Index change attribution method, device and equipment of outbound robot and medium

Abstract

The invention discloses an index change attribution method for a call-out robot, and the method comprises the steps: carrying out the one-time call-out operation of a user with a non-member identifier through the call-out robot based on a preset verbal skill template, and the preset verbal skill template comprises a call-out template statement; when the occurrence of the index change event is detected, obtaining a changed influence factor associated with the index change event; dividing the users into high-willingness users and low-willingness users according to the influence factors; performing attribution analysis according to the outbound template statement corresponding to the user reply content of the high-intention user and the outbound template statement of the user reply content of the low-intention user to obtain an attribution result; according to the method, the index change is deeply analyzed, so that the key factors influencing the business index change are determined, and through the automatic attribution analysis, an enterprise can quickly locate a problem source, the dependence on manual analysis is reduced, and the problem solving efficiency and accuracy are remarkably improved.

Description

Technical Field

[0001] The present application relates to the field of attribution technology, and specifically to a method, device, equipment and medium for attributing indicator changes of an outbound call robot. Background Art

[0002] Attribution methods for analyzing abnormal business indicators usually include: dimension attribution, link attribution, event attribution and other methods. The attribution of abnormal indicators is generally used to help companies better understand the reasons behind various events and decisions, thereby optimizing business processes, improving efficiency and achieving business goals. Attribution is applied to many fields: marketing and advertising, e-commerce, finance and insurance, etc. Therefore, how to reduce labor costs and improve efficiency in the attribution process of abnormal indicators has become an urgent problem to be solved. Summary of the invention

[0003] In view of the above-mentioned shortcomings of the prior art, the present application provides a method, device, equipment and medium for attributing indicator changes of an outbound call robot, which are used to solve at least one defect in the prior art.

[0004] To achieve the above objectives and other objectives, the present application provides an indicator change attribution method for an outbound call robot, the attribution method comprising:

[0005] Performing a one-time outbound call operation on a user with a non-member ID through an outbound call robot based on a preset speech template, wherein the preset speech template includes an outbound call template sentence;

[0006] Detecting an indicator change event and obtaining an impact factor associated with the indicator change event when the indicator change event is detected;

[0007] Classifying the users into high-willingness users and low-willingness users according to the influencing factors;

[0008] Attribution analysis is performed based on the outbound call template sentences corresponding to the user reply content of the high-willingness user and the outbound call template sentences corresponding to the user reply content of the low-willingness user to obtain an attribution result.

[0009] In one embodiment of the present application, the indicator change event includes at least one of the following: an increase or decrease in the number of users with member identification, an increase or decrease in the average call duration, an increase or decrease in the average call rounds, an increase or decrease in the first-round hang-up rate, an increase or decrease in the outbound call conversion rate, and an increase or decrease in the connection conversion rate; the influencing factors include at least one of the following: reply rate, reply willingness rate, reply conversion rate, rounds after reply, setting behavior, purchasing behavior, click behavior, speech hang-up rate, dialogue node stay rounds, dialogue node stay duration, dialogue node reach rate, dialogue node hang-up rate, session willingness rate, and conversion rate in each time period.

[0010] In one embodiment of the present application, if the influencing factors include a response rate and a response willingness rate, obtaining the changed influencing factors associated with the indicator change event includes:

[0011] Acquire the conversation content of at least one round of conversation between two parties, where the two parties include an outbound calling robot and a user with a non-member identifier;

[0012] Identify the conversation content and obtain conversation indicators; the conversation indicators include a first conversation indicator corresponding to the user, a second conversation indicator corresponding to the outbound call robot, and a third conversation indicator, the first conversation indicator is used to indicate the number of responses to the set words, the second conversation indicator is used to indicate the number of triggered set words, and the third conversation indicator includes the number of responses to the set words with high willingness;

[0013] The response rate is obtained based on the first dialogue indicator and the second dialogue indicator, and the response willingness rate is obtained based on the second dialogue indicator and the third dialogue indicator.

[0014] In one embodiment of the present application, the response rate=the first indicator divided by the second indicator; the response willingness rate=the third indicator divided by the second indicator.

