Outbound processing method and device, computer equipment and storage medium

Abstract

The invention provides an outbound processing method and device, computer equipment and a storage medium. The method comprises the following steps: acquiring target driver data of a target driver; wherein the target driver data comprises personal information, behavior data and customer service incoming record information; performing prediction processing on the target driver data based on a trained target prediction model to obtain a corresponding purchase prediction probability and a target importance feature; judging whether the purchase prediction probability is greater than a preset threshold value or not; if yes, constructing a corresponding target verbal skill based on the target importance feature and the target driver data; building a corresponding intelligent customer service based on the target verbal skill; and based on the intelligent customer service, using the target verbal skill to execute outbound processing corresponding to the target driver. According to the invention, based on the use of the target prediction model, the target verbal skill and the intelligent customer service, the intelligence of call-out processing is improved, so that the conversion efficiency is improved, and the user experience is further improved.

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and in particular to an outbound call processing method, device, computer equipment and storage medium. Background Art

[0002] In the current customer management and marketing field, in order to expand the membership base and increase service revenue, many companies have adopted the method of telephone invitation to promote and sell membership services to non-member drivers. In this process, traditional practices often rely on outbound call robots to perform outbound call tasks. Although this method has achieved large-scale and automated customer contact to a certain extent, it faces a series of challenges and shortcomings in actual operation.

[0003] Although outbound call robots can significantly improve outbound call efficiency and reduce costs, they generally use preset scripts for conversations and lack sufficient intelligence and flexibility. This "one-size-fits-all" communication method ignores the individual differences between drivers and cannot be dynamically adjusted based on the drivers' immediate feedback or historical behavior data, thereby reducing the relevance and attractiveness of the conversation, making drivers feel that they are being mechanically processed, and the user experience is greatly reduced. In addition, this lack of personalized outbound call strategy is difficult to effectively stimulate the driver's interest and willingness to buy. Therefore, the existing outbound call processing methods have the problems of low intelligence and low conversion efficiency, and poor user experience. Summary of the invention

[0004] The main purpose of the present invention is to provide an outbound call processing method, device, computer equipment and storage medium, aiming to solve the technical problems that the outbound call processing method adopted in the prior art has low intelligence and low conversion efficiency, as well as poor user experience.

[0005] To achieve the above object, the present invention provides an outbound call processing method, the method comprising:

[0006] Obtain target driver data of the target driver; wherein the target driver data includes personal information, behavior data, and customer service call record information;

[0007] Based on the trained target prediction model, the target driver data is predicted and processed to obtain corresponding purchase prediction probability and target importance characteristics;

[0008] Determining whether the purchase prediction probability is greater than a preset threshold;

[0009] If yes, construct corresponding target speech based on the target importance feature and the target driver data;

[0010] Build corresponding intelligent customer service based on the target words;

[0011] Based on the intelligent customer service, the target script is used to execute outbound call processing corresponding to the target driver.

[0012] Optionally, constructing a corresponding target speech based on the target importance feature and the target driver data includes:

[0013] Call the preset script library;

[0014] Querying the designated speech that matches the target importance feature from the speech library; wherein the designated speech at least includes opening speech, redemption speech, speech asking for reasons for rejection, rights and interests guidance speech, and operation guidance speech;

[0015] Performing speech optimization processing on the specified speech based on the target driver data to obtain corresponding optimized speech;

[0016] The optimized speech is used as the target speech.

[0017] Optionally, the optimizing the specified speech based on the target driver data to obtain the corresponding optimized speech includes:

[0018] Based on the target driver data, the designated speech is processed for address adjustment to obtain a corresponding first speech;

[0019] Performing tone adjustment processing on the first speech to obtain a corresponding second speech;

[0020] Optimizing the wording of the second speech to obtain a corresponding third speech;

[0021] The third speech is used as the target speech.

[0022] Optionally, the building of a corresponding intelligent customer service based on the target speech includes:

[0023] Call the preset initial intelligent customer service;

[0024] Determine the key nodes in the target speech;

[0025] Determining the flow conditions between the key nodes;

[0026] Based on the target speech, the initial intelligent customer service is mapped to a speech, and based on the flow condition, the logic of the initial intelligent customer service is configured to obtain a corresponding designated intelligent customer service;

[0027] Performing performance testing on the designated intelligent customer service;

[0028] If the designated intelligent customer service passes the performance test, the designated intelligent customer service will be used as the intelligent customer service.

[0029] Optionally, the performing a performance test on the designated intelligent customer service includes:

[0030] Conducting a recognition accuracy test on the designated intelligent customer service;

[0031] If the designated intelligent customer service passes the recognition accuracy test, a conversation fluency test is performed on the designated intelligent customer service;

[0032] If the designated intelligent customer service passes the conversation fluency test, a response time test is performed on the designated intelligent customer service;

[0033] If the designated intelligent customer service passes the response time test, it is determined that the designated intelligent customer service has passed the test; otherwise, it is determined that the designated intelligent customer service has failed the test.

[0034] Optionally, before performing prediction processing on the target driver data based on the trained target prediction model to obtain corresponding purchase prediction probability and target importance features, the method further includes:

[0035] Obtaining preset business goals, and collecting corresponding driver data based on the business goals;

[0036] Preprocessing and feature engineering the driver data to obtain corresponding sample data;

[0037] Call the preset initial model;

[0038] Based on the business goal, the initial model is trained using the sample data to obtain a specified model that meets expectations;

[0039] The designated model is used as the target prediction model.

