Intelligent outbound call method, device, equipment and medium

Abstract

The present application relates to the communication technology field in the financial field, disclose a kind of intelligent outbound method, it can be used for the intelligent outbound of target user when financial product is promoted in business, comprising: obtaining user dataset and multiple business scenarios involved by user dataset, and generate multiple outbound lists corresponding to multiple business scenarios;Multiple outbound lists are filtered, and multiple filtered outbound lists are obtained;According to multiple business scenarios, multiple outbound questionnaires corresponding to multiple business scenarios are generated, and outbound tasks containing multiple outbound questionnaires and multiple filtered outbound lists are created;Multiple filtered outbound lists are called based on multiple outbound tasks.This application also proposes a kind of intelligent outbound device, equipment and medium.The present application can improve outbound efficiency and outbound accuracy.

Description

Technical Field

[0001] This invention relates to the field of communication technology, and in particular to an intelligent outbound calling method, apparatus, device, and storage medium. Background Technology

[0002] Outbound calling refers to a company calling users based on the phone numbers they have provided. For example, in the promotion of financial products, calls are made to target users.

[0003] Existing outbound calling technology involves customer service personnel interacting with users via voice calls to provide services such as user follow-up, surveys, and marketing. However, this technology casts a wide net across all users, wasting significant resources and reducing efficiency. Furthermore, it fails to accurately target specific users, resulting in irrelevant calls and increased call rejection rates. Therefore, current technologies suffer from both low efficiency and low accuracy. Summary of the Invention

[0004] This invention provides an intelligent outbound calling method, apparatus, device, and storage medium, the main purpose of which is to improve outbound calling efficiency and accuracy.

[0005] To achieve the above objectives, the present invention provides an intelligent outbound calling method, comprising:

[0006] Obtain a user dataset and multiple business scenarios involved in the user dataset, and generate multiple outbound call lists corresponding to multiple business scenarios;

[0007] Filter the multiple lists of outbound calls to obtain multiple filtered lists of outbound calls.

[0008] Generate multiple outbound call questionnaires corresponding to the multiple business scenarios, and create an outbound call task containing multiple outbound call questionnaires and multiple filtered outbound call lists;

[0009] Multiple outbound call tasks are used to make outbound calls to multiple filtered lists of people to be called.

[0010] Optionally, generating multiple outbound call lists corresponding to multiple business scenarios includes:

[0011] The user dataset is classified according to multiple business scenarios to obtain user data classification results;

[0012] Based on the user data classification results, the user dataset is batch processed to obtain the outbound call list corresponding to each business scenario.

[0013] Optionally, generating multiple outbound call questionnaires corresponding to multiple business scenarios includes:

[0014] Obtain multiple historical questionnaire dialogue sets from multiple business scenarios, and use a preset intent recognition model to perform intent recognition on the multiple historical questionnaire dialogue sets to obtain multiple questionnaire dialogue intents.

[0015] Based on the multiple questionnaire dialogue intentions, the multiple historical questionnaire dialogue sets are mapped into multiple questionnaire dialogue intention sequences, and multiple outbound questionnaires corresponding to the multiple questionnaire dialogue intention sequences are constructed using a preset questionnaire matching template.

[0016] Optionally, mapping multiple historical questionnaire dialogue sets into multiple questionnaire dialogue intent sequences based on multiple questionnaire dialogue intents includes:

[0017] The multiple questionnaire dialogue intentions are sorted according to the time of the dialogue to obtain the sorted questionnaire dialogue intentions.

[0018] The intentions of the sorted questionnaire dialogues are labeled with intention numbers to obtain multiple sequences of the questionnaire dialogue intentions.

[0019] Optionally, the filtering operation on the multiple lists of candidates to be called to obtain multiple filtered lists of candidates to be called includes:

[0020] A cleaning operation is performed on multiple lists of pending outbound calls to obtain multiple cleaned lists of pending outbound calls.

[0021] Retrieve the names and phone numbers of users who made invalid outbound calls from the historical outbound call list;

[0022] The historical outbound call list is compared with multiple cleaned outbound call lists. When there is a user name and user number in multiple cleaned outbound call lists that match the invalid outbound call list in the historical outbound call list, the invalid outbound call list and user number are removed from the multiple cleaned outbound call lists to obtain multiple filtered outbound call lists.

[0023] Optionally, the creation of an outbound call task comprising multiple outbound call questionnaires and multiple filtered lists of candidates to be called includes:

[0024] Extract multiple basic outbound call information from multiple lists of the filtered outbound callers, and configure multiple outbound call times and multiple outbound call logics for multiple lists of the filtered outbound callers;

[0025] Multiple outbound call tasks are generated based on multiple outbound call questionnaires, multiple outbound call basic information, multiple outbound call times, and multiple outbound call logic.

[0026] Optionally, before making outbound calls to multiple filtered lists of candidates based on multiple outbound call tasks, the method further includes:

[0027] Determine whether the outbound call task falls within a preset activity period;

[0028] When the outbound call task falls within a preset activity period, it is determined whether the outbound call task falls within a preset outbound call time period.

[0029] When the outbound call task falls within a preset outbound call time range, the operation of making outbound calls to multiple filtered lists of outbound callers based on multiple outbound call tasks is performed.

[0030] To address the above problems, the present invention also provides an intelligent outbound calling device, the device comprising:

[0031] The list generation module is used to obtain a user dataset and multiple business scenarios involved in the user dataset, and generate multiple lists of outbound calls corresponding to multiple business scenarios;

[0032] The filtering module is used to perform filtering operations on multiple lists of people to be called out, so as to obtain multiple filtered lists of people to be called out.

[0033] The task creation module is used to generate multiple outbound call questionnaires corresponding to multiple business scenarios, and to create outbound call tasks containing multiple outbound call questionnaires and multiple filtered lists of outbound callees;

[0034] The outbound call module is used to make outbound calls to multiple filtered lists of candidates based on multiple outbound call tasks.