[0015] In one embodiment of the present application, the speech template also includes a reply template statement corresponding to the outbound call template statement, and the method for determining the high-willingness setting speech response includes:

[0016] Acquire a reply template sentence corresponding to each round of dialogue and the outbound call template sentence and a user reply content output by the user during each round of dialogue, wherein the reply template sentence includes a high willingness keyword for indicating high willingness;

[0017] Comparing the reply template sentence with the user reply content to identify high-intention keywords in the user reply content;

[0018] Based on the high-willingness keywords in the user reply content, it is determined whether the user reply content output by the user is a high-willingness set speech response.

[0019] In one embodiment of the present application, the method for determining the high-willingness setting speech response includes:

[0020] Obtain the content of each round of dialogue and the user's response content output by the user;

[0021] Extracting features of the user's reply content to obtain voice features and text features;

[0022] Performing emotion recognition on the speech features based on a pre-trained emotion recognition model to obtain the user's emotion classification;

[0023] Performing intent recognition on the text features based on a pre-trained intent recognition model to obtain a user intent classification;

[0024] Recognize the conversation content based on a pre-trained semantic continuity recognition model to obtain a conversation continuity classification;

[0025] Determine whether the user response content output by the user is a high-willing set speech response based on the user's emotion classification, the user's intention classification, and the conversation continuity classification.

[0026] In one embodiment of the present application, the method further includes:

[0027] The outbound call template statements corresponding to the low-willingness users are rewritten according to the outbound call template statements corresponding to the user response content of the high-willingness users.

[0028] To achieve the above-mentioned purpose and other related purposes, the present application provides an indicator change attribution device for an outbound call robot, the attribution device comprising:

[0029] An outbound calling module, used for performing a one-time outbound calling operation on a user with a non-member ID through an outbound calling robot based on a preset speech template, wherein the preset speech template includes an outbound calling template sentence;

[0030] A detection module, used to detect an indicator change event and, when the indicator change event is detected, obtain an impact factor associated with the indicator change event that has changed;

[0031] A user classification module, used for classifying the users into high-willingness users and low-willingness users according to the influencing factors;

[0032] The attribution summary module is used to perform attribution analysis based on the outbound call template sentences corresponding to the user reply content of the high-willingness user and the outbound call template sentences corresponding to the user reply content of the low-willingness user to obtain the attribution result.

[0033] To achieve the above-mentioned purpose and other related purposes, the present application provides an indicator change attribution device for an outbound call robot, including:

[0034] one or more processors; and

[0035] A memory is used to store one or more programs. When the one or more programs are executed by the one or more processors, the memory implements the indicator change attribution method of the outbound call robot.

[0036] To achieve the above-mentioned purpose and other related purposes, the present application provides one or more machine-readable media on which instructions are stored, which, when executed by one or more processors, enable the processor to execute the method for attributing the indicator changes of the outbound call robot.

[0037] Beneficial effects of this application:

[0038] The present application discloses an indicator change attribution method for an outbound call robot, comprising: performing a one-time outbound call operation on a user with a non-member identification through an outbound call robot based on a preset speech template, wherein the preset speech template includes an outbound call template statement; when an indicator change event is detected, obtaining an influence factor associated with the indicator change event that has changed; dividing the user into a high-willingness user and a low-willingness user according to the influence factor; performing attribution analysis based on the outbound call template statement corresponding to the user reply content of the high-willingness user and the outbound call template statement corresponding to the user reply content of the low-willingness user, so as to obtain an attribution result; the present application conducts an in-depth analysis of the indicator changes to determine the key factors affecting the changes in business indicators. Through this automated attribution analysis, the source of the problem can be quickly located, reducing dependence on manual analysis, and significantly improving the efficiency and accuracy of problem solving.