[0040] Optionally, after the outbound call processing corresponding to the target driver is performed based on the intelligent customer service using the target speech, the method further includes:

[0041] After completing the outbound call processing, obtaining the call data between the intelligent customer service and the target driver;

[0042] Analyze and process the call data to obtain corresponding analysis results;

[0043] Based on the analysis result, follow-up processing corresponding to the target driver is performed.

[0044] In addition, to achieve the above-mentioned purpose, the present invention further provides an outbound call processing device, the outbound call processing device comprising:

[0045] A first acquisition module is used to acquire target driver data of a target driver; wherein the target driver data includes personal information, behavior data, and customer service call record information;

[0046] A prediction module, used to perform prediction processing on the target driver data based on the trained target prediction model to obtain corresponding purchase prediction probability and target importance characteristics;

[0047] A judgment module, used to judge whether the purchase prediction probability is greater than a preset threshold;

[0048] A construction module, for constructing corresponding target speech based on the target importance feature and the target driver data;

[0049] A building module is used to build corresponding intelligent customer service based on the target speech;

[0050] The first processing module is used to execute the outbound call processing corresponding to the target driver using the target speech based on the intelligent customer service.

[0051] In order to solve the above technical problems, the embodiment of the present application further provides a computer device, which adopts the following technical solution:

[0052] The computer device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of any one of the outbound call processing methods proposed in the embodiments of the present application are implemented.

[0053] In order to solve the above technical problems, the embodiment of the present application further provides a computer-readable storage medium, which adopts the following technical solution:

[0054] The computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, the steps of any one of the outbound call processing methods proposed in the embodiments of the present application are implemented.

[0055] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0056] The present invention provides an outbound call processing method, device, computer equipment and storage medium, the method comprising: firstly obtaining target driver data of a target driver; wherein the target driver data comprises personal information, behavior data and customer service line record information; then predicting and processing the target driver data based on a trained target prediction model to obtain a corresponding purchase prediction probability and target importance characteristics; then judging whether the purchase prediction probability is greater than a preset threshold; if so, constructing a corresponding target speech based on the target importance characteristics and the target driver data; subsequently constructing a corresponding intelligent customer service based on the target speech; finally, based on the intelligent customer service, using the target speech to execute an outbound call processing corresponding to the target driver. The present application predicts and processes the target driver data based on the use of a trained target prediction model to obtain the corresponding purchase prediction probability and target importance characteristics, and when it is detected that the purchase prediction probability is greater than a preset threshold, it automatically and intelligently constructs the corresponding target speech based on the target importance characteristics and the target driver data, and further builds the corresponding intelligent customer service based on the target speech, so that subsequently based on the intelligent customer service, the target speech is used to execute the outbound call processing corresponding to the target driver, thereby realizing the customization of targeted and matching target speech for drivers with different characteristics, and by controlling the intelligent customer service to use the target speech to execute the outbound call processing corresponding to the target driver, it is possible to communicate and interact with the target driver more effectively, effectively improving the intelligence of the outbound call processing, which is conducive to improving conversion efficiency and improving user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the scheme in the present application, a brief introduction is given below to the drawings required for use in the description of the embodiments of the present application. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0058] Figure 1 is an exemplary system architecture diagram to which the present application may be applied;

[0059] Figure 2 is a flow chart of an outbound call processing method provided by an embodiment of the present invention;

[0060] Figure 3 is a structural schematic diagram of an embodiment of an outbound call processing device according to the present application;

[0061] Figure 4 This is a basic structural block diagram of the computer device in this embodiment. DETAILED DESCRIPTION

[0062] The outbound call processing method provided by the embodiment of the present invention is applied to the outbound call processing device. Unless otherwise defined, all technical and scientific terms used in this article have the same meaning as those generally understood by technicians in the technical field of this application; the terms used in the specification of the application herein are only for the purpose of describing specific embodiments and are not intended to limit this application; the terms "including" and "having" in the specification and claims of this application and the above-mentioned drawings and any variations thereof are intended to cover non-exclusive inclusions. The terms "first", "second", etc. in the specification and claims of this application or the above-mentioned drawings are used to distinguish different objects, not to describe a specific order.

[0063] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0064] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings.

[0065] like Figure 1 As shown, the system architecture 100 may include terminal devices 101, 102, 103, a network 104 and a server 105. The network 104 is used to provide a medium for communication links between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0066] Users can use terminal devices 101, 102, 103 to interact with server 105 through network 104 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 101, 102, 103, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social online platform software, etc.

[0067] Terminal devices 101, 102, 103 can be various electronic devices with display screens and supporting web browsing, including but not limited to smart phones, tablet computers, e-book readers, MP3 players (Moving Picture Experts Group Audio Layer III), MP4 (Moving Picture Experts Group Audio Layer IV) players, laptop computers, desktop computers, etc.

[0068] The server 105 may be a server that provides various services, such as a background server that provides support for web pages displayed on the terminal devices 101 , 102 , and 103 .

[0069] It should be noted that the outbound call processing method provided in the embodiment of the present application is generally executed by a server / terminal device, and accordingly, the outbound call processing device is generally arranged in the server / terminal device.

[0070] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0071] In the current customer management and marketing field, in order to expand the membership base and increase service revenue, many companies have adopted the method of telephone invitation to promote and sell membership services to non-member drivers. In this process, traditional practices often rely on outbound call robots to perform outbound call tasks. Although this method has achieved large-scale and automated customer contact to a certain extent, it faces a series of challenges and shortcomings in actual operation.

