An interactive data processing method, device, equipment, medium and program product applied to an intelligent dialogue robot

Abstract

The application relates to an interactive data processing method and device applied to an intelligent dialogue robot, equipment, a medium and a program product. The method comprises the following steps: screening an interactive data set corresponding to a current user from an interactive database based on a preset statistical period; each piece of interactive data in the interactive data set comprises at least one intention label and at least one performance element; determining a target data set from the interactive data set based on a preset intention label slot value; the target data set comprises all interactive data under a target interactive data corresponding account subject; the target interactive data comprises interactive data meeting the preset intention label slot value; calculating statistical parameters of each performance element in the target data set; generating and displaying distribution data of the statistical parameters based on a preset attention dimension; and the preset attention dimension comprises a subject dimension or a time dimension. The method can realize data screening and display corresponding analysis results based on the characteristics of interactive data, and improve data quality.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of artificial intelligence, in particular to an interactive data processing method, device, equipment, medium and program product applied to an intelligent dialogue robot. BACKGROUND

[0002] With the rapid development of e-commerce and globalization trade, logistics, marketing, financial and other customer service call scenarios play an increasingly important role in modern economy. The complexity and diversity of customer service make customers increasingly demand services such as information inquiry, order tracking, complaint handling. The traditional customer service mode mainly relies on manual customer service, which has the following problems: high labor cost: manual customer service requires a large number of human resources, especially during the peak period of logistics (such as holidays or promotional activities), the work load of customer service personnel increases sharply, resulting in the rise of enterprise operating costs. Slow response: the processing efficiency of manual customer service is limited, customers often need to wait for a long time during the peak period of consultation, affecting customer experience. Limited service time: manual customer service can usually only provide services within fixed working hours, and cannot meet the customers' all-day consultation needs. Limited information processing capacity: manual customer service is prone to errors or omissions when dealing with complex inquiries or large amounts of data, affecting service quality.

[0003] With the development of artificial intelligence technology, intelligent dialogue robots have emerged to replace the above-mentioned manual customer service. The SaaS platform of the intelligent dialogue robot mainly provides a series of capabilities for enterprise users to make intelligent outbound calls through robots, including voice dialogue robot call task configuration, call strategy configuration, workflow configuration, more touch mode configuration, detailed data after dialing and data analysis, etc., to help enterprise users improve business conversion rate.

[0004] However, when managing intelligent outbound calls, a large number of outbound data history parameters are not convenient for enterprise users to display and analyze, making it difficult to obtain useful information from a large amount of outbound data. SUMMARY

[0005] Therefore, it is necessary to provide an interactive data processing method, device, equipment, medium and program product applied to an intelligent dialogue robot, which can realize data filtering based on the characteristics of interactive data and display the corresponding analysis results.

[0006] In a first aspect, the present application provides an interactive data processing method applied to an intelligent dialogue robot, the method comprising:

[0007] Based on a preset statistical period, an interactive data set corresponding to a current user is filtered from an interactive database; each piece of interactive data in the interactive data set includes at least one intent label and at least one fulfillment element;

[0008] determining a target data set from the interaction data set based on a preset intention label slot value; the target data set comprises all the interaction data under a target interaction data corresponding account subject; the target interaction data comprises the interaction data conforming to the preset intention label slot value;

[0009] calculating a statistical parameter of each of the performance elements in the target data set;

[0010] generating and displaying distribution data of the statistical parameter based on a preset attention dimension; the preset attention dimension comprises a subject dimension or a time dimension.

[0011] In one of the embodiments, after the determination of the target data set from the interaction data set, and before the generation and display of the distribution data of the statistical parameter based on the preset attention dimension, the method further comprises:

[0012] In the case where the account identification dimension deduplication rule is configured, the interaction data in the target data set is merged based on the account identification;

[0013] Based on the merged interaction data in the target data set, the statistical parameter corresponding to each dimension value of the attention dimension is determined.

[0014] In one of the embodiments, the method further comprises:

[0015] In the case where the account identification dimension deduplication rule is not configured, or the account identification dimension of the interaction data in the target data set is empty, the interaction data in the target data set is merged based on the outbound call number and / or the business name.