[0039] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The drawings herein are incorporated into the specification and constitute a part of the specification, showing embodiments consistent with the present application, and together with the specification, are used to explain the principles of the present application. Obviously, the drawings described below are only some embodiments of the present application, and for those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0041] Figure 1 This is a flow chart of a method for attributing indicator changes of an outbound call robot according to an embodiment of the present application;

[0042] Figure 2 A flow chart of obtaining a change-related influencing factor associated with an indicator change event according to an embodiment of the present application;

[0043] Figure 3 A flow chart of a method for determining a high-willing setting script to be signed according to an embodiment of the present application;

[0044] Figure 4 A flow chart of a method for determining a high-willing setting speech response according to an embodiment of the present application;

[0045] Figure 5A block diagram of an indicator change attribution device for an outbound call robot shown in an embodiment of the present application;

[0046] Figure 6 A schematic diagram of the structure of a computer system suitable for implementing the memory of an embodiment of the present application is shown. DETAILED DESCRIPTION

[0047] The following describes the embodiments of the present application through specific examples, and those skilled in the art can easily understand other advantages and effects of the present application from the contents disclosed in this specification. The present application can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present application. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.

[0048] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present application, and thus the drawings only show components related to the present application rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed at will, and the component layout may also be more complicated.

[0049] Although the terms "first", "second", "A", and "B", etc. may be used herein to describe various elements, these elements should not be limited by these terms and are only used to distinguish one element from another. For example, a first element may be referred to as a second element, and similarly, a second element may be referred to as a first element without departing from the scope of the following technology. The term "and / or" includes a combination of multiple related items or any of the multiple related items.

[0050] As used herein, unless the context indicates otherwise, the singular form is intended to include the plural form, and it will be understood that the term "comprising" means the presence of stated features, quantities, steps, operations, elements, or combinations thereof, but does not exclude the presence or addition of one or more other features, quantities, steps, operations, elements, components, or combinations thereof.

[0051] Before the detailed description, it is intended to clarify that the division of components in this specification is divided only by the main function of each component. That is, two or more components to be described below can be combined into one component, or can be divided into two or more components according to more detailed functions. In addition to the main function of the component, each component to be described below can also perform some or all of the functions of other components, and some of the main functions of each component can be exclusively performed by other components.

[0052] First, various nouns that may be involved in this application are described.

[0053] Outbound call robot: An outbound call robot is an automated communication tool that can automatically dial a preset phone number through the telephone system and play a pre-recorded voice message to the caller or conduct a simple human-computer conversation according to a pre-set script. This type of robot is often used in telemarketing, market research, customer service and other fields;

[0054] Average call duration: the average call duration of all outbound calls;

[0055] Average call rounds: the average of all call rounds corresponding to an outbound call;

[0056] First-round hang-up rate: the rate at which calls are hung up after the first contact between the two parties;

[0057] Outbound call conversion rate: the percentage of potential customers that are successfully converted into actual purchases or transactions;

[0058] Connection conversion rate: the proportion of calls that are answered and eventually achieve transaction or conversion goals;

[0059] Response rate: the number of responses to a set script divided by the number of scripts triggered;

[0060] Willingness to respond rate: the number of responses to a given script with high willingness divided by the number of responses to a given script;

[0061] Response conversion rate: the proportion of customers who receive a response and eventually complete a purchase or other predetermined goal;

[0062] Post-response turn: the conversation turn from the beginning of a conversation to the end of the conversation;

[0063] Hang-up rate: the ratio of calls that were actively hung up by the other party before the scheduled content was completed to the total number of calls;

[0064] Dialogue node dwell rounds: the number of times a specific dialogue node (i.e. a topic or question) can be repeatedly discussed or responded to in a dialogue system;

[0065] Dialogue node dwell time: the time a participant spends on a specific dialogue node (i.e., topic or question point);

[0066] Dialogue node reach rate: the percentage of correct responses to questions or instructions in the total number of exchanges;

[0067] Dialog node hang-up rate: the percentage of users ending the conversation at a specific dialog node;

[0068] Conversation willingness rate: the tendency or probability that a person is willing to communicate or have a conversation with others in a specific situation;

[0069] Conversion rate in different time periods: the efficiency or proportion of a certain behavior or activity in achieving its goals in different time periods.