[0072] Although outbound call robots can significantly improve outbound call efficiency and reduce costs, they generally use preset scripts for conversations and lack sufficient intelligence and flexibility. This "one-size-fits-all" communication method ignores the individual differences between drivers and cannot be dynamically adjusted based on the drivers' immediate feedback or historical behavior data, thereby reducing the relevance and attractiveness of the conversation, making drivers feel that they are being mechanically processed, and the user experience is greatly reduced. In addition, this lack of personalized outbound call strategy is difficult to effectively stimulate the driver's interest and willingness to buy. Therefore, the existing outbound call processing methods have the problems of low intelligence and low conversion efficiency, and poor user experience.

[0073] Continue to refer Figure 2 , shows a flow chart of an embodiment of the outbound call processing method proposed in the present application. The embodiment of the present application can acquire and process relevant data based on artificial intelligence technology.

[0074] The outbound call processing method provided by the embodiment of the present invention comprises the following steps:

[0075] S210, obtaining target driver data of a target driver; wherein the target driver data includes personal information, behavior data, and customer service call record information.

[0076] In this step, the present invention can be applied to the business scenario of making outbound calls to drivers to invite them to purchase memberships, and the executor of the present invention can be specifically an outbound call processing system, or system. Among them, the target driver data includes personal information, behavioral data, and customer service call record information. Personal information may include: age, gender, deposit, behavioral points, wallet balance, vehicle type, etc. Behavioral data may include: inviting new drivers, inviting new passengers, completing class appointments, etc. Customer service call record information may include: unable to receive orders, using APP.

[0077] S220, performing prediction processing on the target driver data based on the trained target prediction model to obtain corresponding purchase prediction probability and target importance characteristics.

[0078] In this step, the target driver data can be input into the target prediction model, so that the input target driver data can be predicted and processed by the target prediction model and the corresponding purchase prediction probability and target importance features can be output. Among them, the purchase prediction probability can be the probability of the target driver purchasing a membership. The above-mentioned target importance features refer to the features that are ranked at the top after sorting the features according to the importance scores of the features. The obtained target importance features help to identify which features have the greatest impact on the model prediction, so that the corresponding target words can be designed for these features in the future. In addition, the importance scores of the features can be extracted from the trained target prediction model, which can be obtained through the interface or post-processing steps provided by the model. And there is no limitation on the determination of the features ranked at the top, for example, the features ranked in the top three in importance can be used. In addition, the specific construction process of the above-mentioned target prediction model will be further described in detail in the subsequent specific embodiments of this application, and will not be elaborated on here.

[0079] S230, determining whether the purchase prediction probability is greater than a preset threshold.

[0080] In this step, there is no specific limitation on the value of the preset threshold, which can be set according to actual business needs, for example, 0.6. If the purchase prediction probability is greater than the preset threshold, it is determined that the user is more likely to purchase a membership, and the subsequent outbound call processing will be performed. If the purchase prediction probability is less than the preset threshold, it is determined that the user is less likely to purchase a membership, and the subsequent outbound call processing will not be performed, so as to reduce the ineffective workload of the outbound call processing and improve the processing intelligence of the outbound call processing.

[0081] S240: If yes, construct corresponding target speech based on the target importance feature and the target driver data.

[0082] In this step, the specific implementation process of constructing the corresponding target speech based on the target importance characteristics and the target driver data will be further described in detail in the subsequent specific embodiments of this application and will not be elaborated here.

[0083] S250, building corresponding intelligent customer service based on the target speech.

[0084] In this step, the specific implementation process of building the corresponding intelligent customer service based on the target speech will be further described in detail in the subsequent specific embodiments of this application, and will not be elaborated here.

[0085] S260, based on the intelligent customer service, use the target speech to execute the outbound call processing corresponding to the target driver.

[0086] In this step, after the intelligent customer service is built, the intelligent customer service is launched online and used to make outbound calls to the target driver. The intelligent customer service uses the above-mentioned target words to communicate with the user during the outbound call process. This application conducts more refined hierarchical management of drivers, follows up and develops specific driver needs and levels, and ultimately efficiently converts drivers; drivers with different characteristics receive different types of marketing words, so that the corresponding drivers can personally experience the words related to themselves and are more likely to be converted by outbound calls.

[0087] In an embodiment of the present invention, the target driver data of the target driver is first obtained; wherein the target driver data includes personal information, behavior data and customer service call record information; then, the target driver data is predicted and processed based on the trained target prediction model to obtain the corresponding purchase prediction probability and target importance characteristics; then, it is determined whether the purchase prediction probability is greater than a preset threshold; if so, a corresponding target script is constructed based on the target importance characteristics and the target driver data; subsequently, a corresponding intelligent customer service is constructed based on the target script; finally, based on the intelligent customer service, the target script is used to execute the outbound call processing corresponding to the target driver. The present application predicts and processes the target driver data based on the use of a trained target prediction model to obtain the corresponding purchase prediction probability and target importance characteristics, and when it is detected that the purchase prediction probability is greater than a preset threshold, it automatically and intelligently constructs the corresponding target speech based on the target importance characteristics and the target driver data, and further builds the corresponding intelligent customer service based on the target speech, so that subsequently based on the intelligent customer service, the target speech is used to execute the outbound call processing corresponding to the target driver, thereby realizing the customization of targeted and matching target speech for drivers with different characteristics, and by controlling the intelligent customer service to use the target speech to execute the outbound call processing corresponding to the target driver, it is possible to communicate and interact with the target driver more effectively, effectively improving the intelligence of the outbound call processing, which is conducive to improving conversion efficiency and improving user experience.

[0088] Optionally, constructing a corresponding target speech based on the target importance feature and the target driver data includes:

[0089] Call the preset script library.