[0016] In one of the embodiments, after the determination of the statistical parameter corresponding to each dimension value of the attention dimension, the method further comprises:

[0017] In the case where the account identification corresponds to multiple outbound call numbers, the outbound call number with the most recent outbound call time is determined as the outbound target number corresponding to the account identification, and the outbound target number corresponding to each account identification corresponding to each dimension value is displayed.

[0018] In one of the embodiments, the performance elements at least comprise a performance date; the method further comprises:

[0019] periodically updating the interaction data set;

[0020] In the case where the preset intention label slot value of the target interaction data is converted into a non-pre-set intention label slot value, in the case where the performance date arrives, the remaining performance elements are updated, and the distribution data is corrected based on the updated performance elements.

[0021] In one of the embodiments, the method further comprises:

[0022] Based on the current user's data acquisition request for the to-be-queried account subject, the target interaction data and non-target interaction data corresponding to the to-be-queried account subject in the preset statistical period are screened out; the non-target interaction data includes the interaction data that does not conform to the preset intent label slot value;

[0023] The performance elements of the target interaction data and / or the performance elements of the non-target interaction data are displayed.

[0024] In one of the embodiments, the method further comprises:

[0025] When the target interaction data or non-target interaction data corresponding to the to-be-queried account subject is not screened out in the preset statistical period, the performance elements in the interaction data corresponding to the most recent interaction time of the to-be-queried account subject in the preset statistical period are displayed.

[0026] In one of the embodiments, the method further comprises:

[0027] Based on user configuration information, a dialogue summary field is added in the distribution data; the user configuration information includes NER slot description and event label;

[0028] The dialogue summary field is generated based on the intent label in each piece of the interaction data in the target data set, the hit NER slot description, the hit NER slot value, and the hit event label.

[0029] In a second aspect, the application further provides an intelligent dialogue robot interaction method, which comprises:

[0030] Obtaining a distribution data display request configured by a current user; the distribution data display request includes a preset statistical period, a to-be-displayed performance element, a preset intent label slot value, a to-be-displayed statistical parameter, and a preset attention dimension;

[0031] Based on the statistical period, an interaction data set corresponding to the current user is screened from an interaction database; each piece of interaction data in the interaction data set includes at least one intent label and at least one performance element;

[0032] Based on the preset intent label slot value, a target data set is determined from the interaction data set; the target data set includes all the interaction data under the account subject corresponding to the target interaction data; the target interaction data includes the interaction data that conforms to the preset intent label slot value;

[0033] calculating the statistical parameter of each of the performance elements in the target data set;

[0034] generating and displaying distribution data of the statistical parameter based on the preset attention dimension; the preset attention dimension includes a subject dimension or a time dimension.

[0035] In one of the embodiments, the method further comprises:

[0036] obtaining a deduplication rule configured by the current user;

[0037] when the deduplication rule is an account identifier dimension deduplication rule, merging the interaction data in the target data set based on account identifier;

[0038] based on the merged interaction data in the target data set, determining the statistical parameter corresponding to each dimension value of the attention dimension.

[0039] In one of the embodiments, the method further comprises:

[0040] obtaining a data acquisition request configured by the current user, the data acquisition request including a to-be-queried account subject;

[0041] based on the data acquisition request, filtering out the target interaction data and non-target interaction data corresponding to the to-be-queried account subject in the preset statistical period; the non-target interaction data includes the interaction data not meeting the preset intent label slot value;

[0042] displaying the performance elements of the target interaction data and / or the performance elements in the non-target interaction data.

[0043] In one of the embodiments, the method further comprises:

[0044] obtaining user configuration information; the user configuration information includes NER slot description and event label;

[0045] based on the user configuration information, displaying a conversation summary field in the distribution data; the conversation summary field is generated based on the intent label in each of the interaction data in the target data set, the hit NER slot description, the hit NER slot value, and the hit event label.