[0070] See also Figure 1 , Figure 1 This is a flow chart of a method for attributing index changes of an outbound call robot according to an embodiment of the present application. Figure 1 As shown, the indicator change attribution method of the outbound call robot at least includes steps S110 to S140:

[0071] Step S110, performing a one-time outbound call operation on a user with a non-member ID through an outbound call robot based on a preset speech template, wherein the preset speech template includes an outbound call template sentence;

[0072] When conducting outbound business through an outbound call robot, different speech templates will be designed according to different types of business. Take the telephone notification of the outbound voice robot of xxx company to notify customers of renewal as an example. The basic process is: Node 1: Hello, I am Xiao Wang, the customer service specialist of XX company. Are you Li Ming? When the user answers yes, jump to node 2: Your xxx product "XXX" is about to renew. Please renew in time. I would like to inform you. When the question at node 2 ends, or the answer to node 1 is no, jump to node 3: Thank you for listening, I wish you a happy life, and goodbye! Therefore, the speech template of the telephone notification business of the outbound call robot of xxx company to notify customers of renewal has three nodes. Node 1 confirms whether it is the customer himself. If it is the customer himself, jump to node 2 to notify. If it is not the customer himself, jump to node 3 to broadcast the sorry ending.

[0073] A user with a non-member ID is a non-member user, and the corresponding user is a user with a member ID, i.e. a member user. A one-time outbound call means that the outbound call robot only makes one phone call to a certain customer. This phone call can include at least one round of dialogue. A round of dialogue means that both parties in the dialogue have made a voice output once, such as the outbound call robot asks a question and the user answers the question.

[0074] Step S120, detecting an indicator change event and obtaining a changed influencing factor associated with the indicator change event when the indicator change event is detected to have occurred;

[0075] An indicator change event indicates that an indicator has changed, including an increase or decrease in an indicator. Detection of an indicator change event indicates that a decrease or increase in the size of one or more indicators has been detected. An impact factor associated with an indicator change event indicates that the impact factor is related to the indicator change event, and a change in the impact factor will cause the indicator to change. Detecting the impact factor associated with an indicator change event indicates detecting whether the impact factor has changed. If the impact factor has changed, the impact factor that has changed is obtained, and the impact factor that has not changed is not considered for the time being. The change mentioned here refers to an increase or decrease in the value of the impact factor.

[0076] Step S130, dividing users into high-willingness users and low-willingness users according to the influencing factors;

[0077] The impact factor changes, including the increase or decrease of the impact factor. The size of the impact factor can be used to distinguish high-willing users from low-willing users. High-willing users refer to users with a higher probability of completing a certain behavior (purchasing a membership), while low-willing users refer to users with a lower probability of completing a certain behavior (purchasing a membership). If the impact factor includes the response rate, then the size of the response rate can be used to distinguish high-willing users from low-willing users. Users with a response rate greater than a certain value are defined as high-willing users, and users with a response rate lower than a certain value are defined as low-willing users.

[0078] Step S140, performing attribution analysis based on the outbound call template sentences corresponding to the user reply contents of the high-willingness users and the outbound call template sentences corresponding to the user reply contents of the low-willingness users, to obtain an attribution result.

[0079] After determining the high-willingness users and low-willingness users, obtain the outbound call template sentence corresponding to the user reply content of the high-willingness users and use it as the first outbound call template sentence, and obtain the outbound call template sentence corresponding to the user reply content of the low-willingness users and use it as the second outbound call template sentence. Then, perform attribution analysis based on the first outbound call template sentence and the second outbound call template sentence to find the cause of the indicator change event; for example, you can find some keywords related to high willingness and keywords related to low willingness, so as to get the attribution result, that is, the cause of the indicator change event is the use of some keywords that make users have high willingness, which increases the number of member users, or the use of some keywords that make users have low willingness, which reduces the number of member users.

[0080] It should be noted that in this application, the changes in indicators and influencing factors are determined based on two outbound calls. The corresponding indicators and influencing factors will be obtained after the previous outbound call test is completed. The speech template used in the next outbound call test is obtained by making corresponding adjustments based on the speech template used in the previous outbound call test. The corresponding indicators and influencing factors will be obtained after completing the next outbound call test. By comparing the two results, it is determined whether there is a change.