[0090] In this step, a certain number of characteristic groups that have passed the purchase invitation, that is, the manual invitation records of specific members, including telephone recordings, text records, etc., can be obtained. And the invitation records are analyzed in the process, including the analysis of the opening speech, the analysis of the recovery speech, the analysis of the speech of asking for the reason for rejection, the analysis of the rights and interests guidance speech, and the operation guidance speech. Specifically, the analysis of the opening speech includes: paying attention to how the inviter establishes contact with the driver and the opening speech they use. The analysis of the recovery speech includes: analyzing how the inviter recovers when the driver shows hesitation or refusal. The speech of asking for the reason for refusal includes: when the driver refuses, how the inviter politely asks the reason. The rights and interests guidance speech includes: how the inviter introduces the membership rights to the driver and how they emphasize the value of these rights. The analysis of the operation guidance speech includes: how the inviter guides them to complete the purchase process after the driver decides to purchase the membership. Then, according to the analysis results obtained by the process analysis of the invitation records, the speech corresponding to different characteristic groups is characterized and classified. For example, for drivers with high deposits but low behavior scores, it may be necessary to emphasize the additional protection and benefits of membership, while for drivers with high wallet balances, the money-saving advantage of membership can be highlighted. The data after feature classification is stored in a preset database to obtain the corresponding script library. The script library contains scripts prepared for drivers with different characteristics and needs, and is stored in a way that is easy to retrieve and match.

[0091] The designated speech matching the target importance feature is searched from the speech library, wherein the designated speech includes at least opening speech, redemption speech, speech asking for rejection reason, rights and interests guidance speech and operation guidance speech.

[0092] In this step, the matching standards can be formulated based on the importance characteristics predicted by the model and the needs of the driver. The matching standards can include characteristics such as the driver's age, gender, occupation, behavior score, wallet balance, and other elements such as the type, focus, and tone of the speech. By using a programming language or database query tool, you can search for speech that highly matches the above-mentioned target importance characteristics in the above-mentioned speech library. Among them, methods such as fuzzy matching, keyword search, or machine learning algorithms can be used to improve the accuracy and efficiency of matching. In addition, select the best speech from the searched matching speech. Give priority to those speeches that highly match the target importance characteristics and have been verified to have good results.

[0093] The designated speech is optimized based on the target driver data to obtain corresponding optimized speech.

[0094] In this step, the specific implementation process of optimizing the specified speech based on the target driver data to obtain the corresponding optimized speech will be further described in detail in subsequent specific embodiments of the present application and will not be elaborated here.

[0095] The optimized speech is used as the target speech.

[0096] In this step, the preset script library can be continuously optimized and updated according to needs to adapt to market changes and changes in customer needs.

[0097] In this embodiment, a preset script library is called; then the specified script that matches the target importance feature is queried from the script library; wherein the specified script at least includes opening script, recovery script, script asking for rejection reason, rights guidance script and operation guidance script; then the specified script is optimized based on the target driver data to obtain the corresponding optimized script; subsequently, the optimized script is used as the target script. This application queries the specified script that matches the target importance feature based on the use of the script library, and optimizes the specified script based on the target driver data to obtain the corresponding optimized script as the final target script. The target script that can be effectively obtained is a script that highly matches the target importance feature and has a good effect after optimization, which is conducive to improving the conversion rate of outbound calls and customer satisfaction.

[0098] Optionally, the optimizing the specified speech based on the target driver data to obtain the corresponding optimized speech includes:

[0099] Based on the target driver data, the designated speech is adjusted to obtain a corresponding first speech.

[0100] In this step, appropriate titles can be selected in the specified speech according to the age and gender of the target driver in the above target driver data, such as using friendly titles such as "handsome guy" and "beautiful girl" for young drivers, and using respectful titles such as "sir" and "madam" for older drivers, thereby obtaining the first speech after the title adjustment.

[0101] The tone of the first speech is adjusted to obtain a corresponding second speech.

[0102] In this step, the tone of the first speech can be adjusted according to the personality and behavioral characteristics of the target driver in the above target driver data, such as using a light-hearted and humorous tone for enthusiastic and cheerful drivers, and a gentle and sincere tone for introverted and cautious drivers, thereby obtaining a second speech with adjusted tone.

[0103] The wording of the second speech is optimized to obtain a corresponding third speech.

[0104] In this step, appropriate wording can be selected in the second speech according to the occupation and education level of the target driver in the above target driver data, such as using professional terms and precise expressions for highly educated drivers and using easy-to-understand language for low-educated drivers, thereby obtaining a third speech with optimized wording.

[0105] The third speech is used as the target speech.

[0106] In this embodiment, the designated speech is processed for address adjustment based on the target driver data to obtain the corresponding first speech; then the tone of voice is adjusted for the first speech to obtain the corresponding second speech; then the wording of the second speech is optimized to obtain the corresponding third speech; and the third speech is subsequently used as the target speech. This application efficiently and accurately completes the wording optimization processing of the designated speech by adjusting the address, tone of voice, and wording of the designated speech based on the use of target driver data, thereby ensuring that the details of the generated target speech are more in line with the driver's personality and preferences, thereby improving the conversion rate of outbound calls and customer satisfaction.

[0107] Optionally, the building of a corresponding intelligent customer service based on the target speech includes:

[0108] Call the preset initial intelligent customer service.

[0109] In this step, the initial intelligent customer service may specifically be a pre-built AI outbound call robot.

[0110] Determine the key nodes in the target speech.

[0111] In this step, the target speech can be analyzed to obtain key nodes in the target speech. There are multiple key nodes, including opening remarks, redemption speech, inquiry about rejection reasons, rights guidance, operation instructions, etc. These key nodes will serve as the basis for the initial intelligent customer service speech mapping.