[0046] In a third aspect, the application further provides an interaction data processing device applied to an intelligent dialogue robot, the device comprising:

[0047] The first data acquisition module is configured to filter an interaction data set corresponding to the current user from an interaction database based on a preset statistical period; each piece of interaction data in the interaction data set comprises at least one intention label and at least one performance element;

[0048] The first screening module is configured to determine a target data set from the interaction data set based on a preset intention label slot value; the target data set comprises all the interaction data under a target interaction data corresponding account subject; and the target interaction data comprises the interaction data meeting the preset intention label slot value.

[0049] The first statistical parameter determination module is configured to calculate statistical parameters of each performance element in the target data set.

[0050] The first display module is configured to generate and display distribution data of the statistical parameters based on a preset attention dimension; and the preset attention dimension comprises a subject dimension or a time dimension.

[0051] In a fourth aspect, the present application further provides an intelligent dialogue robot interaction device, which comprises:

[0052] The request acquisition module is configured to acquire a distribution data display request configured by a current user; and the distribution data display request comprises a preset statistical period, a performance element to be displayed, a preset intention label slot value, a statistical parameter to be displayed, and a preset attention dimension.

[0053] The second data acquisition module is configured to filter an interaction data set corresponding to the current user from an interaction database based on the statistical period; each piece of interaction data in the interaction data set comprises at least one intention label and at least one performance element.

[0054] The second screening module is configured to determine a target data set from the interaction data set based on the preset intention label slot value; the target data set comprises all the interaction data under a target interaction data corresponding account subject; and the target interaction data comprises the interaction data meeting the preset intention label slot value.

[0055] The second statistical parameter determination module is configured to calculate the statistical parameters of each performance element in the target data set.

[0056] The second display module is configured to generate and display distribution data of the statistical parameters based on the preset attention dimension; and the preset attention dimension comprises a subject dimension or a time dimension.

[0057] In a fifth aspect, the present application further provides a computer device comprising a memory and a processor; the memory stores a computer program; and the processor implements the steps of the method in any one of the above-mentioned embodiments when executing the computer program.

[0058] In a sixth aspect, the present application also provides a computer readable storage medium, having stored thereon a computer program, which when executed by a processor implements the steps of the method in any one of the above embodiments.

[0059] In a seventh aspect, the present application also provides a computer program product comprising a computer program, which when executed by a processor implements the steps of the method in any one of the above embodiments.

[0060] The above-mentioned interaction data processing method, device, equipment, medium and program product applied to the intelligent dialogue robot, based on a preset statistical period, the interaction database is filtered from the interaction data set corresponding to the current user; each piece of interaction data in the interaction data set includes at least one intention label and at least one performance element; based on the preset intention label slot value, the target data set is determined from the interaction data set; the target data set includes all interaction data under the target interaction data corresponding account subject; the target interaction data includes the interaction data meeting the preset intention label slot value; the statistical parameters of each performance element in the target data set are calculated; the distribution data of the statistical parameters based on the preset attention dimension is generated and displayed; the preset attention dimension includes the subject dimension or the time dimension, so that the interaction data set in the interaction database is filtered based on the unique intention label of the interaction data to obtain the target data set, and then the statistical parameters of each performance element in the target data set are calculated, so that the demand based on the preset attention dimension is displayed, useful information is obtained, and the quality of the interaction data is improved. BRIEF DESCRIPTION OF DRAWINGS

[0061] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related art, the drawings needed to be used in the description of the embodiments of the present application or the related art will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other related drawings can also be obtained without creative labor on the basis of these drawings.

[0062] Figure 1 An application environment diagram of the interaction data processing method applied to the intelligent dialogue robot in an embodiment;

[0063] Figure 2 A processing schematic diagram in the process of voice interaction between the dialogue robot and the user terminal in an embodiment;

[0064] Figure 3 A flowchart of the interaction data processing method applied to the intelligent dialogue robot in an embodiment;

[0065] Figure 4 A flowchart of the target data merging step in an embodiment;

[0066] Figure 5 a flowchart of an intelligent dialogue robot interaction method in an embodiment;

[0067] Figure 6 a structural block diagram of an apparatus for an interaction data processing method applied to an intelligent dialogue robot in an embodiment;

[0068] Figure 7 a structural block diagram of an intelligent dialogue robot interaction apparatus in an embodiment;

[0069] Figure 8 an internal structure diagram of a computer device in an embodiment. DETAILED DESCRIPTION

[0070] In order to make the purposes, technical solutions and advantages of the present application clearer, the present application is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the present application.