[0081] This application obtains the influencing factors associated with the indicator change events that have changed, and then divides the users into high-willingness users and low-willingness users according to the influencing factors, and analyzes the outbound call template statements corresponding to high-willingness users and low-willingness users, so as to determine the reasons for the indicator changes. In addition, the reply template statements can be adjusted according to the attribution results to optimize the outbound call effect and improve user satisfaction. This application conducts an in-depth analysis of indicator changes to determine the key factors that affect changes in business indicators. Through this automated attribution analysis, enterprises can quickly locate the source of the problem, reduce reliance on manual analysis, and significantly improve the efficiency and accuracy of problem solving.

[0082] In one embodiment, the indicator change event includes at least one of the following: an increase or decrease in the number of users with member identification, an increase or decrease in the average call duration, an increase or decrease in the average call rounds, an increase or decrease in the first round hang-up rate, an increase or decrease in the outbound call conversion rate, and an increase or decrease in the connection conversion rate; the influencing factors include at least one of the following: reply rate, reply willingness rate, reply conversion rate, rounds after reply, setting behavior, purchase behavior, click behavior, speech hang-up rate, dialogue node stay rounds, dialogue node stay duration, dialogue node reach rate, dialogue node hang-up rate, session willingness rate, and conversion rate in each time period.

[0083] See also Figure 2 , Figure 2 This is a flow chart of obtaining the influence factors associated with the indicator change event in one embodiment of the present application. Figure 2 As shown, the factors affecting the indicator change event that have been changed include:

[0084] Step S210, obtaining the conversation content of at least one round of conversation between two parties, where the two parties include the outbound calling robot and the user with a non-member identifier;

[0085] Here, at least one round of dialogue refers to at least one interaction between two parties in a certain period of time. The content of the dialogue refers to the speeches of the two parties in the process of interaction within the period of time.

[0086] Step S220, identifying the conversation content and obtaining conversation indicators; the conversation indicators include a first conversation indicator corresponding to the user, a second conversation indicator corresponding to the outbound call robot, and a third conversation indicator, wherein the first conversation indicator is used to indicate the number of responses to the set words, the second conversation indicator is used to indicate the number of triggered set words, and the third conversation indicator includes the number of responses to the set words with high willingness;

[0087] The trigger-set script here refers to a series of pre-set dialogues or answers. When specific conditions or keywords are activated, the outbound call robot will automatically respond or execute these pre-set scripts.

[0088] The set script response here refers to the user's corresponding response to some scripts set by the outbound calling machine.

[0089] Step S230, obtaining a response rate based on the first dialogue indicator and the second dialogue indicator, and obtaining a response willingness rate based on the second dialogue indicator and the third dialogue indicator.

[0090] In one embodiment, the response rate=the first indicator divided by the second indicator; the response willingness rate=the third indicator divided by the second indicator.

[0091] See also Figure 3 , Figure 3 This is a flow chart of a method for determining a high-willing setting script to be signed according to an embodiment of the present application. Figure 3 In the example, the speech template also includes a reply template statement corresponding to the outbound call template statement, and the method for determining the high-willingness setting speech response includes:

[0092] Step S310, obtaining a reply template sentence corresponding to each round of dialogue and the outbound call template sentence and a user reply content output by the user during each round of dialogue, wherein the reply template sentence includes a high willingness keyword for indicating high willingness;

[0093] Step S320, comparing the reply template sentence with the user reply content to identify high-intention keywords in the user reply content;

[0094] Step S330, based on the high-willingness keywords in the user's reply content, determine whether the user reply content output by the user is a high-willingness set speech response.

[0095] Since the reply template sentence includes high-willingness keywords, the reply template sentence is compared with the user's reply content to determine whether there are high-willingness keywords in the user's reply content. If there are high-willingness keywords, the user's reply in the current round of conversation is considered to be a high-willingness reply. On the contrary, if there are low-willingness keywords in the reply template sentence, if, through comparison, low-willingness keywords are found in the user's reply content, the user's reply in the current round of conversation is considered to be a low-willingness reply.