[0112] Determine the flow conditions between the key nodes.

[0113] In this step, the flow conditions between each key node are determined according to the actual processing requirements. For example, when the driver shows the intention to refuse to purchase the membership, the intelligent customer service should jump to the node of redemption speech; when the driver asks about the membership rights, the intelligent customer service should jump to the node of rights guidance.

[0114] The initial intelligent customer service is mapped with speech based on the target speech, and the logic of the initial intelligent customer service is configured based on the flow condition to obtain the corresponding designated intelligent customer service.

[0115] In this step, the target speech can be mapped to the nodes of the initial intelligent customer service, including inputting specific opening speech, redemption speech, speech asking for rejection reasons, rights guidance speech, and operation guidance speech into the corresponding nodes of the initial intelligent customer service. In addition, the logic of the initial intelligent customer service is configured according to the flow conditions. Specifically, it includes the design of the dialogue process to ensure that the initial intelligent customer service can intelligently jump to the next node according to the driver's response.

[0116] Perform a performance test on the designated intelligent customer service.

[0117] In this step, the specific implementation process of the performance test of the designated intelligent customer service will be further described in detail in subsequent specific embodiments of the present application, and will not be elaborated in detail here.

[0118] If the designated intelligent customer service passes the performance test, the designated intelligent customer service will be used as the intelligent customer service.

[0119] In this embodiment, by calling the preset initial intelligent customer service; then determining the key nodes in the target speech; and determining the flow conditions between the key nodes; then mapping the speech of the initial intelligent customer service based on the target speech, and configuring the logic of the initial intelligent customer service based on the flow conditions to obtain the corresponding designated intelligent customer service; subsequently performing a performance test on the designated intelligent customer service; if the designated intelligent customer service passes the performance test, the designated intelligent customer service is used as the intelligent customer service. This application determines the key nodes in the target speech; and determines the flow conditions between the key nodes, then mapping the speech of the initial intelligent customer service based on the target speech, and configuring the logic of the initial intelligent customer service based on the flow conditions to obtain the corresponding designated intelligent customer service; performing a performance test on the designated intelligent customer service, and using the designated intelligent customer service that passes the test as the final intelligent customer service, thereby realizing the efficient and accurate construction of an intelligent customer service that corresponds to the target driver and can be applied to the target driver's driver characteristics and performs outbound call processing with a speech that matches the target driver's characteristics, and then based on the use of the target driver, it can provide the target driver with an outbound call service that meets the needs, thereby greatly improving the conversion rate of outbound calls.

[0120] Optionally, the performing a performance test on the designated intelligent customer service includes:

[0121] Perform a recognition accuracy test on the designated intelligent customer service.

[0122] In this step, the recognition accuracy refers to the proportion of the specified intelligent customer service that correctly recognizes and understands the user's intention. Specifically, the recognition accuracy test includes: first designing test cases, including positive cases (specifying common user inputs that the intelligent customer service should correctly recognize), negative cases (specifying inputs that the intelligent customer service can misunderstand, such as spelling errors, grammatical errors, irrelevant topics, etc.), and edge cases (specifying the performance of the intelligent customer service under specific conditions, such as low resources and high concurrency). Then, perform automated testing according to the test cases, that is, use scripts or test frameworks to batch execute test cases, and then calculate the accuracy, that is, the proportion of correctly recognized test cases to the total number of test cases. If the accuracy is greater than or equal to the preset accuracy threshold, the specified intelligent customer service is determined to have passed the recognition accuracy test, and if the accuracy is less than the accuracy threshold, otherwise the specified intelligent customer service is determined to have failed the recognition accuracy test. Among them, there is no specific limitation on the value of the above-mentioned accuracy threshold, which can be set according to actual business needs.

[0123] If the designated intelligent customer service passes the recognition accuracy test, a conversation fluency test is performed on the designated intelligent customer service.

[0124] In this step, conversation fluency refers to the ability of the designated intelligent customer service to maintain logical coherence and contextual consistency in the conversation. Specifically, the process of conducting a conversation fluency test on the designated intelligent customer service includes: testing the performance of the designated intelligent customer service in multiple consecutive rounds of conversations, testing the ability of the designated intelligent customer service to maintain contextual consistency in the conversation, and testing the response of the intelligent service when the user changes the topic. Then, the tester conducts a simulated conversation with the designated intelligent customer service and records the conversation process. Invite real users to participate in the test and collect their evaluation of the conversation fluency. Then, evaluate the conversation fluency of the designated intelligent customer service based on the conversation records and user feedback. If the conversation fluency meets the preset fluency expectations, the designated intelligent customer service is determined to have passed the conversation fluency test, otherwise, the designated intelligent customer service is determined to have failed the conversation fluency test.

[0125] If the designated intelligent customer service passes the conversation fluency test, a response time test is performed on the designated intelligent customer service.

[0126] In this step, the above-mentioned response time refers to the time taken by the designated intelligent customer service from receiving user input to generating and sending a reply. Specifically, the process of testing the response time of the designated intelligent customer service includes: testing the response time of the designated intelligent customer service under a single user request, then testing the response time of the designated intelligent customer service under high concurrent requests, and testing the response time of the designated intelligent customer service under resource constraints (such as CPU, memory). Then, use a test tool or script to simulate user requests and record the response time of the designated intelligent customer service. And use monitoring tools to track the performance indicators of the designated intelligent customer service in real time, such as CPU utilization, memory occupancy, etc. Analyze the results and calculate the average response time. If the average response time is less than the preset time threshold, it is determined that the designated intelligent customer service has passed the response time test, otherwise it is determined that the designated intelligent customer service has not passed the response time test.

