to provide the target object with policy data for reducing the operation level; a fourth determination submodule, used to determine the first target service and the second target service as the target services.

[0079] In an exemplary embodiment, the above-mentioned device also includes: a first push module, which is used to generate the target service based on the above-mentioned operation level and target parameters, and then push the above-mentioned target service to the above-mentioned target account according to the object characteristics of the above-mentioned target object; a first collection module, which is used to collect feedback information of the above-mentioned target object on the above-mentioned target service; and a first update module, which is used to update the service rule library based on the above-mentioned feedback information, wherein the above-mentioned service rule library includes service rules for lowering the above-mentioned operation level.

[0080] An embodiment of the present application provides a computer program product, including a computer program, and when the computer program is executed by a processor, the steps in any one of the above method embodiments are implemented.

[0081] An embodiment of the present application further provides a computer-readable storage medium, in which a computer program is stored, wherein the computer program is configured to execute the steps of any of the above method embodiments when running.

[0082] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0083] An embodiment of the present application further provides an electronic device, including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0084] In an exemplary embodiment, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0085] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail herein.

[0086] Obviously, those skilled in the art should understand that the above modules or steps of the present application can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order from that herein, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present application is not limited to any specific combination of hardware and software.

[0087] The above are only preferred embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method for generating a target service, characterized in that: include: During the conversation interaction with the target account, obtaining first text data and target voice data generated by the target object through the target account during the conversation interaction; Extracting N emotion features of the target object from the target text data and the target voice data, wherein the target text data is determined by the following steps: converting the target voice data into second text data, merging the first text data and the second text data to obtain the target text data, and N is a natural number greater than 1; Determining, based on the N emotion features, an operation level at which the target object performs a target operation on the conversation interaction; A target service is generated based on the operation level and target parameters, wherein the target parameters include a first parameter and a second parameter, the first parameter includes object characteristics of the target object, the second parameter includes dynamic operation characteristics performed by the target object through the target account within a historical time period, and the target service is a service corresponding to the target operation.

2. The method according to claim 1, characterized in that In a process of performing a conversation interaction with a target account, obtaining first text data and target voice data generated by a target object through the target account during the conversation interaction includes: Acquiring text data and voice data output by the target object through the target account; Based on text keywords, data within a preset text range is acquired from the text data to obtain the first text data, wherein the text keywords are keywords determined based on the conversation topic of the conversation interaction; Based on voice keywords, data within a preset time range is acquired from the voice data to obtain the target voice data, wherein the voice keywords are keywords determined based on the conversation topic of the conversation interaction.

3. The method according to claim 1, characterized in that Extracting N emotion features of the target object from the target text data and the target voice data includes: Extracting a target text feature from the target text data based on the text emotion feature to obtain a first emotion feature, wherein the text emotion feature is used to represent the text emotion expression of the target object; Extracting a target speech feature from the target speech data based on the acoustic feature to obtain a second emotion feature, wherein the acoustic feature is used to represent the speech emotion expression of the target object; The first emotion feature and the second emotion feature are determined as the emotion features.

4. The method according to claim 1, characterized in that Determining the operation level of the target object performing the target operation on the conversation interaction based on the N emotion features includes: Performing a classification operation on the N emotion features to obtain M emotion feature groups, wherein M is a natural number greater than or equal to 1 and less than or equal to N, wherein each emotion feature group includes at least one emotion feature, and each emotion feature group has a different emotion level; Determining the emotion value of each of the emotion feature groups according to the preset weight of the emotion level; The operation level is determined based on the M emotion values.

5. The method according to claim 1, characterized in that Based on the operation level and the target parameters, generating a target service includes: Based on the operation level and the first parameter, a first service rule is obtained from a service rule library to generate a first target service, wherein the service rule library includes a service rule for reducing the operation level, the service rules include the first service rule, and the first target service is used to provide the target object with session interaction data for reducing the operation level; Based on the operation level and the second parameter, obtaining a second service rule from the service rule library to generate a second target service, wherein the service rule includes the second service rule, and the second target service is used to provide policy data for reducing the operation level to the target object; The first target service and the second target service are determined as the target services.

6. The method according to claim 1, characterized in that After generating the target service based on the operation level and the target parameter, the method further includes: Pushing the target service to the target account according to the object characteristics of the target object; Collecting feedback information of the target object on the target service; Based on the feedback information, a service rule base is updated, wherein the service rule base includes a service rule for reducing the operation level.

7. A device for generating a target service, characterized in that: include: A first acquisition module is used to acquire, during a conversation interaction with a target account, first text data and target voice data generated by a target object through the target account during the conversation interaction; A first extraction module is used to extract N emotion features of the target object from the target text data and the target voice data, wherein the target text data is determined by the following steps: converting the target voice data into second text data, merging the first text data and the second text data to obtain the target text data, and N is a natural number greater than 1; A first determination module is used to determine the operation level of the target object performing the target operation on the conversation interaction based on the N emotion features; The first generation module is used to generate a target service based on the operation level and the target parameters, wherein the target parameters include a first parameter and a second parameter, the first parameter includes the object characteristics of the target object, the second parameter includes the dynamic operation characteristics performed by the target object through the target account within a historical time period, and the target service is a service corresponding to the target operation.

8. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method described in any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein the computer program implements the steps of the method described in any one of claims 1 to 6 when executed by a processor.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method described in any one of claims 1 to 6 are implemented.

Images