[0035] To address the above problems, the present invention also provides an electronic device, the electronic device comprising:

[0036] Memory, storing at least one computer program; and

[0037] The processor executes the computer program stored in the memory to implement the intelligent outbound calling method described above.

[0038] To address the aforementioned problems, the present invention also provides a computer-readable storage medium storing at least one computer program, which is executed by a processor in an electronic device to implement the intelligent outbound calling method described above.

[0039] In this embodiment of the invention, after acquiring the user dataset and the multiple business scenarios involved in the user dataset, firstly, multiple outbound call lists corresponding to multiple business scenarios are generated. This allows for the determination of the outbound call list for each business scenario, enabling precise targeting of target users and improving outbound call accuracy. Secondly, by filtering the multiple outbound call lists, multiple filtered outbound call lists are obtained. This further filters out invalid users in the outbound call lists, resulting in more accurate target users. This facilitates reducing outbound call costs and improving outbound call efficiency and accuracy during subsequent outbound calls. Finally, by generating multiple outbound call questionnaires corresponding to multiple business scenarios, more user information can be obtained in a short time during subsequent outbound calls, improving outbound call efficiency. Furthermore, outbound call tasks containing multiple outbound call questionnaires and multiple filtered outbound call lists are created. Outbound calls are made to the multiple filtered outbound call lists based on these multiple outbound call tasks, further improving outbound call efficiency and accuracy. Therefore, the intelligent outbound call method, device, equipment, and storage medium proposed in this embodiment of the invention can improve outbound call efficiency and accuracy. Attached Figure Description

[0040] Figure 1 This is a flowchart illustrating an embodiment of the intelligent outbound calling method provided by the present invention;

[0041] Figure 2 This is a detailed flowchart illustrating one step of an intelligent outbound calling method provided in an embodiment of the present invention;

[0042] Figure 3 This is a detailed flowchart illustrating one step of an intelligent outbound calling method provided in an embodiment of the present invention;

[0043] Figure 4 This is a schematic diagram of a module of an intelligent outbound calling device provided in an embodiment of the present invention;

[0044] Figure 5 This is a schematic diagram of the internal structure of an electronic device for implementing an intelligent outbound calling method according to an embodiment of the present invention.

[0045] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0046] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0047] This invention provides an intelligent outbound calling method. The executing entity of the intelligent outbound calling method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application embodiment: a server, a terminal, etc. In other words, the intelligent outbound calling method can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster.

[0048] Reference Figure 1 The flowchart shown is a schematic diagram of an intelligent outbound calling method provided in an embodiment of the present invention. In this embodiment, the intelligent outbound calling method includes the following steps S1-S4:

[0049] S1. Obtain the user dataset and the multiple business scenarios involved in the user dataset, and generate multiple lists of outbound calls corresponding to the multiple business scenarios.

[0050] In this embodiment of the invention, the user dataset is a collection of user data based on actual business scenarios, and the user dataset may include information on multiple users. For example, users who have registered or joined a health and wellness community or users who have registered on an insurance network platform.

[0051] In this embodiment of the invention, the multiple business scenarios involved in the user dataset refer to multiple actual scenarios involved in the user dataset. For example, the user dataset selected by the bank that meets the requirements for credit card application involves the scenario of credit card business promotion; the user dataset that has registered and browsed insurance products on the insurance network platform involves the scenario of insurance product purchase; the user dataset in the health and wellness community that has met the requirements but has not yet joined or has joined involves the scenario of user qualification scenarios in the health and wellness community; the target activity users selected by the health and wellness community involve the scenario of offline activities in the health and wellness community; the scenario for all users who have registered or joined the health and wellness community involves the user life scenarios in the health and wellness community (such as the scenario of birthday wishes for users); and the scenario for all users in the health and wellness community located in hotspot areas involves the scenario of user-related hotspot greetings in the health and wellness community (such as the scenario of user care when an earthquake or rainstorm occurs in the user's local area).

[0052] In this embodiment of the invention, the list of users to be called refers to the list of users to be called for each business scenario, wherein the list of users to be called includes user name, age and telephone information, etc.

[0053] In this embodiment, the list of users to be called is generated based on the data in the user dataset. Therefore, the users in the list of users to be called belong to the users in the user dataset.

[0054] This invention, through obtaining a user dataset and multiple business scenarios involved in the user dataset, generates multiple outbound call lists for these business scenarios. Based on the segmentation of business scenarios, it can achieve precise selection of outbound call lists, improve user acceptance of outbound call content, facilitate the reduction of user complaints, and improve the accuracy and reach of outbound calls.

[0055] As an embodiment of the present invention, reference Figure 2 As shown, generating multiple outbound call lists corresponding to multiple business scenarios includes the following steps S11-S12:

[0056] S11. Classify the user dataset according to multiple business scenarios to obtain user data classification results;

[0057] S12. Perform batch processing on the user dataset according to the user data classification results to obtain the outbound call list corresponding to each business scenario.

[0058] The user data classification results include multiple types of user datasets, and each type of user dataset is a set of user data corresponding to each business scenario.

[0059] Specifically, classifying the user dataset according to multiple business scenarios includes: obtaining scenario identifiers for multiple business scenarios (the scenario identifier can be the name of the business scenario, and each business scenario name is unique), and matching the user dataset with the multiple scenario identifiers. The matching result between the user dataset and the business scenario is the classification result. For example, if there is a scenario A with the scenario identifier "Health and Wellness Community" for user qualification, all users in the user dataset who meet the membership requirements are filtered according to the conditions of this scenario, and all users who meet the membership requirements are marked, then the marked user dataset is matched with scenario A.

[0060] For example, user classification results for scenarios involving meeting the criteria for senior living communities can include users who meet the entry requirements and those who have already signed up; user classification results for scenarios involving offline activities in senior living communities can include users who have participated in historical activities frequently, provided positive feedback, and received high ratings; user classification results for scenarios involving daily life in senior living communities can include registered or current members; user classification results for scenarios involving greetings related to trending topics in senior living communities can include users associated with trending areas; and user classification results for scenarios involving purchasing insurance products can include registered users on insurance online platforms.