[0127] If the designated intelligent customer service passes the response time test, it is determined that the designated intelligent customer service has passed the test; otherwise, it is determined that the designated intelligent customer service has failed the test.

[0128] In this step, only when it is detected that the designated intelligent customer service has passed the recognition accuracy test, the conversation fluency test and the response time test at the same time, will it be determined that the designated intelligent customer service has passed the performance test. If the designated intelligent customer service has failed at least one test, such as the recognition accuracy test, the conversation fluency test, or the response time test, then it is determined that the designated intelligent customer service has failed the performance test.

[0129] In this embodiment, the designated intelligent customer service is tested for recognition accuracy; if the designated intelligent customer service passes the recognition accuracy test, the designated intelligent customer service is tested for conversation fluency; if the designated intelligent customer service passes the conversation fluency test, the designated intelligent customer service is tested for response time; if the designated intelligent customer service passes the response time test, the designated intelligent customer service is determined to have passed the test, otherwise the designated intelligent customer service is determined to have failed the test. This application performs recognition accuracy tests, conversation fluency tests, and conversation fluency tests on designated intelligent customer service, thereby achieving efficient and accurate performance testing of designated intelligent customer service from a high dimension, effectively ensuring the accuracy of the performance test results of the generated designated intelligent customer service, and then accurately constructing an intelligent customer service that corresponds to the target driver and can be applied to the target driver's driver characteristics to process outbound calls, so that subsequent use based on the target driver can provide the target driver with outbound call services that meet the needs, thereby greatly helping to improve the conversion rate of outbound calls.

[0130] Optionally, before performing prediction processing on the target driver data based on the trained target prediction model to obtain corresponding purchase prediction probability and target importance features, the method further includes:

[0131] Obtain preset business goals and collect corresponding driver data based on the business goals.

[0132] In this step, the above business objectives can be set according to the actual outbound call needs, such as inviting drivers to purchase membership. This business objective will guide subsequent data collection and model training.

[0133] The driver data is preprocessed and feature-engineered to obtain corresponding sample data.

[0134] In this step, relevant driver data, i.e., driver's personal information, behavioral characteristics, customer service line data, etc., can be collected according to business objectives. The specific types of driver data can refer to the above-mentioned target driver types, and these data can come from the company's database, CRM system or third-party data source. Then the collected driver data is preprocessed and feature engineered, including data cleaning, missing value processing, outlier detection, feature selection, feature scaling, etc. Ensure the quality and consistency of the data. And further determine the input features: According to the business logic and data availability, select features that are helpful for model prediction as input. For example, the driver's age, gender, deposit, behavioral points, wallet balance, vehicle type, etc. And determine the output: Determine the output of the model according to the business objectives. In this embodiment, the output is whether the driver purchases a membership, which is a binary classification problem.

[0135] Recall the preset initial model.

[0136] In this step, a suitable model can be selected as the above initial model according to the type of problem and the characteristics of the data. Since it is a binary classification problem and the data may contain complex nonlinear relationships, tree models such as XGBoost, LightGBM or GBDT are selected for training. Among them, the model configuration includes: configuring the parameters of the model, such as learning rate, maximum depth, number of leaf nodes, etc. These parameters can be optimized through methods such as cross-validation and grid search.

[0137] Based on the business objectives, the initial model is trained using the sample data to obtain a specified model that meets expectations.

[0138] In this step, training is performed using the processed sample data and the configured initial model. During the training process, the initial model learns the mapping relationship between input features and outputs. After training is completed, the performance of the model is evaluated using the validation set. Evaluation indicators may include accuracy, recall, F1 score, etc. Based on the evaluation results, the parameters of the initial model are adjusted to obtain a specified model with expected performance.

[0139] The designated model is used as the target prediction model.

[0140] In this step, the driver data corresponding to the new driver is predicted by using the target prediction model whose performance meets expectations. The output predictable results include the probability of the driver purchasing a membership and the features that the model considers important, namely, importance features.

[0141] In this embodiment, by obtaining a preset business goal, and collecting corresponding driver data based on the business goal; then preprocessing and feature engineering the driver data to obtain corresponding sample data; then calling the preset initial model; subsequently based on the business goal, using the sample data to train the initial model to obtain a specified model that meets expectations; finally, using the specified model as the target prediction model. This application obtains business goals, collects corresponding driver data based on the business goals, and preprocesses and feature engineering the driver data to obtain sample data, and then based on the business goals, uses the sample data to train the initial model to obtain a target prediction model that meets expectations, which effectively improves the construction efficiency of the target prediction model and ensures the model effect of the target prediction model obtained.

[0142] Optionally, after the outbound call processing corresponding to the target driver is performed based on the intelligent customer service using the target speech, the method further includes:

[0143] After completing the outbound call processing, the call data between the intelligent customer service and the target driver is obtained.

[0144] In this step, after the outbound call processing is completed, the call data between the intelligent customer service and the target driver can be further recorded and obtained.

[0145] The call data is analyzed and processed to obtain corresponding analysis results.

[0146] In this step, the call data is analyzed to obtain corresponding analysis results, including conversion success or conversion failure.

[0147] Based on the analysis result, follow-up processing corresponding to the target driver is performed.

[0148] In this step, if the analysis result is that the conversion failed, a follow-up plan will be formulated based on the feedback. If the analysis result is that the conversion was successful, caring information will be sent to the converted drivers regularly to improve satisfaction and loyalty.