[0071] It should be noted that the terms "first", "second", etc. used in the present application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "include" and "have" used in the present application and any variations thereof are intended to cover non-exclusive inclusion. The term "a plurality of" used in the present application means two and more than two. The term "and / or" used in the present application means one of the options or any combination of a plurality of options.

[0072] The interaction data processing method of the intelligent dialogue robot provided by the embodiments of the present application can be applied in an application environment as shown in Figure 1 The intelligent dialogue robot platform can provide dialogue robots for enterprise users, and the dialogue robots communicate with various user terminals. The intelligent dialogue robot platform includes functions such as dialogue robot call task configuration, call strategy configuration, workflow configuration, more touch mode configuration, detailed data after dialing, and data analysis.

[0073] Specifically, in combination with the application environment shown in Figure 2 Figure 2 ​A processing schematic diagram in a voice interaction process between a dialogue robot and a user terminal in an embodiment, wherein after receiving voice information sent by the user terminal, the voice information is subjected to voice recognition, semantic understanding, dialogue management and reply generation to complete a round of voice interaction. Specifically, the voice recognition includes processing the voice information by ASR (Automatic Speech Recognition) to obtain a data stream status, robot speaking stage interruption logic processing, user speaking judgment logic processing and ASR text generation logic processing. The semantic understanding fills corresponding slots based on the results of intent recognition, specifically, corresponding entities can be recognized by NER (Named Entity Recognition) and corresponding slots are filled. The dialogue management includes state tracking and dialogue strategy, wherein the state tracking includes t time state, t time system behavior and t+1 time user behavior (including intent, slot and variable), the dialogue strategy can include label inheritance strategy, slot filling strategy, variable system, objection processing logic, function execution logic and free question and answer strategy. The reply generation includes variable splicing, slot splicing, dialogue generation logic and audio broadcast process and the like.

[0074] Wherein the enterprise user can manage the above dialogue logic process and obtain dialogue history record data through the SaaS platform. The current call detail interface can display the task name, dialogue robot model, outbound number, call time and user intent corresponding to each call ID; for each outbound task, the call duration, ring duration, ring classification, whether to transfer to manual, dialogue round, total call number, hang-up type, ASR result and the like can be displayed, but the enterprise user analyzes the data inefficiently.

[0075] To solve the above technical problems, an interactive data processing method applied to an intelligent dialogue robot is proposed in the present application, comprising: filtering the interactive data set corresponding to the current user from the interactive database based on a preset statistical period; each piece of interactive data in the interactive data set includes at least one intent label and at least one performance element; determining a target data set from the interactive data set based on a preset intent label slot value; the target data set includes all interactive data under the account subject corresponding to the target interactive data; the target interactive data includes interactive data meeting the preset intent label slot value; calculating the statistical parameters of each performance element in the target data set; generating and displaying the distribution data of the statistical parameters based on a preset attention dimension; the preset attention dimension includes a subject dimension or a time dimension, so that the demand can be statistically displayed based on the preset attention dimension, useful information is obtained, the quality of the interactive data is improved, and multi-dimensional data details can be displayed according to the attention dimension of the enterprise user, helping the enterprise user better understand customer behavior and interaction effect.

[0076] The terminal can be, but is not limited to, various personal computers, notebook computers, smart phones, tablet computers, Internet of Things devices, and portable wearable devices. The Internet of Things device can be a smart speaker, a smart television, a smart air conditioner, a smart vehicle device, a projection device, etc. The portable wearable device can be a smart watch, a smart bracelet, a head-mounted device, etc. The head-mounted device can be a virtual reality (VR) device, an augmented reality (AR) device, smart glasses, etc. The intelligent conversational robot platform can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0077] In an exemplary embodiment, as shown in Figure 3 An interactive data processing method applied to an intelligent conversational robot is provided. The method is applied to an intelligent conversational robot platform in Figure 1 for example, and includes the following steps S302 to S308. In the steps

[0078] S302: Based on a preset statistical period, filtering an interactive data set corresponding to a current user from an interactive database. Each piece of interactive data in the interactive data set includes at least one intent label and at least one performance element.