[0096] See also Figure 4 , Figure 4 This is a flow chart of a method for determining a high-willing setting speech response according to an embodiment of the present application. Figure 4 As shown, the method for determining the high-willingness setting speech response includes:

[0097] Step S410, obtaining the content of each round of dialogue and the user response content output by the user;

[0098] Step S420, extracting features of the user's reply content to obtain voice features and text features;

[0099] In one embodiment, voice features include parameters such as pitch, volume, speaking speed and timbre, which can reflect the user's emotional state and the intensity of language expression in the conversation. Text features involve the extraction of keywords, including but not limited to positive words, negative words, interrogative sentences, declarative sentences, etc. in the user's reply. These features help analyze the user's willingness to reply. By comprehensively analyzing these features, the user's willingness to reply can be judged more accurately, thereby providing a basis for subsequent dialogue strategies.

[0100] Step S430, performing emotion recognition on the speech features based on the pre-trained emotion recognition model to obtain the user's emotion classification;

[0101] The emotion recognition model is obtained by training a neural network model. After the speech features are input into the emotion recognition model and the speech features are processed, the emotion classification of the user is obtained.

[0102] Step S440, performing intent recognition on text features based on a pre-trained intent recognition model to obtain a user intent classification;

[0103] The intent recognition model is trained by a neural network model. The text features are input into the intent recognition model, and the intent recognition model processes the input text features to obtain the user's intent classification.

[0104] Step S450, identifying the conversation content based on the pre-trained semantic continuity recognition model to obtain a conversation continuity classification;

[0105] The semantic continuity recognition model is trained by a neural network model. The conversation content is input into the semantic continuity recognition model. The semantic continuity recognition model processes the input conversation content to obtain the user's speech continuity classification.

[0106] Step S460, determining whether the user response content output by the user is a high-willing set speech response based on the user's emotion classification, the user's intention classification, and the conversation continuity classification.

[0107] If the user's emotion classification shows positive, the intent classification shows that the user has a clear need, and the conversation continuity classification shows that the user is highly concerned about the topic, the user's reply content is determined to be a high-willingness set speech response. On the contrary, if any of the indicators of the user's emotion classification, intent classification, or conversation continuity classification does not support a high-willingness response, the user's reply content is determined to be a low-willingness set speech response.

[0108] In one embodiment, the method further comprises:

[0109] The outbound call template statements corresponding to the low-willingness users are rewritten according to the outbound call template statements corresponding to the user response content of the high-willingness users.

[0110] In order to improve the success rate of outbound calls, the outbound call template statements of high-willing users are analyzed, key information and positive expressions are extracted, and incorporated into the outbound call template statements of low-willing users. In this way, even when users show low willingness, the outbound call robot can interact with users in a more personalized and attractive way, thereby improving user engagement and response rate. In addition, the outbound call template statements will be continuously optimized and adjusted based on user feedback and interaction results to ensure that the outbound call strategy is always consistent with the user's needs and emotional state.

[0111] It should be understood that the size of the serial numbers of the steps in the above embodiments does not 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 embodiments of the present application.

[0112] Figure 5 FIG. 1 is a block diagram of an indicator change attribution device for an outbound call robot according to an embodiment of the present application. Figure 5 As shown, an indicator change attribution device of an outbound call robot includes:

[0113] An outbound calling module 510 is used to perform a one-time outbound calling operation on a user with a non-member ID through an outbound calling robot based on a preset speech template, wherein the preset speech template includes an outbound calling template sentence;

[0114] The detection module 520 is used to detect the indicator change event and obtain the changed influencing factor associated with the indicator change event when the indicator change event is detected.

[0115] A user classification module 530, for classifying users into high-willing users and low-willing users according to the influencing factors;

[0116] The attribution summary module 540 is used to perform attribution analysis based on the outbound call template sentences corresponding to the user reply contents of high-willingness users and the outbound call template sentences corresponding to the user reply contents of low-willingness users to obtain attribution results.

[0117] It should be noted that the indicator change attribution device for the outbound call robot provided in the above embodiment and the indicator change attribution method for the outbound call robot provided in the above embodiment belong to the same concept, wherein the specific manner in which each module and unit performs the operation has been described in detail in the method embodiment and will not be repeated here. In actual application, the indicator change attribution device for the outbound call robot provided in the above embodiment can allocate the above functions to different functional modules as needed, that is, divide the internal structure of the device into different functional modules to complete all or part of the functions described above, and this is not limited here.

