City grouping-based outbound management method, apparatus and device, and storage medium

Abstract

The invention discloses an outbound call management method and device based on city grouping, equipment and a storage medium. The method comprises the steps that historical dialogue texts between an AI robot and all users in different cities are acquired; determining task conversion rates of different cities; performing dialogue analysis on the historical dialogue text of each city by using a preset large model to obtain a dialogue analysis result; according to the dialogue analysis result and the task conversion rate, performing grouping processing on each city to obtain a city grouping result; and based on a city grouping result, constructing outbound call skills of the AI robots of different cities so as to use the outbound call skills to carry out dialogue with the user. According to the method, dialogue recognition is carried out on the historical dialogue text by using the large model, grouping processing is carried out on each city in combination with the recognized features and the task conversion rate of each city, different outbound calling means are used for management for different types of cities, and the online conversion rate of each city is effectively improved.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to an outbound call management method, device, equipment and storage medium based on city grouping. Background Art

[0002] In the field of voice dialogue business, we usually deploy AI robots on the platform to interact with users to achieve efficient customer service or marketing goals. Take the recommendation of membership service as an example. When conducting outbound marketing, AI robots need to explain in detail the various benefits of becoming a member to users and skillfully guide users to complete the purchase during the conversation. In this process, AI robots usually follow a set of standard words to communicate with users in order to achieve the best sales results.

[0003] However, due to the significant differences in market environment, consumption habits, competitive situation, etc. among different types of cities, if AI robots use the same rhetoric for all cities, they may ignore the specific needs and pain points of users in specific cities, thereby affecting sales conversion rates.

[0004] Secondly, users in different cities have different levels of acceptance of company brands and membership payment models. A universal set of rhetoric may not be able to fully address the concerns of specific user groups, especially when they have different views on the cost-effectiveness, service content, and long-term value of paid memberships. In this case, users may be unwilling to pay membership fees due to concerns or distrust, which directly leads to a decrease in sales conversion rate. Summary of the invention

[0005] Based on this, it is necessary to provide an outbound call management method, device, equipment and storage medium based on city grouping to address the above technical problems, so as to solve at least one of the above technical problems.

[0006] The present invention provides an outbound call management method based on city grouping, comprising:

[0007] Obtain historical conversation texts between the AI ​​robot and various users in different cities;

[0008] Determine the task conversion rates corresponding to different cities;

[0009] Using the preset large model, the historical dialogue texts of each city are analyzed to obtain the dialogue analysis results;

[0010] Based on the dialogue analysis results and the task conversion rate, each city is grouped to obtain a city grouping result;

[0011] Based on the city clustering results, outbound call scripts for AI robots in different cities are constructed to use the outbound call scripts to communicate with users.

[0012] Optionally, according to an outbound call management method based on city grouping provided by the present invention, based on the city grouping result, constructing outbound call scripts of AI robots in different cities, so as to use the outbound call scripts to communicate with users, including:

[0013] Based on each of the city grouping results, construct an outbound call script of an AI robot corresponding to different cities according to the pre-collected manual scripts;

[0014] Determine the city location corresponding to the user to be called out;

[0015] Based on the city location, a corresponding AI robot is selected to communicate with the user to be called out.

[0016] Optionally, according to an outbound call management method based on city grouping provided by the present invention, the step of selecting a corresponding AI robot to conduct a conversation with the outbound call user based on the city location includes:

[0017] Determine the cluster to which the city location belongs;

[0018] Based on the group to which the city location belongs, an AI robot corresponding to the group is selected to communicate with the user to be called.

[0019] Optionally, according to an outbound call management method based on city grouping provided by the present invention, the conversation analysis result includes an intention recognition result and multi-dimensional indicator features;

[0020] The preset large model is used to perform dialogue analysis on the historical dialogue texts of each city to obtain dialogue analysis results, including:

[0021] Classifying the cities based on the task conversion rates corresponding to the different cities;

[0022] Using the preset large model, the intent recognition is performed on the historical conversation texts of each city corresponding to each classification type to obtain the intent recognition results;

[0023] Use the preset large model to extract the multi-dimensional indicator features corresponding to each historical conversation text.

[0024] Optionally, according to an outbound call management method based on city grouping provided by the present invention, the method of extracting multi-dimensional indicator features corresponding to each historical conversation text by using a preset large model includes:

[0025] For any city:

[0026] Based on each of the historical conversation texts, extracting behavior features and conversation features;

[0027] Performing sentiment analysis on each of the historical dialogue texts to obtain sentiment distribution features;

[0028] The multi-dimensional indicator feature is formed based on the behavior feature, the conversation feature, and the emotion distribution feature.