[0079] The preset statistical period is determined by the intelligent conversational robot platform based on analysis requirements. The preset statistical period can be automatically determined by the intelligent conversational robot platform, or determined based on a setting of an enterprise user. For example, the intelligent conversational robot platform can receive a statistical period input by an enterprise user through an interface. Optionally, the statistical period can be one day, one week, one month, one year, etc. In other embodiments, the statistical period can also be other values, which are not specifically limited here.

[0080] The interactive database is a database in the intelligent conversational robot platform for storing interactive data between the intelligent conversational robot and the terminal. The interactive database can store interactive data of several users, or each user corresponds to an interactive database, which is not specifically limited here. The user refers to an enterprise user. Different enterprises correspond to different intelligent conversational robots. For example, after an enterprise accesses the intelligent conversational robot platform, the enterprise can be allocated a corresponding intelligent conversational robot, so as to establish a mapping relationship between the enterprise and the intelligent conversational robot. When there is a business requirement in the corresponding enterprise, the corresponding intelligent conversational robot is called based on the mapping relationship to process the corresponding business.

[0081] The interactive data set can include several pieces of interactive data, each piece of interactive data including at least one intent label and at least one performance element.

[0082] The intent tag is the information of the corresponding slot obtained by processing the interaction data based on the NER model during the semantic understanding process. The slot value of the intent tag can be limited based on the business scenario. In this application, the business scenario can include, but is not limited to, financial scenarios, marketing scenarios, and logistics scenarios. In the financial scenario, the slot value of the intent tag can include a confirmation of repayment tag and a confirmation of non-repayment tag. In the marketing scenario, the slot value of the intent tag can include a marketing success tag and a marketing failure tag. In the logistics scenario, the slot value of the intent tag can include a payment tag and a non-payment tag, etc. The above scenarios and corresponding intent tags are only illustrative examples. In other embodiments, those skilled in the art can set the corresponding intent tags based on specific business scenarios.

[0083] Performance elements include account tags, amount, and time. The account tag refers to the tag of the entity, which is the account. Different tags can be set based on the business scenario to indicate the account category. For example, in a financial scenario, the account tag could include Broken PTP account (which can be understood as an account with confirmed outstanding payments) and PTP account (which can be understood as an account with confirmed repayments). The amount can be determined based on the business scenario, such as the performance amount. The time indicates the performance date, i.e., the time of the intended action.

[0084] When intelligent chatbots process interaction data, they can determine the intent tags and fulfillment elements corresponding to the interaction data and store them in the interaction database.

[0085] The business data insight module of the intelligent chatbot can obtain the statistical period and retrieve the interaction dataset corresponding to the current user from the interaction database based on the preset statistical period. Specifically, the business data insight module obtains the interaction dataset between the intelligent chatbot and the terminal for the current user within the time period corresponding to the preset statistical period.

[0086] S304: Determine the target dataset from the interaction dataset based on the preset intent tag slot values; the target dataset includes all interaction data under the account subject corresponding to the target interaction data; the target interaction data includes interaction data that conforms to the preset intent tag slot values.

[0087] The preset intent tag slot values ​​are determined based on the analysis requirements. For example, the interaction data in this application can be financial scenario interaction data, marketing scenario interaction data, or logistics scenario interaction data. In the financial scenario, the analysis requirements include target interaction data with value transfer confirmation, so the preset intent tag slot values ​​can include a repayment confirmation tag. In the marketing scenario, the analysis requirements include target interaction data with successful marketing and target interaction data with unsuccessful marketing, so the preset intent tag slot values ​​can include a marketing success tag and a marketing failure tag. In the logistics scenario, the analysis requirements include target interaction data with value transfer confirmation corresponding to waybills or goods, so the preset intent tag slot values ​​can include a payment tag and a non-payment tag. In other embodiments, other business scenarios may also be included, and the analysis requirements corresponding to the business scenarios can also be set based on user needs, without specific limitations. Therefore, in this application, the analysis requirements can be determined based on the business scenario, and the analysis requirements correspond to the type of target interaction data. For example, in the marketing scenario, the target interaction data includes target interaction data with successful marketing and target interaction data with unsuccessful marketing.