[0061] In one embodiment of the present invention, the batch processing refers to the operation of generating outbound call lists in batches on the user dataset. It can generate batch data of the user dataset by calling a pre-configured user interface based on the user data classification results, and set the trigger time for generating the outbound call list, and generate the outbound call list corresponding to the batch data according to the trigger time.

[0062] S2. Perform a filtering operation on the multiple lists of pending outbound calls to obtain multiple filtered lists of pending outbound calls.

[0063] In this embodiment of the invention, the multiple lists of outbound calls to be filtered refer to lists after filtering out invalid users from the outbound call lists.

[0064] This invention provides a method to filter multiple lists of users to be called, thereby obtaining multiple filtered lists. This allows for further screening of invalid users in the lists to obtain more accurate target users, which in turn reduces outbound calling costs and improves outbound calling efficiency and accuracy.

[0065] As an embodiment of the present invention, the filtering operation on the plurality of lists of candidates to be called to obtain a plurality of filtered lists of candidates to be called to include:

[0066] The process involves cleaning multiple lists of pending outbound calls to obtain multiple cleaned pending outbound call lists; obtaining the names and numbers of users who made invalid outbound calls in the historical outbound call list; comparing the historical outbound call list with the multiple cleaned pending outbound call lists; and removing the names and numbers of users who made invalid outbound calls in the historical outbound call list from the multiple cleaned pending outbound call lists to obtain multiple filtered pending outbound call lists.

[0067] The purpose of cleaning multiple outbound call lists is to remove duplicates from multiple outbound call lists, thereby improving the quality of the lists.

[0068] In one embodiment of the present invention, the names and numbers of users who made invalid outbound calls include, but are not limited to, wrong numbers, empty numbers, deactivated numbers, and names and numbers in blacklists. The comparison involves comparing all the names and numbers of users who made invalid outbound calls in the historical outbound call list with all the names and numbers of users in multiple cleaned outbound call lists. If there are user names and numbers in multiple cleaned outbound call lists that match the names and numbers of invalid outbound calls in the historical outbound call list, then the user names and numbers in the cleaned outbound call lists that match the names and numbers of invalid outbound calls in the historical outbound call list are removed, resulting in multiple filtered outbound call lists.

[0069] Furthermore, if multiple cleaned outbound call lists do not contain user names and user numbers that are invalid outbound calls in the historical outbound call list, then the multiple cleaned outbound call lists are multiple filtered outbound call lists.

[0070] S3. Generate multiple outbound call questionnaires corresponding to the multiple business scenarios, and create an outbound call task containing multiple outbound call questionnaires and multiple filtered outbound call lists.

[0071] In this embodiment of the invention, the multiple outbound call questionnaires refer to a series of business questions for users set based on actual business needs. For example, the outbound call task refers to an outbound call task generated for each user in each filtered outbound call list.

[0072] This invention generates multiple outbound call questionnaires based on various business scenarios, allowing for the creation of specialized questionnaires for each scenario. This facilitates the acquisition of more user information in a shorter time during outbound calls, thereby improving outbound call efficiency.

[0073] As an embodiment of the present invention, reference Figure 3 As shown, the step of generating multiple outbound call questionnaires corresponding to multiple business scenarios includes the following steps S31-S32:

[0074] S31. Obtain multiple historical questionnaire dialogue sets for multiple business scenarios, and use a preset intent recognition model to perform intent recognition on the multiple historical questionnaire dialogue sets to obtain multiple questionnaire dialogue intents.

[0075] S32. Based on the multiple questionnaire dialogue intentions, map the multiple historical questionnaire dialogue sets into multiple questionnaire dialogue intention sequences, and use a preset questionnaire matching template to construct multiple outbound questionnaires corresponding to the multiple questionnaire dialogue intention sequences.

[0076] The aforementioned multiple historical questionnaire dialogue sets refer to the dataset of dialogues between users and corresponding customer service representatives in each business scenario over a historical period, concerning questionnaire questions and responses.

[0077] In one embodiment of the present invention, the preset intent recognition model can be a BERT or TextCNN model; the multiple questionnaire dialogue intents refer to the intent corresponding to each text in the historical questionnaire dialogue set. For example, if a dialogue text is "What are the signing standards for joining the health and wellness community?", then the corresponding intent is "the standards for joining the health and wellness community".

[0078] This invention utilizes a pre-defined intent recognition model to identify the intent of multiple historical questionnaire dialogue sets, thereby accurately identifying the questionnaire dialogue intent of the initial historical questionnaire dialogue sets and improving the accuracy of questionnaire construction. The intent recognition involves segmenting multiple historical questionnaire dialogue sets using the intent recognition model to obtain multiple dialogue word vector sets, extracting features from these dialogue word vector sets to obtain multiple dialogue feature vector sets, and finally outputting the multiple questionnaire dialogue intents from the multiple dialogue feature vector sets through the activation function of the model.

[0079] In this embodiment of the invention, the preset questionnaire matching template can be a matching template between the questionnaire dialogue intent and the questionnaire questions and scripts for each business scenario. Through this matching template, the questionnaire questions and communication scripts for each business scenario can be determined, so as to select the appropriate outbound questionnaire for each business scenario.

[0080] Further, the step of mapping multiple historical questionnaire dialogue sets into multiple questionnaire dialogue intent sequences based on multiple questionnaire dialogue intents includes:

[0081] The multiple questionnaire dialogue intentions are sorted according to the time of occurrence to obtain sorted questionnaire dialogue intentions; the sorted questionnaire dialogue intentions are labeled with intention sequence numbers to obtain multiple questionnaire dialogue intention sequences.

[0082] The sorted questionnaire dialogue intents are questionnaire dialogue intents arranged according to the time sequence of the user's dialogue with customer service; the intent sequence number annotation can be achieved by calling the Deepdive annotation tool, the main function of which is to annotate and associate the sorted questionnaire dialogue intents according to the intent sequence number, and the annotation result is the questionnaire dialogue intent sequence.