[0029] Optionally, according to an outbound call management method based on city grouping provided by the present invention, the city grouping results are obtained by grouping the cities based on the conversation analysis results and the task conversion rate, including:

[0030] Compare and analyze the intent recognition results corresponding to the historical dialogue texts of each city;

[0031] Based on the comparison results, target feature information is extracted from the multi-dimensional indicator features;

[0032] Based on the target feature information and the task conversion rate, each city is grouped to obtain a city grouping result.

[0033] Optionally, according to an outbound call management method based on city grouping provided by the present invention, the determining of task conversion rates corresponding to different cities includes:

[0034] For each city;

[0035] Determine the number of users in the city who have engaged in conversations with the AI ​​bot;

[0036] Determine the outbound call result corresponding to each historical conversation text in the city;

[0037] Determine the number of completed tasks based on the outbound call results;

[0038] The task conversion rate is determined based on the number of users and the number of completed tasks.

[0039] The present invention also provides an outbound call management device based on city grouping, comprising:

[0040] The acquisition module is used to obtain the historical conversation texts between the AI ​​robot and various users in different cities;

[0041] A determination module is used to determine the task conversion rates corresponding to different cities;

[0042] An analysis module is used to use a preset large model to conduct dialogue analysis on the historical dialogue texts of each city to obtain dialogue analysis results;

[0043] A grouping module, used to group each city based on the dialogue analysis result and the task conversion rate to obtain a city grouping result;

[0044] A construction module is used to construct outbound call scripts of AI robots in different cities based on the city grouping results, so as to use the outbound call scripts to communicate with users.

[0045] The present invention also provides a computer device, including a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, wherein the processor implements the above-mentioned outbound call management method based on city grouping when executing the computer-readable instructions.

[0046] The present invention also provides one or more readable storage media storing computer-readable instructions, and when the computer-readable instructions are executed by a processor, the above-mentioned outbound call management method based on city grouping is implemented.

[0047] The above-mentioned outbound call management method, device, equipment and storage medium based on city clustering, the present invention uses a large model to perform dialogue recognition on historical dialogue texts to obtain dialogue analysis results; based on the dialogue analysis results and the task conversion rate of each city, each city is grouped, and based on the city clustering results, the outbound call scripts of AI robots in different cities are constructed, so as to use different outbound call means for management of different types of cities, effectively improving the online conversion rate of each city. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.

[0049] Figure 1 It is a flow chart of an outbound call management method based on city grouping in one embodiment of the present invention;

[0050] Figure 2 It is a structural diagram of an outbound call management device based on city grouping in one embodiment of the present invention;

[0051] Figure 3 is a schematic diagram of a computer device in one embodiment of the present invention. DETAILED DESCRIPTION

[0052] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0053] The terms used in one or more embodiments of the present invention are only for the purpose of describing specific embodiments, and are not intended to limit one or more embodiments of the present invention. The singular forms of "a", "said" and "the" used in one or more embodiments of the present invention are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used in one or more embodiments of the present invention refers to and includes any or all possible combinations of one or more associated listed items.

[0054] The outbound call management method based on city grouping provided in the embodiment of the present invention can be applied in different outbound call scenarios, for example, adjusting the sales talk to meet the specific needs of users in different cities, user authentication, using appropriate talk to remind and persuade users in different cities to pay a deposit, user churn recall, using different talk to sell cars to drivers in different types of cities, and other scenarios.

[0055] In one embodiment, specifically, Figure 1 As shown, Figure 1 FIG. 1 is a flow chart of an outbound call management method based on city grouping in an embodiment of the present invention. The embodiment of the present invention provides an outbound call management method based on city grouping, comprising the following steps:

[0056] Step S11, obtaining historical conversation texts between the AI ​​robot and various users in different cities;

[0057] Specifically, the AI ​​robot uses the same dialogue to communicate with users in various cities, collects historical conversation records between the AI ​​robot and users in different cities, and then performs text recognition on each historical conversation record to obtain the historical conversation text.

[0058] Step S12, determining the task conversion rates corresponding to different cities;

[0059] It should be noted that after the AI ​​robot makes an outbound call to a user, the platform will record the outbound call results of the current conversation. For example, in the scenario where the AI ​​robot sells memberships, it will record whether the user accepts to open a membership.