[0149] In this embodiment, after the outbound call processing is completed, the call data between the intelligent customer service and the target driver is obtained; then the call data is analyzed and processed to obtain the corresponding analysis results; and then the follow-up processing corresponding to the target driver is performed based on the analysis results. After completing the outbound call processing, this application will further analyze and process the call data to obtain the corresponding analysis results, and then intelligently perform follow-up processing corresponding to the target driver based on the analysis results, which is conducive to improving the conversion rate of outbound calls and the personal satisfaction of drivers.

[0150] Further references Figure 3 , as a response to the above Figure 2 In order to realize the method shown in the figure, the present application provides an embodiment of an outbound call processing device 300, which is similar to Figure 2 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0151] An outbound call processing device 300 provided in an embodiment of the present invention includes:

[0152] The first acquisition module 310 is used to acquire target driver data of a target driver; wherein the target driver data includes personal information, behavior data, and customer service call record information;

[0153] Prediction module 320, used to perform prediction processing on the target driver data based on the trained target prediction model to obtain corresponding purchase prediction probability and target importance characteristics;

[0154] A determination module 330 is used to determine whether the purchase prediction probability is greater than a preset threshold;

[0155] A construction module 340 is used to construct a corresponding target speech based on the target importance feature and the target driver data;

[0156] A building module 350 is used to build a corresponding intelligent customer service based on the target speech;

[0157] The first processing module 360 ​​is used to perform outbound call processing corresponding to the target driver based on the intelligent customer service and using the target speech.

[0158] Optionally, the construction module 340 includes:

[0159] The first calling submodule is used to call the preset speech library;

[0160] A query submodule, used to query the specified speech matching the target importance feature from the speech library; wherein the specified speech includes at least opening speech, redemption speech, speech asking for rejection reason, rights and interests guidance speech and operation guidance speech;

[0161] An optimization submodule, used for performing speech optimization processing on the specified speech based on the target driver data to obtain corresponding optimized speech;

[0162] The first determination submodule is used to use the optimized speech as the target speech.

[0163] Optionally, the optimization submodule includes:

[0164] A first processing unit is used to adjust the address of the specified speech based on the target driver data to obtain a corresponding first speech;

[0165] A second processing unit is used to adjust the tone of the first speech to obtain a corresponding second speech;

[0166] A third processing unit is used to optimize the wording of the second speech to obtain a corresponding third speech;

[0167] A determination unit is used to use the third speech as the target speech.

[0168] Optionally, the building module 350 includes:

[0169] The second calling submodule is used to call the preset initial intelligent customer service;

[0170] A second determination submodule is used to determine the key nodes in the target speech;

[0171] A third determination submodule is used to determine the flow conditions between the key nodes;

[0172] A processing submodule, configured to map the initial intelligent customer service to a speech based on the target speech, and to configure the logic of the initial intelligent customer service based on the flow condition to obtain a corresponding designated intelligent customer service;

[0173] A testing submodule, used to perform a performance test on the designated intelligent customer service;

[0174] The fourth determination submodule is used to use the designated intelligent customer service as the intelligent customer service if the designated intelligent customer service passes the performance test.

[0175] Optionally, the test submodule includes:

[0176] A first testing unit, used to test the recognition accuracy of the designated intelligent customer service;

[0177] A second testing unit is used to perform a conversation fluency test on the designated intelligent customer service if the designated intelligent customer service passes the recognition accuracy test;

[0178] A third testing unit is configured to perform a response time test on the designated intelligent customer service if the designated intelligent customer service passes the conversation fluency test;

[0179] The determination unit is configured to determine that the designated intelligent customer service has passed the test if the designated intelligent customer service has passed the response time test, and otherwise determine that the designated intelligent customer service has failed the test.

[0180] Optionally, the outbound call processing device 300 further includes:

[0181] A second acquisition module is used to acquire a preset business target and collect corresponding driver data based on the business target;

[0182] A second processing module is used to perform preprocessing and feature engineering on the driver data to obtain corresponding sample data;

[0183] A calling module is used to call the preset initial model;

[0184] A training module, used to train the initial model using the sample data based on the business goal to obtain a specified model that meets expectations;

[0185] A determination module is used to use the specified model as the target prediction model.

[0186] Optionally, the outbound call processing device 300 further includes:

[0187] A third acquisition module is used to obtain the call data between the intelligent customer service and the target driver after completing the outbound call processing;

[0188] An analysis module, used to analyze and process the call data to obtain corresponding analysis results;

[0189] A follow-up module is used to perform follow-up processing corresponding to the target driver based on the analysis result.

[0190] To solve the above technical problems, the present application also provides a computer device. Figure 4 , Figure 4 This is a basic structural block diagram of the computer device in this embodiment.

[0191] The computer device 4 includes a memory 41, a processor 42, and a network interface 43 that are interconnected through a system bus. It should be noted that the figure only shows a computer device 4 with components 41-43, but it should be understood that it is not required to implement all the components shown, and more or fewer components can be implemented instead. Among them, those skilled in the art can understand that the computer device here is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application specific integrated circuits (Application Specific Integrated Circuit, ASIC), programmable gate arrays (Field-Programmable Gate Array, FPGA), digital processors (Digital Signal Processor, DSP), embedded devices, etc.

[0192] The computer device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The computer device may interact with a user through a keyboard, a mouse, a remote controller, a touch pad, or a voice control device.