[0083] For example, sorting the intents of user-customer service conversations in a questionnaire by conversation time can yield intents such as inquiries about joining a senior living community, intents to meet membership requirements, and intents to sign a membership agreement. After annotation, these intents can be mapped to a sequence of questionnaire conversation intents: intents. Figure 1 Intent to participate in membership activities, Figure 2 Intent and purpose of meeting membership requirements Figure 3 The intention is to sign a membership agreement.

[0084] As an embodiment of the present invention, the creation of an outbound calling task comprising multiple outbound calling questionnaires and multiple filtered lists of candidates to be called includes:

[0085] Extract basic outbound call information from multiple lists of candidates to be called, and configure multiple outbound call times and multiple outbound call logics for multiple lists of candidates to be called; generate multiple outbound call tasks based on multiple outbound call questionnaires, multiple basic outbound call information, multiple outbound call times and multiple outbound call logics.

[0086] Among them, the multiple outbound call basic information can be all user names and all user numbers included in the list to be called; the multiple outbound call times refer to the outbound call time periods set based on the actual business scenario, usually from 9:00 to 11:30 am and from 2:00 to 6:00 pm, and the outbound call time periods can be customized and adjusted in special time periods (such as weekends, holidays, etc.) to improve outbound call efficiency.

[0087] In this embodiment of the invention, the multiple outbound call logics are outbound call strategies for each outbound call list, which may include strategies such as call interval duration, number of calls per call, call start time and end time, automatic call duration, outbound call method, and outbound call filtering conditions. Among them, the outbound call method can be a sequential calling method for multiple numbers of the same customer or a calling method based on number sequence, etc., when the same customer has multiple numbers. The outbound call filtering conditions can be custom filtering conditions. For example, if a health and wellness community has an offline activity G that requires users to reach activity level B to participate, then the outbound call users who have reached activity level B can be determined through condition filtering.

[0088] Furthermore, in this embodiment of the invention, the outbound call task also specifies the time limit for making outbound calls after the outbound call list is issued and the maximum number of outbound calls to rejected customers. For example, if an offline event A lasts for three days, the effective time limit for making outbound calls is to make outbound calls to users who have previously rejected calls within three days of the event. The maximum number of outbound calls is set to two to prevent users from getting annoyed by the event and to optimize the user's listening experience.

[0089] S4. Make outbound calls to multiple filtered lists of outbound callers based on multiple outbound call tasks.

[0090] In this embodiment of the invention, the outbound call can be an outbound call initiated by AI customer service to users in the filtered outbound call list.

[0091] In an optional embodiment of the present invention, before making outbound calls to multiple filtered lists of candidates based on multiple outbound call tasks, the method further includes:

[0092] Determine whether the outbound call task falls within a preset activity period; if the outbound call task falls within a preset activity period, identify whether the outbound call task falls within a preset outbound call time period; if the outbound call task falls within a preset outbound call time period, perform the operation of making outbound calls to multiple filtered outbound call lists based on multiple outbound call tasks.

[0093] The preset activity period refers to activities held in actual business operations, such as the points redemption program for members of health and wellness communities; the preset outbound call time range includes outbound call time ranges for special dates such as holidays.

[0094] In one embodiment of the present invention, by determining whether the outbound call task belongs to a preset activity period and identifying whether the outbound call task belongs to a preset outbound call time period, the validity of outbound call results at different times during the activity period can be guaranteed, thereby improving the user participation rate and the user reach rate of the activity.

[0095] In this embodiment of the invention, after obtaining the user dataset and the multiple business scenarios involved in the user dataset, firstly, multiple outbound call lists corresponding to multiple business scenarios are generated. This allows for the determination of the outbound call list for each business scenario, enabling precise targeting of target users and improving outbound call accuracy. Secondly, by filtering the multiple outbound call lists, multiple filtered outbound call lists are obtained. This further filters out invalid users in the outbound call lists, resulting in more accurate target users. This facilitates reducing outbound call costs and improving outbound call efficiency and accuracy during subsequent outbound calls. Finally, by generating multiple outbound call questionnaires corresponding to multiple business scenarios, more user information can be obtained in a short time during subsequent outbound calls, improving outbound call efficiency. Furthermore, outbound call tasks containing multiple outbound call questionnaires and multiple filtered outbound call lists are created. Outbound calls are made to the multiple filtered outbound call lists based on multiple outbound call tasks, further improving outbound call efficiency and accuracy. Therefore, the intelligent outbound call method proposed in this embodiment of the invention can improve outbound call efficiency and accuracy.

[0096] like Figure 4 As shown, the intelligent outbound calling device 100 of the present invention can be installed in an electronic device. Depending on the functions implemented, the intelligent outbound calling device may include a list generation module 101, a filtering module 102, a task creation module 103, and an outbound calling module 104. The module in this invention can also be called a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, and which are stored in the memory of the electronic device.

[0097] In this embodiment, the functions of each module / unit are as follows:

[0098] The list generation module 101 is used to acquire a user dataset and multiple business scenarios involved in the user dataset, and generate multiple lists of outbound calls corresponding to multiple business scenarios.

[0099] In this embodiment of the invention, the user dataset is a collection of user data based on actual business scenarios, and the user dataset may include information on multiple users. For example, users who have registered or joined a health and wellness community or users who have registered on an insurance network platform.

[0100] In this embodiment of the invention, the multiple business scenarios involved in the user dataset refer to multiple actual scenarios involved in the user dataset. For example, the user dataset selected by the bank that meets the requirements for credit card application involves the scenario of credit card business promotion; the user dataset that has registered and browsed insurance products on the insurance network platform involves the scenario of insurance product purchase; the user dataset in the health and wellness community that has met the requirements but has not yet joined or has joined involves the scenario of user qualification scenarios in the health and wellness community; the target activity users selected by the health and wellness community involve the scenario of offline activities in the health and wellness community; the scenario for all users who have registered or joined the health and wellness community involves the user life scenarios in the health and wellness community (such as the scenario of birthday wishes for users); and the scenario for all users in the health and wellness community located in hotspot areas involves the scenario of user-related hotspot greetings in the health and wellness community (such as the scenario of user care when an earthquake or rainstorm occurs in the user's local area).