[0060] Specifically, the following operations are performed for each city: the number of users who have had conversations with the AI ​​robot in the city is determined; and the number of task completions is determined based on the outbound call results corresponding to each historical conversation text in the city, for example, the number of transactions accepted for membership activation is determined; further, based on the number of users and the number of task completions, the task conversion rate is determined, so that the user conversion situation in each city can be understood.

[0061] Step S13, using a preset large model, performing dialogue analysis on the historical dialogue texts of each city to obtain dialogue analysis results;

[0062] It should be noted that the dialogue analysis results include intention recognition results and multi-dimensional indicator features; the multi-dimensional indicator features include behavioral features, dialogue features and emotion distribution features, among which behavioral features include connection rate, hang-up rate, silence rate and task conversion time. Emotion distribution features include positive emotion distribution results and negative emotion distribution results. Positive emotion distribution results refer to the proportion distribution of positive emotions (such as happiness, satisfaction, etc.) in the overall text after sentiment analysis of the dialogue text, and negative emotion distribution results refer to the proportion distribution of negative emotions (such as anger, dissatisfaction, etc.) in the overall text after sentiment analysis of the dialogue text. Behavioral features include user question rate, dialogue time length, dialogue turns and dialogue intention classification.

[0063] Specifically, in one embodiment, the historical dialogue texts of each city are subjected to intent recognition using a preset large model to obtain intent recognition results; and the multi-dimensional indicator features corresponding to each historical dialogue text are extracted using the preset large model. In another embodiment, each city is classified based on the task conversion rates corresponding to the different cities, for example, divided into three types of cities with high conversion rates, medium conversion rates, and low conversion rates, and then the historical dialogue texts of each city corresponding to each classification type are subjected to intent recognition using the preset large model to obtain intent recognition results; and the multi-dimensional indicator features corresponding to each historical dialogue text are extracted using the preset large model.

[0064] Step S14, based on the dialogue analysis result and the task conversion rate, grouping the cities to obtain city grouping results;

[0065] It should be noted that clustering is an unsupervised learning technique used to group similar data points together to form different clusters. For example, clustering is performed using algorithms such as cosine similarity and clustering algorithms.

[0066] Specifically, according to the intention recognition results in the dialogue analysis results, the intention differences of cities of different classification types are analyzed, and then according to the intention differences of different cities, the target features that can group the cities are extracted from the multi-dimensional indicator features. Furthermore, each city is grouped in combination with the target features and task conversion rate to obtain the city clustering results.

[0067] Step S15, based on the city grouping results, construct outbound call scripts for AI robots in different cities, so as to use the outbound call scripts to communicate with users.

[0068] Specifically, for each of the city clustering results, a corresponding outbound call script of the AI ​​robot is constructed. For example, for the city clusters that question the credibility of things, the focus of the outbound call script is to convey real information and increase credibility. For the city clusters that are sensitive to prices, the focus of the outbound call script is to continuously convey low prices and high returns. Therefore, different outbound call scripts are used for different types of cities to communicate with users.

[0069] The embodiment of the present invention, through the above scheme, includes: obtaining historical conversation texts between the AI ​​robot and each user in different cities; determining the task conversion rate corresponding to different cities; using a preset large model to perform conversation analysis on the historical conversation texts of each city to obtain conversation analysis results; based on the conversation analysis results and the task conversion rate, grouping each city to obtain a city grouping result; based on the city grouping result, constructing outbound call speech for the AI ​​robot in different cities, so as to use the outbound call speech to communicate with users. The embodiment of the present invention uses a large model to perform conversation recognition on historical conversation texts, combines the recognized features and the task conversion rate of each city, groups each city, and then uses different outbound call methods to manage different types of cities, effectively improving the online conversion rate of each city.

[0070] In one embodiment of the present invention, based on the city grouping results, outbound call scripts of AI robots in different cities are constructed to use the outbound call scripts to communicate with users, including:

[0071] Based on each of the city grouping results, the outbound call scripts of the AI ​​robots corresponding to different cities are constructed according to the pre-collected manual scripts; the city location corresponding to the outbound call user is determined; based on the city location, the corresponding AI robot is selected to communicate with the outbound call user.