[0088] The account subject is the dialogue partner of the intelligent chatbot, and this account subject can be distinguished based on the account identifier (case ID). The target dataset includes all interaction data under the account subject corresponding to the target interaction data; all interaction data belongs to the target dataset.

[0089] In this application, after determining the interaction dataset, the interaction data in the interaction dataset is filtered based on the preset intent label slot value to obtain the target interaction data that meets the preset intent label slot value.

[0090] Furthermore, this application selects the intent behavior time as the valid value from the interaction dataset, and the slot value of the intent label is the interaction data of the target label as the target interaction data. Optionally, each target label can be determined first based on the analysis requirements, and then the intent behavior time is selected as the valid value from the interaction dataset, and the slot value of the intent label is the target interaction data of each target label. For example, in a financial scenario, only the intent behavior time as the valid value needs to be selected, and the intent label data is the target interaction data of the confirmed value transfer label. In a marketing scenario, each target label can be determined first based on the analysis requirements, such as the marketing success label and the marketing failure label, and then the intent behavior time as the valid value is selected from the interaction data, and the slot value of the intent label is the target interaction data of the marketing success label and the target interaction data of the marketing failure label, respectively. Optionally, different processing methods or the same processing method can be used for different types of target interaction data, and no specific limitation is made here.

[0091] The intent action time (PTP date) mentioned here refers to the slot information obtained by processing the interaction data based on the NER model during semantic understanding. Generally, the slot corresponding to the intent action time PTP date is promise_to_pay_time. The interaction dataset can be filtered based on whether there is a valid value for this slot and whether the slot value of the intent label matches the target label.

[0092] S306: Calculate the statistical parameters of each performance element in the target dataset.

[0093] The limitations on the performance elements can be found above and will not be repeated here. The statistical parameters include the results obtained based on various statistical methods. In this application, the statistical parameters corresponding to different business scenarios may be the same or different, and no specific limitations are made here.

[0094] Taking a financial scenario as an example, statistical parameters can include the number of different types of account tags, the number of time periods, and the number of amounts. For example, in a financial scenario, the statistical parameters can include the total number of accounts corresponding to the target account tag, the amount of the target intent tag of each account corresponding to the target account tag, the total number of accounts corresponding to non-target account tags, and the amount of the non-target intent tag of each account corresponding to the non-target account tag.

[0095] S308: Generate and display the distribution data of statistical parameters based on preset focus dimensions; preset focus dimensions include subject dimension or time dimension.

[0096] The preset attention dimensions for different business scenarios can be the same or different. Preset attention dimensions can include subject dimensions or time dimensions. Time dimensions can include the time of intent behavior, other time dimensions, spatial dimensions, etc., without specific limitations. Attention dimensions include different dimension values. For example, if the preset attention dimension is the time of intent behavior, then the corresponding dimension value can be a specific time value.

[0097] The subject dimension generates and displays corresponding statistical parameters based on different account subjects, such as the total amount of money associated with each account subject. The time dimension represents the statistical parameters corresponding to each moment within a preset statistical period, increasing by a preset step size, such as the number of account subjects and the total amount of money associated with each date. This allows for a clear and intuitive display of the target interaction data. By filtering and processing this data, data quality can be improved, and the intuitive display of distribution helps enterprise users better understand customer behavior and interaction effects.