[0101] In this embodiment of the invention, the list of users to be called refers to the list of users to be called for each business scenario, wherein the list of users to be called includes user name, age and telephone information, etc.

[0102] In this embodiment, the list of users to be called is generated based on the data in the user dataset. Therefore, the users in the list of users to be called belong to the users in the user dataset.

[0103] This invention, through obtaining a user dataset and multiple business scenarios involved in the user dataset, generates multiple outbound call lists for these business scenarios. Based on the segmentation of business scenarios, it can achieve precise selection of outbound call lists, improve user acceptance of outbound call content, facilitate the reduction of user complaints, and improve the accuracy and reach of outbound calls.

[0104] As an embodiment of the present invention, the list generation module 101 generates multiple lists of outbound calls corresponding to multiple business scenarios by performing the following operations:

[0105] The user dataset is classified according to multiple business scenarios to obtain user data classification results;

[0106] Based on the user data classification results, the user dataset is batch processed to obtain the outbound call list corresponding to each business scenario.

[0107] The user data classification results include multiple types of user datasets, and each type of user dataset is a set of user data corresponding to each business scenario.

[0108] Specifically, classifying the user dataset according to multiple business scenarios includes: obtaining scenario identifiers for multiple business scenarios (the scenario identifier can be the name of the business scenario, and each business scenario name is unique), and matching the user dataset with the multiple scenario identifiers. The matching result between the user dataset and the business scenario is the classification result. For example, if there is a scenario A with the scenario identifier "Health and Wellness Community" for user qualification, all users in the user dataset who meet the membership requirements are filtered according to the conditions of this scenario, and all users who meet the membership requirements are marked, then the marked user dataset is matched with scenario A.

[0109] For example, user classification results for scenarios involving meeting the criteria for senior living communities can include users who meet the entry requirements and those who have already signed up; user classification results for scenarios involving offline activities in senior living communities can include users who have participated in historical activities frequently, provided positive feedback, and received high ratings; user classification results for scenarios involving daily life in senior living communities can include registered or current members; user classification results for scenarios involving greetings related to trending topics in senior living communities can include users associated with trending areas; and user classification results for scenarios involving purchasing insurance products can include registered users on insurance online platforms.

[0110] In one embodiment of the present invention, the batch processing refers to the operation of generating outbound call lists in batches on the user dataset. It can generate batch data of the user dataset by calling a pre-configured user interface based on the user data classification results, and set the trigger time for generating the outbound call list, and generate the outbound call list corresponding to the batch data according to the trigger time.

[0111] The filtering module 102 is used to perform filtering operations on multiple lists of outbound calls to obtain multiple filtered lists of outbound calls.

[0112] In this embodiment of the invention, the multiple lists of outbound calls to be filtered refer to lists after filtering out invalid users from the outbound call lists.

[0113] This invention provides a method to filter multiple lists of users to be called, thereby obtaining multiple filtered lists. This allows for further screening of invalid users in the lists to obtain more accurate target users, which in turn reduces outbound calling costs and improves outbound calling efficiency and accuracy.

[0114] As an embodiment of the present invention, the filtering module 102 performs a filtering operation on multiple lists of candidates to be called by performing the following operations to obtain multiple filtered lists of candidates to be called:

[0115] A cleaning operation is performed on the multiple lists of pending outbound calls to obtain multiple cleaned lists of pending outbound calls.

[0116] Retrieve the names and phone numbers of users who made invalid outbound calls from the historical outbound call list;

[0117] The historical outbound call list is compared with multiple cleaned outbound call lists. When there is a user name and user number in multiple cleaned outbound call lists that match the invalid outbound call list in the historical outbound call list, the invalid outbound call list and user number are removed from the multiple cleaned outbound call lists to obtain multiple filtered outbound call lists.

[0118] The purpose of cleaning multiple outbound call lists is to remove duplicates from multiple outbound call lists, thereby improving the quality of the lists.

[0119] In one embodiment of the present invention, the names and numbers of users who made invalid outbound calls include, but are not limited to, wrong numbers, empty numbers, deactivated numbers, and names and numbers in blacklists. The comparison involves comparing all the names and numbers of users who made invalid outbound calls in the historical outbound call list with all the names and numbers of users in multiple cleaned outbound call lists. If there are user names and numbers in multiple cleaned outbound call lists that match the names and numbers of invalid outbound calls in the historical outbound call list, then the user names and numbers in the cleaned outbound call lists that match the names and numbers of invalid outbound calls in the historical outbound call list are removed, resulting in multiple filtered outbound call lists.

[0120] Furthermore, if multiple cleaned outbound call lists do not contain user names and user numbers that are invalid outbound calls in the historical outbound call list, then the multiple cleaned outbound call lists are multiple filtered outbound call lists.

[0121] The task creation module 103 is used to generate multiple outbound call questionnaires corresponding to multiple business scenarios, and to create an outbound call task containing multiple outbound call questionnaires and multiple filtered outbound call lists.

[0122] In this embodiment of the invention, the multiple outbound call questionnaires refer to a series of business questions for users set based on actual business needs. For example, the outbound call task refers to an outbound call task generated for each user in each filtered outbound call list.

[0123] This invention generates multiple outbound call questionnaires based on various business scenarios, allowing for the creation of specialized questionnaires for each scenario. This facilitates the acquisition of more user information in a shorter time during outbound calls, thereby improving outbound call efficiency.