[0072] Specifically, for each group in the city grouping results: Combined with the pre-collected manual scripts, construct the outbound call scripts of the AI ​​robot corresponding to each city in each group, so as to use different outbound call scripts for different cities to manage outbound calls. Furthermore, after building the outbound call scripts of the AI ​​robot, for the users to be called out (for example, newly registered users), it is necessary to determine the city location corresponding to the users to be called out (for example, newly registered users), and then determine the group to which the city location belongs, so as to select the AI ​​robot corresponding to the group based on the group to which the city location belongs, so as to communicate with the users to be called out according to the outbound call scripts corresponding to the AI ​​robot.

[0073] The embodiment of the present invention, through the above scheme, includes: based on each group in the city grouping results, constructing the outbound call speech of the AI ​​robot corresponding to different cities according to the pre-collected manual speech; determining the city landing point corresponding to the user to be called out; based on the city landing point, selecting the corresponding AI robot to have a conversation with the user to be called out. After the city grouping is performed, the embodiment of the present invention constructs the outbound call speech of the AI ​​robot corresponding to each city in each group, and then according to the city landing point corresponding to the user to be called out, based on the city landing point, selects the corresponding AI robot to have a conversation with the user to be called out. It realizes the use of different outbound call speech methods for different cities to carry out outbound call management, effectively improving the online conversion rate of each city.

[0074] In one embodiment of the present invention, a preset large model is used to perform dialogue analysis on the historical dialogue texts of various cities to obtain dialogue analysis results, including:

[0075] Based on the task conversion rates corresponding to the different cities, the cities are classified; using the preset big model, the historical conversation texts of each city corresponding to each classification type are subjected to intent recognition to obtain intent recognition results; and the multi-dimensional indicator features corresponding to each historical conversation text are extracted using the preset big model.

[0076] Specifically, based on the task conversion rate corresponding to each city, each city is classified according to a preset conversion rate range, wherein the conversion rate range can be set according to the actual situation, and is not specifically limited here. For example, the city is divided into three types: high conversion rate, medium conversion rate, and low conversion rate. The historical conversation texts of each city corresponding to each classification type are subjected to intention recognition using a preset large model to obtain an intention recognition result; in addition, for each city: the behavior features and conversation features corresponding to each historical conversation text are extracted using a preset large model, and the emotion analysis of each of the historical conversation texts is performed using a preset large model to obtain an emotion distribution feature, wherein the behavior features include connection rate, hang-up rate, silence rate, and task conversion time. The emotion distribution feature includes positive emotion distribution results and negative emotion distribution results. The behavior features include user question rate, conversation time length, conversation turn, and conversation intention classification. Further, based on the behavior features, the conversation features, and the emotion distribution features, the multi-dimensional indicator features are formed.

[0077] The embodiment of the present invention adopts the above scheme, including: classifying each city based on the task conversion rate corresponding to the different cities; using the preset big model to perform intent recognition on the historical conversation texts of each city corresponding to each classification type to obtain intent recognition results; using the preset big model to extract multi-dimensional indicator features corresponding to each historical conversation text. It realizes full use of the excellent natural language understanding ability of the big model, analyzes the conversation features of different types of cities, and extracts features that can group cities.

[0078] In one embodiment of the present invention, based on the dialogue analysis result and the task conversion rate, each city is grouped to obtain a city grouping result, including:

[0079] A comparative analysis is performed on the intention recognition results corresponding to the historical conversation texts of each city; based on the comparative results, target feature information is extracted from the multi-dimensional indicator features; based on the target feature information and the task conversion rate, each city is grouped to obtain a city grouping result.

[0080] Specifically, the intent recognition results corresponding to each historical dialogue text of each city are classified, and then the intent classification results corresponding to different cities are compared and analyzed to analyze the intention differences of cities of different classification types; further, according to the analysis results, target feature information with large numerical differences is extracted from the multi-dimensional indicator features, that is, not all features are used to perform city clustering. For the multi-dimensional features of each city, features with large numerical differences between cities are selected for city clustering. Then, the intent recognition results, target feature information and task conversion rates of each city are combined to perform clustering on each city to obtain city clustering results.

[0081] The embodiment of the present invention, through the above scheme, includes: comparative analysis of the intention recognition results corresponding to the historical dialogue texts of each city; based on the comparative results, extracting the target feature information from the multi-dimensional indicator features; based on the target feature information and the task conversion rate, grouping each city to obtain a city grouping result. The use of a large model to classify the intentions of the historical dialogue texts between AI and drivers is realized, and then the cities are grouped in combination with the characteristics of the city's task conversion rate, so that different outbound call methods are used for outbound call management for cities in different groups.