[0118] An embodiment of the present application also provides a device, including: one or more processors; and a memory for storing one or more programs. When the one or more programs are executed by one or more processors, the memory implements the indicator change attribution method of the outbound call robot in the above embodiment.

[0119] An embodiment of the present application also provides one or more machine-readable media on which instructions are stored, which, when executed by one or more processors, enable the processor to execute the indicator change attribution method of the outbound call robot in the above embodiment.

[0120] Figure 6 The schematic diagram of the structure of a computer system suitable for implementing the memory of the embodiment of the present application is shown. It should be noted that: Figure 6 The computer system of the memory shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0121] like Figure 6 As shown, the computer system includes a central processing unit (CPU), which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) or the program loaded from the storage part to the random access memory (RAM), such as the method in the above embodiment. In the RAM, various programs and data required for system operation are also stored. The CPU, ROM and RAM are connected to each other through a bus. The input / output (I / O) interface is also connected to the bus.

[0122] The following components are connected to the I / O interface: an input part including a keyboard, a mouse, etc.; an output part including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker; a storage part including a hard disk, etc.; and a communication part including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication part performs communication processing via a network such as the Internet. A drive is also connected to the I / O interface as needed. Removable media, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., are installed on the drive as needed so that a computer program read therefrom is installed into the storage part as needed.

[0123] In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains a computer program for executing the aforementioned implementation of the outbound call robot indicator change attribution method. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part, and / or installed from a removable medium. When the computer program is executed by the central processing unit (CPU), the various functions defined in the system of the present application are executed.

[0124] It should be noted that the computer-readable medium shown in the embodiment of the present application can be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium can be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, wherein a computer-readable computer program is carried. This propagated data signal can take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. A computer program contained on a computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0125] The flowchart and block diagram in the accompanying drawings illustrate the possible architecture, functions and operations of the system, method and computer program product according to various embodiments of the present application. Wherein, each box in the flowchart or block diagram can represent a module, a program segment, or a part of the code, and the above-mentioned module, program segment, or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0126] The units involved in the embodiments described in this application may be implemented by software or hardware, and the units described may also be set in a processor. The names of these units do not, in some cases, constitute limitations on the units themselves.

[0127] Another aspect of the present application further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor of a computer, causes the computer to execute the aforementioned index change attribution method for an outbound call robot. The computer-readable storage medium may be included in the memory described in the above embodiment, or may exist independently without being assembled into the memory.

[0128] Another aspect of the present application further provides a computer program product or a computer program, which includes a computer instruction stored in a computer-readable storage medium. A processor of a computer device reads the computer instruction from the computer-readable storage medium, and the processor executes the computer instruction, so that the computer device executes the indicator change attribution method of the outbound call robot provided in each of the above embodiments.

[0129] The above embodiments are merely illustrative of the principles and effects of the present application, and are not intended to limit the present application. Anyone familiar with the technology may modify or change the above embodiments without violating the spirit and scope of the present application. Therefore, all equivalent modifications or changes made by a person of ordinary skill in the art without departing from the spirit and technical ideas disclosed in the present application shall still be covered by the claims of the present application.

Claims

1. A method for attributing index changes of an outbound call robot, characterized in that: The attribution methods include: Performing a one-time outbound call operation on a user with a non-member ID through an outbound call robot based on a preset speech template, wherein the preset speech template includes an outbound call template sentence; Detecting an indicator change event and obtaining an impact factor associated with the indicator change event when the indicator change event is detected; Classifying the users into high-willingness users and low-willingness users according to the influencing factors; Attribution analysis is performed based on the outbound call template sentences corresponding to the user reply content of the high-willingness user and the outbound call template sentences corresponding to the user reply content of the low-willingness user to obtain an attribution result.