[0124] As an embodiment of the present invention, the task creation module 103 generates multiple outbound call questionnaires corresponding to multiple business scenarios by performing the following operations:

[0125] Obtain multiple historical questionnaire dialogue sets from multiple business scenarios, and use a preset intent recognition model to perform intent recognition on the multiple historical questionnaire dialogue sets to obtain multiple questionnaire dialogue intents.

[0126] Based on the multiple questionnaire dialogue intentions, the multiple historical questionnaire dialogue sets are mapped into multiple questionnaire dialogue intention sequences, and multiple outbound questionnaires corresponding to the multiple questionnaire dialogue intention sequences are constructed using a preset questionnaire matching template.

[0127] The aforementioned multiple historical questionnaire dialogue sets refer to the dataset of dialogues between users and corresponding customer service representatives in each business scenario over a historical period, concerning questionnaire questions and responses.

[0128] In one embodiment of the present invention, the preset intent recognition model can be a BERT or TextCNN model; the multiple questionnaire dialogue intents refer to the intent corresponding to each text in the historical questionnaire dialogue set. For example, if a dialogue text is "What are the signing standards for joining the health and wellness community?", then the corresponding intent is "the standards for joining the health and wellness community".

[0129] This invention utilizes a pre-defined intent recognition model to identify the intent of multiple historical questionnaire dialogue sets, thereby accurately identifying the questionnaire dialogue intent of the initial historical questionnaire dialogue sets and improving the accuracy of questionnaire construction. The intent recognition involves segmenting multiple historical questionnaire dialogue sets using the intent recognition model to obtain multiple dialogue word vector sets, extracting features from these dialogue word vector sets to obtain multiple dialogue feature vector sets, and finally outputting the multiple questionnaire dialogue intents from the multiple dialogue feature vector sets through the activation function of the model.

[0130] In this embodiment of the invention, the preset questionnaire matching template can be a matching template between the questionnaire dialogue intent and the questionnaire questions and scripts for each business scenario. Through this matching template, the questionnaire questions and communication scripts for each business scenario can be determined, so as to select the appropriate outbound questionnaire for each business scenario.

[0131] Further, the step of mapping multiple historical questionnaire dialogue sets into multiple questionnaire dialogue intent sequences based on multiple questionnaire dialogue intents includes:

[0132] The multiple questionnaire dialogue intentions are sorted according to the time of occurrence to obtain sorted questionnaire dialogue intentions; the sorted questionnaire dialogue intentions are labeled with intention sequence numbers to obtain multiple questionnaire dialogue intention sequences.

[0133] The sorted questionnaire dialogue intents are questionnaire dialogue intents arranged according to the time sequence of the user's dialogue with customer service; the intent sequence number annotation can be achieved by calling the Deepdive annotation tool, the main function of which is to annotate and associate the sorted questionnaire dialogue intents according to the intent sequence number, and the annotation result is the questionnaire dialogue intent sequence.

[0134] For example, sorting the intents of user-customer service conversations in a questionnaire by conversation time can yield intents such as inquiries about joining a senior living community, intents to meet membership requirements, and intents to sign a membership agreement. After annotation, these intents can be mapped to a sequence of questionnaire conversation intents: intents. Figure 1 Intent to participate in membership activities, Figure 2 Intent and purpose of meeting membership requirements Figure 3 The intention is to sign a membership agreement.

[0135] As an embodiment of the present invention, the creation of an outbound calling task comprising multiple outbound calling questionnaires and multiple filtered lists of candidates to be called includes:

[0136] Extract basic outbound call information from multiple lists of candidates to be called, and configure multiple outbound call times and multiple outbound call logics for multiple lists of candidates to be called; generate multiple outbound call tasks based on multiple outbound call questionnaires, multiple basic outbound call information, multiple outbound call times and multiple outbound call logics.

[0137] Among them, the multiple outbound call basic information can be all user names and all user numbers included in the list to be called; the multiple outbound call times refer to the outbound call time periods set based on the actual business scenario, usually from 9:00 to 11:30 am and from 2:00 to 6:00 pm, and the outbound call time periods can be customized and adjusted in special time periods (such as weekends, holidays, etc.) to improve outbound call efficiency.

[0138] In this embodiment of the invention, the multiple outbound call logics are outbound call strategies for each outbound call list, which may include strategies such as call interval duration, number of calls per call, call start time and end time, automatic call duration, outbound call method, and outbound call filtering conditions. Among them, the outbound call method can be a sequential calling method for multiple numbers of the same customer or a calling method based on number sequence, etc., when the same customer has multiple numbers. The outbound call filtering conditions can be custom filtering conditions. For example, if a health and wellness community has an offline activity G that requires users to reach activity level B to participate, then the outbound call users who have reached activity level B can be determined through condition filtering.

[0139] Furthermore, in this embodiment of the invention, the outbound call task also specifies the time limit for making outbound calls after the outbound call list is issued and the maximum number of outbound calls to rejected customers. For example, if an offline event A lasts for three days, the effective time limit for making outbound calls is to make outbound calls to users who have previously rejected calls within three days of the event. The maximum number of outbound calls is set to two to prevent users from getting annoyed by the event and to optimize the user's listening experience.

[0140] The outbound call module 104 is used to make outbound calls to multiple filtered lists of outbound callees based on multiple outbound call tasks.

[0141] In this embodiment of the invention, the outbound call can be an outbound call initiated by AI customer service to users in the filtered outbound call list.

[0142] The outbound call module 104 can also be used for:

[0143] Before making outbound calls to multiple filtered lists of candidates based on multiple outbound call tasks, it is determined whether the outbound call task belongs to a preset activity period; if the outbound call task belongs to the preset activity period, it is identified whether the outbound call task belongs to a preset outbound call time period; if the outbound call task belongs to the preset outbound call time period, the operation of making outbound calls to multiple filtered lists of candidates based on multiple outbound call tasks is executed.

[0144] The preset activity period refers to activities held in actual business operations, such as the points redemption program for members of health and wellness communities; the preset outbound call time range includes outbound call time ranges for special dates such as holidays.