[0082] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0083] In one embodiment, an outbound call management device based on city groups is provided, and the outbound call management device based on city groups corresponds one-to-one to the outbound call management method based on city groups in the above embodiment. Figure 2 As shown, Figure 2 1 is a schematic diagram of a structure of an outbound call management device based on city groups in one embodiment of the present invention. The outbound call management device based on city groups includes:

[0084] An acquisition module 21 is used to acquire historical conversation texts between the AI ​​robot and various users in different cities;

[0085] A determination module 22, used to determine the task conversion rates corresponding to different cities;

[0086] An analysis module 23 is used to perform dialogue analysis on the historical dialogue texts of each city using a preset large model to obtain dialogue analysis results;

[0087] A grouping module 24, configured to group each city based on the conversation analysis result and the task conversion rate to obtain a city grouping result;

[0088] The construction module 25 is used to construct the outbound call scripts of the AI ​​robot in different cities based on the city grouping results, so as to use the outbound call scripts to communicate with users.

[0089] The building block 25 is also used to:

[0090] Based on each of the city grouping results, construct an outbound call script of an AI robot corresponding to different cities according to the pre-collected manual scripts;

[0091] Determine the city location corresponding to the user to be called out;

[0092] Based on the city location, a corresponding AI robot is selected to communicate with the user to be called out.

[0093] The building block 25 is also used to:

[0094] Determine the cluster to which the city location belongs;

[0095] Based on the group to which the city location belongs, an AI robot corresponding to the group is selected to communicate with the user to be called.

[0096] The analysis module 23 is also used for:

[0097] Classifying the cities based on the task conversion rates corresponding to the different cities;

[0098] Using the preset large model, the intent recognition is performed on the historical conversation texts of each city corresponding to each classification type to obtain the intent recognition results;

[0099] Use the preset large model to extract the multi-dimensional indicator features corresponding to each historical conversation text.

[0100] The analysis module 23 is also used for:

[0101] For any city:

[0102] Based on each of the historical conversation texts, extracting behavior features and conversation features;

[0103] Performing sentiment analysis on each of the historical dialogue texts to obtain sentiment distribution features;

[0104] The multi-dimensional indicator feature is formed based on the behavior feature, the conversation feature, and the emotion distribution feature.

[0105] The grouping module 24 is also used for:

[0106] Compare and analyze the intent recognition results corresponding to the historical dialogue texts of each city;

[0107] Based on the comparison results, target feature information is extracted from the multi-dimensional indicator features;

[0108] Based on the target feature information and the task conversion rate, each city is grouped to obtain a city grouping result.

[0109] The determination module 22 is further used for:

[0110] For each city;

[0111] Determine the number of users in the city who have engaged in conversations with the AI ​​bot;

[0112] Determine the outbound call result corresponding to each historical conversation text in the city;

[0113] Determine the number of completed tasks based on the outbound call results;

[0114] The task conversion rate is determined based on the number of users and the number of completed tasks.

[0115] For the specific definition of the outbound call management device based on city clustering, please refer to the definition of the outbound call management method based on city clustering above, which will not be repeated here. Each module in the above-mentioned outbound call management device based on city clustering can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0116] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 3 As shown, Figure 3 : is a schematic diagram of a computer device in one embodiment of the present invention. The computer device includes a processor, a memory, a network interface and a database connected by a device bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a readable storage medium and an internal memory. The readable storage medium stores an operating device, a computer-readable instruction and a database. The internal memory provides an environment for the operation of the operating device and the computer-readable instructions in the readable storage medium. The database of the computer device is used to store data involved in the outbound call management method based on city clustering. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer-readable instruction is executed by the processor, an outbound call management method based on city clustering is implemented. The readable storage medium provided in this embodiment includes a non-volatile readable storage medium and a volatile readable storage medium.

[0117] In one embodiment, a computer device is provided. The computer device may be a terminal device, and its internal structure diagram may be as follows: Figure 3 As shown. The computer device includes a processor, a memory, and a network interface connected through a device bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a readable storage medium. The readable storage medium stores computer-readable instructions. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer-readable instructions are executed by the processor, an outbound call management method based on city grouping is implemented. The readable storage medium provided in this embodiment includes a non-volatile readable storage medium and a volatile readable storage medium.

[0118] In one embodiment, a computer device is provided, including a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor. When the processor executes the computer-readable instructions, the steps of the outbound call management method based on city grouping as described above are implemented.