2. The indicator change attribution method of the outbound call robot according to claim 1 is characterized in that: The indicator change events include at least one of the following: an increase or decrease in the number of users with member identification, an increase or decrease in the average call duration, an increase or decrease in the average call rounds, an increase or decrease in the first round hang-up rate, an increase or decrease in the outbound call conversion rate, and an increase or decrease in the connection conversion rate; the influencing factors include at least one of the following: reply rate, reply willingness rate, reply conversion rate, rounds after reply, setting behavior, purchase behavior, click behavior, speech hang-up rate, dialogue node stay rounds, dialogue node stay duration, dialogue node reach rate, dialogue node hang-up rate, session willingness rate, and conversion rate in each time period.

3. The indicator change attribution method of the outbound call robot according to claim 2 is characterized in that: If the influencing factors include the response rate and the response willingness rate, obtaining the changed influencing factors associated with the indicator change event includes: Acquire the conversation content of at least one round of conversation between two parties, where the two parties include an outbound calling robot and a user with a non-member identifier; Identify the conversation content and obtain conversation indicators; the conversation indicators include a first conversation indicator corresponding to the user, a second conversation indicator corresponding to the outbound call robot, and a third conversation indicator, the first conversation indicator is used to indicate the number of responses to the set words, the second conversation indicator is used to indicate the number of triggered set words, and the third conversation indicator includes the number of responses to the set words with high willingness; The response rate is obtained based on the first dialogue indicator and the second dialogue indicator, and the response willingness rate is obtained based on the second dialogue indicator and the third dialogue indicator.

4. The indicator change attribution method of the outbound call robot according to claim 3 is characterized in that: The response rate = the first index divided by the second index; the response willingness rate = the third index divided by the second index.

5. The indicator change attribution method of the outbound call robot according to claim 3 is characterized in that: The speech template also includes a reply template statement corresponding to the outbound call template statement, and the method for determining the high-willingness setting speech response includes: Acquire a reply template sentence corresponding to each round of dialogue and the outbound call template sentence and a user reply content output by the user during each round of dialogue, wherein the reply template sentence includes a high willingness keyword for indicating high willingness; Comparing the reply template sentence with the user reply content to identify high-intention keywords in the user reply content; Based on the high-willingness keywords in the user reply content, it is determined whether the user reply content output by the user is a high-willingness set speech response.

6. The indicator change attribution method of the outbound call robot according to claim 3 is characterized in that: The method for determining the high-willingness setting speech response includes: Obtain the content of each round of dialogue and the user's response content output by the user; Extracting features of the user's reply content to obtain voice features and text features; Performing emotion recognition on the speech features based on a pre-trained emotion recognition model to obtain the user's emotion classification; Performing intent recognition on the text features based on a pre-trained intent recognition model to obtain a user intent classification; Recognize the conversation content based on a pre-trained semantic continuity recognition model to obtain a conversation continuity classification; Determine whether the user response content output by the user is a high-willing set speech response based on the user's emotion classification, the user's intention classification, and the conversation continuity classification.

7. The indicator change attribution method of the outbound call robot according to claim 5 or 6, characterized in that: The method further comprises: The outbound call template statements corresponding to the low-willingness users are rewritten according to the outbound call template statements corresponding to the user response content of the high-willingness users.

8. An indicator change attribution device for an outbound call robot, characterized in that: The attribution device comprises: An outbound calling module, used for performing a one-time outbound calling operation on a user with a non-member ID through an outbound calling robot based on a preset speech template, wherein the preset speech template includes an outbound calling template sentence; A detection module, used to detect an indicator change event and, when the indicator change event is detected, obtain an impact factor associated with the indicator change event that has changed; A user classification module, used for classifying the users into high-willingness users and low-willingness users according to the influencing factors; The attribution summary module is used to perform attribution analysis based on the outbound call template sentences corresponding to the user reply content of the high-willingness user and the outbound call template sentences corresponding to the user reply content of the low-willingness user to obtain the attribution result.

9. An indicator change attribution device, characterized in that: include: one or more processors; and A memory, used to store one or more programs. When the one or more programs are executed by the one or more processors, the memory implements the indicator change attribution method as described in any one of claims 1 to 7.

10. A machine-readable medium, characterized in that Instructions are stored thereon, which, when executed by one or more processors, enable the processors to execute the indicator change attribution method as described in any one of claims 1-7.

Images