[0145] In one embodiment of the present invention, by determining whether the outbound call task belongs to a preset activity period and identifying whether the outbound call task belongs to a preset outbound call time period, the validity of outbound call results at different times during the activity period can be guaranteed, thereby improving the user participation rate and the user reach rate of the activity.

[0146] In this embodiment of the invention, after acquiring the user dataset and the multiple business scenarios involved in the user dataset, firstly, multiple outbound call lists corresponding to multiple business scenarios are generated. This allows for the determination of the outbound call list for each business scenario, enabling precise targeting of target users and improving outbound call accuracy. Secondly, by filtering the multiple outbound call lists, multiple filtered outbound call lists are obtained. This further filters out invalid users in the outbound call lists, resulting in more accurate target users. This facilitates reducing outbound call costs and improving outbound call efficiency and accuracy during subsequent outbound calls. Finally, by generating multiple outbound call questionnaires corresponding to multiple business scenarios, more user information can be obtained in a short time during subsequent outbound calls, improving outbound call efficiency. Furthermore, outbound call tasks containing multiple outbound call questionnaires and multiple filtered outbound call lists are created. Outbound calls are made to the multiple filtered outbound call lists based on these multiple outbound call tasks, further improving outbound call efficiency and accuracy. Therefore, the intelligent outbound call device proposed in this embodiment of the invention can improve outbound call efficiency and accuracy.

[0147] like Figure 5 The diagram shown is a structural schematic of the electronic device that implements the intelligent outbound calling method of the present invention.

[0148] The electronic device may include a processor 10, a memory 11, a communication bus 12 and a communication interface 13, and may also include a computer program, such as an intelligent outbound calling program, stored in the memory 11 and capable of running on the processor 10.

[0149] The memory 11 includes at least one type of medium, such as flash memory, portable hard drive, multimedia card, card-type memory (e.g., SD or DX memory), magnetic memory, local disk, optical disk, etc. In some embodiments, the memory 11 can be an internal storage unit of an electronic device, such as a portable hard drive. In other embodiments, the memory 11 can be an external storage device of the electronic device, such as a plug-in portable hard drive, Smart Media Card (SMC), Secure Digital (SD) card, Flash Card, etc. Furthermore, the memory 11 can include both internal and external storage units of the electronic device. The memory 11 can be used not only to store application software and various types of data installed on the electronic device, such as the code of a smart outbound calling program, but also to temporarily store data that has been output or will be output.

[0150] In some embodiments, the processor 10 may be composed of integrated circuits, such as a single packaged integrated circuit or multiple integrated circuits with the same or different functions, including combinations of one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. The processor 10 is the control unit of the electronic device, connecting various components of the entire electronic device through various interfaces and lines. It executes programs or modules (such as intelligent outbound calling programs) stored in the memory 11, and calls data stored in the memory 11 to perform various functions of the electronic device and process data.

[0151] The communication bus 12 can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This bus can be divided into an address bus, a data bus, a control bus, etc. The communication bus 12 is configured to enable communication between the memory 11 and at least one processor 10, etc. For ease of illustration, only one thick line is used in the figure, but this does not indicate that there is only one bus or one type of bus.

[0152] Figure 5 Only electronic devices with components are shown; it will be understood by those skilled in the art that... Figure 5 The structure shown does not constitute a limitation on the electronic device and may include fewer or more components than shown, or combine certain components, or have different component arrangements.

[0153] For example, although not shown, the electronic device may also include a power supply (such as a battery) to power the various components. Preferably, the power supply can be logically connected to the at least one processor 10 through a power management device, thereby enabling functions such as charging management, discharging management, and power consumption management. The power supply may also include one or more DC or AC power supplies, recharging devices, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components. The electronic device may also include various sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be described in detail here.

[0154] Optionally, the communication interface 13 may include a wired interface and / or a wireless interface (such as a Wi-Fi interface, a Bluetooth interface, etc.), which is typically used to establish communication connections between the electronic device and other electronic devices.

[0155] Optionally, the communication interface 13 may further include a user interface, which may be a display, an input unit (such as a keyboard), or, optionally, a standard wired or wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen, etc. The display may also be appropriately referred to as a screen or display unit, used to display information processed in the electronic device and to display a visual user interface.

[0156] It should be understood that the embodiments described are for illustrative purposes only and are not limited to this structure in the scope of the patent application.

[0157] The intelligent outbound calling program stored in the memory 11 of the electronic device is a combination of multiple computer programs. When run in the processor 10, it can achieve the following:

[0158] Obtain a user dataset and multiple business scenarios involved in the user dataset, and generate multiple outbound call lists corresponding to multiple business scenarios;

[0159] Filter the multiple lists of outbound calls to obtain multiple filtered lists of outbound calls.

[0160] Generate multiple outbound call questionnaires corresponding to the multiple business scenarios, and create an outbound call task containing multiple outbound call questionnaires and multiple filtered outbound call lists;

[0161] Multiple outbound call tasks are used to make outbound calls to multiple filtered lists of people to be called.

[0162] Specifically, the processor 10's implementation method of the above-mentioned computer program can be found in [reference needed]. Figure 1 The descriptions of the relevant steps in the corresponding embodiments are not repeated here.

[0163] Furthermore, if the modules / units integrated into the electronic device are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable medium. The computer-readable medium can be non-volatile or volatile. The computer-readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, or a read-only memory (ROM).

[0164] This invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor of an electronic device, can perform the following:

[0165] Obtain a user dataset and multiple business scenarios involved in the user dataset, and generate multiple outbound call lists corresponding to multiple business scenarios;

[0166] Filter the multiple lists of outbound calls to obtain multiple filtered lists of outbound calls.

[0167] Generate multiple outbound call questionnaires corresponding to the multiple business scenarios, and create an outbound call task containing multiple outbound call questionnaires and multiple filtered outbound call lists;

[0168] Multiple outbound call tasks are used to make outbound calls to multiple filtered lists of people to be called.