[0119] In one embodiment, a readable storage medium is provided, the readable storage medium stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the steps of the outbound call management method based on city grouping as described above are implemented. A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through computer-readable instructions, and the computer-readable instructions can be stored in a non-volatile readable storage medium or a volatile readable storage medium. When the computer-readable instructions are executed, they may include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0120] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0121] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. An outbound call management method based on city grouping, characterized in that: include: Obtain historical conversation texts between the AI ​​robot and various users in different cities; Determine the task conversion rates corresponding to different cities; Using the preset large model, the historical dialogue texts of each city are analyzed to obtain the dialogue analysis results; Based on the dialogue analysis results and the task conversion rate, each city is grouped to obtain a city grouping result; Based on the city clustering results, outbound call scripts for AI robots in different cities are constructed to use the outbound call scripts to communicate with users.

2. The outbound call management method based on city grouping according to claim 1 is characterized in that: The step of constructing outbound call scripts of AI robots in different cities based on the city grouping results, and using the outbound call scripts to communicate with users, includes: Based on each of the city grouping results, construct an outbound call script of an AI robot corresponding to different cities according to the pre-collected manual scripts; Determine the city location corresponding to the user to be called out; Based on the city location, a corresponding AI robot is selected to communicate with the user to be called out.

3. The outbound call management method based on city grouping according to claim 2 is characterized in that: The selecting a corresponding AI robot to communicate with the outbound user based on the city location includes: Determine the cluster to which the city location belongs; Based on the group to which the city location belongs, an AI robot corresponding to the group is selected to communicate with the user to be called.

4. The outbound call management method based on city grouping according to claim 1 is characterized in that: The conversation analysis results include intention recognition results and multi-dimensional indicator features; The preset large model is used to perform dialogue analysis on the historical dialogue texts of each city to obtain dialogue analysis results, including: Classifying the cities based on the task conversion rates corresponding to the different cities; Using the preset large model, the intent recognition is performed on the historical conversation texts of each city corresponding to each classification type to obtain the intent recognition results; Use the preset large model to extract the multi-dimensional indicator features corresponding to each historical conversation text.

5. The outbound call management method based on city grouping according to claim 4 is characterized in that: The method of using a preset large model to extract multi-dimensional indicator features corresponding to each historical dialogue text includes: For any city: Based on each of the historical conversation texts, extracting behavior features and conversation features; Performing sentiment analysis on each of the historical dialogue texts to obtain sentiment distribution features; The multi-dimensional indicator feature is formed based on the behavior feature, the conversation feature, and the emotion distribution feature.

6. The outbound call management method based on city grouping according to claim 4 is characterized in that: The step of grouping the cities based on the conversation analysis results and the task conversion rate to obtain city grouping results includes: Compare and analyze the intent recognition results corresponding to the historical dialogue texts of each city; Based on the comparison results, target feature information is extracted from the multi-dimensional indicator features; Based on the target feature information and the task conversion rate, each city is grouped to obtain a city grouping result.

7. The outbound call management method based on city grouping according to claim 1 is characterized in that: Determining the task conversion rates corresponding to different cities includes: For each city; Determine the number of users in the city who have engaged in conversations with the AI ​​bot; Determine the outbound call result corresponding to each historical conversation text in the city; Determine the number of completed tasks based on the outbound call results; The task conversion rate is determined based on the number of users and the number of completed tasks.

8. An outbound call management device based on city grouping, characterized in that: include: The acquisition module is used to obtain the historical conversation texts between the AI ​​robot and various users in different cities; A determination module is used to determine the task conversion rates corresponding to different cities; An analysis module is used to use a preset large model to conduct dialogue analysis on the historical dialogue texts of each city to obtain dialogue analysis results; A grouping module, used to group each city based on the dialogue analysis result and the task conversion rate to obtain a city grouping result; A construction module is used to construct outbound call scripts of AI robots in different cities based on the city grouping results, so as to use the outbound call scripts to communicate with users.

9. A computer device comprising a memory, a processor, and computer-readable instructions stored in the memory and executed on the processor, characterized in that: When the processor executes the computer-readable instructions, the outbound call management method based on city grouping as described in any one of claims 1 to 7 is implemented.

10. A readable storage medium having computer readable instructions stored thereon, characterized in that: When the computer-readable instructions are executed by a processor, the outbound call management method based on city grouping as described in any one of claims 1 to 7 is implemented.

Images