[0193] The memory 41 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 41 can be an internal storage unit of the computer device 4, such as a hard disk or memory of the computer device 4. In other embodiments, the memory 41 can also be an external storage device of the computer device 4, such as a plug-in hard disk equipped on the computer device 4, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (FlashCard), etc. Of course, the memory 41 can also include both the internal storage unit of the computer device 4 and its external storage device. In this embodiment, the memory 41 is generally used to store the operating system and various application software installed on the computer device 4, such as the program code of the outbound call processing method, etc. In addition, the memory 41 can also be used to temporarily store various types of data that have been output or are to be output.

[0194] The processor 42 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chips in some embodiments. The processor 42 is generally used to control the overall operation of the computer device 4. In this embodiment, the processor 42 is used to run the program code stored in the memory 41 or process data, such as running the program code of the outbound call processing method.

[0195] The network interface 43 may include a wireless network interface or a wired network interface. The network interface 43 is generally used to establish a communication connection between the computer device 4 and other electronic devices.

[0196] The present application also provides another implementation, namely, providing a computer-readable storage medium, wherein the computer-readable storage medium stores the application crash processing program, and the application crash processing program can be executed by at least one processor to enable the at least one processor to perform the steps of the outbound call processing method as described above.

[0197] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware online 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, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0198] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0199] Obviously, the embodiments described above are only some embodiments of the present application, rather than all embodiments. The preferred embodiments of the present application are given in the accompanying drawings, but they do not limit the patent scope of the present application. The present application can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present application more thorough and comprehensive. Although the present application is described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions recorded in the aforementioned specific implementation methods, or to perform equivalent replacement of some of the technical features therein. Any equivalent structure made using the contents of the specification and drawings of this application, directly or indirectly used in other related technical fields, is similarly within the scope of patent protection of this application.

Claims

1. An outbound call processing method, characterized in that: include: Obtain target driver data of the target driver; wherein the target driver data includes personal information, behavior data, and customer service call record information; Based on the trained target prediction model, the target driver data is predicted and processed to obtain corresponding purchase prediction probability and target importance characteristics; Determining whether the purchase prediction probability is greater than a preset threshold; If yes, construct corresponding target speech based on the target importance feature and the target driver data; Build corresponding intelligent customer service based on the target words; Based on the intelligent customer service, the target script is used to execute outbound call processing corresponding to the target driver.

2. The method according to claim 1, characterized in that The constructing corresponding target speech based on the target importance feature and the target driver data includes: Call the preset script library; Querying the designated speech that matches the target importance feature from the speech library; wherein the designated speech at least includes opening speech, redemption speech, speech asking for reasons for rejection, rights and interests guidance speech, and operation guidance speech; Performing speech optimization processing on the specified speech based on the target driver data to obtain corresponding optimized speech; The optimized speech is used as the target speech.

3. The method according to claim 2, characterized in that The optimizing the specified speech based on the target driver data to obtain the corresponding optimized speech includes: Based on the target driver data, the designated speech is processed for address adjustment to obtain a corresponding first speech; Performing tone adjustment processing on the first speech to obtain a corresponding second speech; Optimizing the wording of the second speech to obtain a corresponding third speech; The third speech is used as the target speech.

4. The method according to claim 1, characterized in that The corresponding intelligent customer service is built based on the target speech, including: Call the preset initial intelligent customer service; Determine the key nodes in the target speech; Determining the flow conditions between the key nodes; Based on the target speech, the initial intelligent customer service is mapped to a speech, and based on the flow condition, the logic of the initial intelligent customer service is configured to obtain a corresponding designated intelligent customer service; Performing performance testing on the designated intelligent customer service; If the designated intelligent customer service passes the performance test, the designated intelligent customer service will be used as the intelligent customer service.

5. The method according to claim 4, characterized in that The performance test of the designated intelligent customer service comprises: Conducting a recognition accuracy test on the designated intelligent customer service; If the designated intelligent customer service passes the recognition accuracy test, a conversation fluency test is performed on the designated intelligent customer service; If the designated intelligent customer service passes the conversation fluency test, a response time test is performed on the designated intelligent customer service; If the designated intelligent customer service passes the response time test, it is determined that the designated intelligent customer service has passed the test; otherwise, it is determined that the designated intelligent customer service has failed the test.

6. The method according to claim 1, characterized in that Before performing prediction processing on the target driver data based on the trained target prediction model to obtain the corresponding purchase prediction probability and target importance features, the method further includes: Obtaining preset business goals, and collecting corresponding driver data based on the business goals; Preprocessing and feature engineering the driver data to obtain corresponding sample data; Call the preset initial model; Based on the business goal, the initial model is trained using the sample data to obtain a specified model that meets expectations; The designated model is used as the target prediction model.

7. The method according to claim 1, characterized in that After the outbound call processing corresponding to the target driver is performed based on the intelligent customer service using the target speech, the method further includes: After completing the outbound call processing, obtaining the call data between the intelligent customer service and the target driver; Analyze and process the call data to obtain corresponding analysis results; Based on the analysis result, follow-up processing corresponding to the target driver is performed.

8. An outbound call processing device, characterized in that: include: A first acquisition module is used to acquire target driver data of a target driver; wherein the target driver data includes personal information, behavior data, and customer service call record information; A prediction module, used to perform prediction processing on the target driver data based on the trained target prediction model to obtain corresponding purchase prediction probability and target importance characteristics; A judgment module, used to judge whether the purchase prediction probability is greater than a preset threshold; A construction module, for constructing corresponding target speech based on the target importance feature and the target driver data; A building module is used to build corresponding intelligent customer service based on the target speech; The first processing module is used to execute the outbound call processing corresponding to the target driver using the target speech based on the intelligent customer service.

9. A computer device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the outbound call processing method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the outbound call processing method according to any one of claims 1 to 7 are implemented.

Images