[0169] Furthermore, the computer-readable storage medium may primarily include a program storage area and a data storage area, wherein the program storage area may store the operating system, at least one application required for a function, etc.; and the data storage area may store data created based on the use of blockchain nodes, etc.

[0170] In the several embodiments provided by this invention, it should be understood that the disclosed media, devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0171] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0172] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0173] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0174] Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be embraced within the invention. No appended diagram markings in the claims should be construed as limiting the scope of the claims.

[0175] The blockchain referred to in this invention is a novel application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanisms, and encryption algorithms. Essentially, a blockchain is a decentralized database, a chain of data blocks linked together using cryptographic methods. Each data block contains information about a batch of network transactions, used to verify the validity of the information (anti-counterfeiting) and generate the next block. A blockchain can include an underlying blockchain platform, a platform product service layer, and an application service layer.

[0176] Furthermore, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices recited in a system claim may also be implemented by a single unit or device through software or hardware. The term "second class" is used to indicate names and does not indicate any specific order.

[0177] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. An intelligent outbound calling method, characterized in that, The method includes: Obtain a user dataset and multiple business scenarios involved in the user dataset, and generate multiple outbound call lists corresponding to multiple business scenarios; Filter the multiple lists of outbound calls to obtain multiple filtered lists of outbound calls. Multiple historical questionnaire dialogue sets for multiple business scenarios are obtained, and the intent of the multiple historical questionnaire dialogue sets is identified using a preset intent recognition model to obtain multiple questionnaire dialogue intents. Based on the multiple questionnaire dialogue intents, the multiple historical questionnaire dialogue sets are mapped into multiple questionnaire dialogue intent sequences. Multiple outbound questionnaires corresponding to the multiple questionnaire dialogue intent sequences are constructed using a preset questionnaire matching template. An outbound task containing multiple outbound questionnaires and multiple filtered outbound call lists is created. In the multiple outbound call tasks, the outbound call time limit after the multiple outbound call lists are issued and the maximum number of outbound calls to rejected customers are set. Based on the configured multiple outbound call tasks, multiple filtered lists of candidates to be called are used to make outbound calls.

2. The intelligent outbound calling method as described in claim 1, characterized in that, The generation of multiple outbound call lists corresponding to multiple business scenarios includes: The user dataset is classified according to multiple business scenarios to obtain user data classification results; Based on the user data classification results, the user dataset is batch processed to obtain the outbound call list corresponding to each business scenario.

3. The intelligent outbound calling method as described in claim 1, characterized in that, The step of mapping multiple historical questionnaire dialogue sets into multiple questionnaire dialogue intent sequences based on multiple questionnaire dialogue intents includes: The multiple questionnaire dialogue intentions are sorted according to the time of the dialogue to obtain the sorted questionnaire dialogue intentions. The intentions of the sorted questionnaire dialogues are labeled with intention numbers to obtain multiple sequences of the questionnaire dialogue intentions.

4. The intelligent outbound calling method as described in any one of claims 1-3, characterized in that, The filtering operation on the multiple lists of candidates to be called, resulting in multiple filtered lists of candidates to be called, includes: A cleaning operation is performed on the multiple lists of pending outbound calls to obtain multiple cleaned lists of pending outbound calls. Retrieve the names and phone numbers of users who made invalid outbound calls from the historical outbound call list; The historical outbound call list is compared with multiple cleaned outbound call lists. When there is a user name and user number in multiple cleaned outbound call lists that match the invalid outbound call list in the historical outbound call list, the invalid outbound call list and user number are removed from the multiple cleaned outbound call lists to obtain multiple filtered outbound call lists.

5. The intelligent outbound calling method as described in any one of claims 1-3, characterized in that, The creation of an outbound calling task that includes multiple outbound call questionnaires and multiple filtered lists of candidates to be called includes: Extract multiple basic outbound call information from multiple lists of the filtered outbound callers, and configure multiple outbound call times and multiple outbound call logics for multiple lists of the filtered outbound callers; Multiple outbound call tasks are generated based on multiple outbound call questionnaires, multiple outbound call basic information, multiple outbound call times, and multiple outbound call logic.

6. The intelligent outbound calling method as described in any one of claims 1-3, characterized in that, Before making outbound calls to multiple filtered lists of candidates based on multiple outbound call tasks, the method further includes: Determine whether the outbound call task falls within a preset activity period; When the outbound call task falls within a preset activity period, it is determined whether the outbound call task falls within a preset outbound call time period. When the outbound call task falls within a preset outbound call time range, the operation of making outbound calls to multiple filtered lists of outbound callers based on multiple outbound call tasks is performed.

7. An intelligent outbound calling device, characterized in that, The device includes: The list generation module is used to obtain a user dataset and multiple business scenarios involved in the user dataset, and generate multiple lists of outbound calls corresponding to multiple business scenarios; The filtering module is used to perform filtering operations on multiple lists of people to be called out, so as to obtain multiple filtered lists of people to be called out. The task creation module is used to acquire multiple historical questionnaire dialogue sets of multiple business scenarios, and use a preset intent recognition model to perform intent recognition on the multiple historical questionnaire dialogue sets to obtain multiple questionnaire dialogue intents. Based on the multiple questionnaire dialogue intents, the multiple historical questionnaire dialogue sets are mapped into multiple questionnaire dialogue intent sequences. A preset questionnaire matching template is used to construct multiple outbound questionnaires corresponding to the multiple questionnaire dialogue intent sequences. An outbound task containing multiple outbound questionnaires and multiple filtered outbound call lists is created. In the multiple outbound call tasks, the outbound call dialing time limit after the multiple outbound call lists are issued and the maximum number of outbound calls to rejected customers are set. The outbound call module is used to make outbound calls to multiple filtered lists of outbound callees based on multiple outbound call tasks that have been set.

8. An electronic device, characterized in that, The electronic device includes: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the intelligent outbound calling method as described in any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the intelligent outbound calling method as described in any one of claims 1 to 6.

Images