[0098] The aforementioned interactive data processing method applied to intelligent chatbots filters the interactive dataset corresponding to the current user from the interactive database based on a preset statistical period. Each piece of interactive data in the interactive dataset includes at least one intent tag and at least one fulfillment element. Based on preset intent tag slot values, a target dataset is determined from the interactive dataset. The target dataset includes all interactive data under the account subject corresponding to the target interactive data. The target interactive data includes interactive data that meets the preset intent tag slot values. Statistical parameters of each fulfillment element in the target dataset are calculated. Distribution data of statistical parameters based on preset attention dimensions is generated and displayed. The preset attention dimensions include subject dimension or time dimension. In this way, the interactive dataset in the interactive database is filtered based on the unique intent tags of the interactive data to obtain the target dataset. Then, the statistical parameters of each fulfillment element in the target dataset are calculated. This allows for the statistical display of needs based on preset attention dimensions, obtaining useful information and improving the quality of interactive data.

[0099] In one optional embodiment, after determining the target dataset from the interaction dataset and before generating and displaying the distribution data of statistical parameters based on the preset attention dimension, the method further includes: merging the target interaction data in the target dataset based on the account identifier when the deduplication rule for the account identifier dimension is configured; and determining the statistical parameter corresponding to each dimension value of the attention dimension based on the target interaction data in the merged target dataset.

[0100] In one optional embodiment, the method further includes: merging the target interaction data in the target dataset based on the outbound call number and / or service name when no deduplication rule for the account identifier dimension is configured, or when the account identifier dimension of the target interaction data in the target dataset is empty.

[0101] In this embodiment, the business data management module of the intelligent chatbot processes the data. After the business data insight module of the intelligent chatbot obtains the interaction data and filters the data to obtain the target interaction data, the business data insight module can send the target interaction data to the business data management module so that the business data can process the target interaction data to obtain multi-dimensional data details (detailed information of each item parameter) within the preset attention dimensions required by the user, which facilitates the subsequent display of multi-dimensional data details.

[0102] The business data management module can merge target interaction data to deduplicate it. This prevents data distortion caused by different results from multiple outbound calls, especially in scenarios such as logistics, marketing, and finance. Distorted data can lead to incorrect customer intent statistics for enterprise users, affecting the execution of subsequent business plans and the configuration of outbound call plans. By first merging the data and then processing the interaction data based on the merged data, the module avoids data distortion caused by different results from multiple outbound calls.

[0103] Specifically, since one account can correspond to multiple outbound numbers, this application, with deduplication rules configured at the account identifier dimension, merges the target interaction data based on the account identifier. Specifically, for cases where one account identifier (case ID) makes multiple calls to different outbound numbers, statistics are compiled based on the account identifier (case ID), and call results are updated according to call duration. Calls without an account identifier (case ID) are not included in the statistics.

[0104] If no deduplication rules are configured for the account identifier dimension, or if the account identifier case ID dimension is empty, the target interaction data can be merged according to the enterprise user's deduplication rules. Specifically, the target interaction data can be merged based on the outbound call number and / or business name. This may result in the same outbound call number belonging to two different account identifier case IDs, which is more in line with user scenarios. Optionally, deduplication can also be performed according to the business name configured by the enterprise user (multiple account identifier case IDs or outbound call numbers can be added under the same business name) to merge target interaction data for some key businesses together, facilitating user analysis.

[0105] For ease of understanding, the following is combined Figure 4 As shown, Figure 4 This is a flowchart of the target interaction data merging step in one embodiment. After obtaining the target interaction data, the deduplication rule configuration information is first determined. If the deduplication rule configuration information indicates that the account identifier dimension deduplication rule is configured, the target interaction data is merged based on the account identifier. For example, multiple target interaction data with different outbound call numbers but the same account identifier are merged. Target interaction data without an account identifier is not included in the statistics.

[0106] If the deduplication rule configuration information indicates that no deduplication rule is configured for the account identifier dimension, or if the account identifier dimension of the target interaction data is empty, then the target interaction data will be merged according to the deduplication rule configured by the enterprise user. Specifically, if the enterprise user configures the deduplication rule based on the outbound call number dimension, then the target interaction data will be merged based on the outbound call number; that is, target interaction data with the same outbound call number will be merged together. If the enterprise user configures the deduplication rule based on the business name, then target interaction data with the same business name will be merged together. For example, target interaction data corresponding to different goods with the same order number can be merged together.

[0107] Since a user or a single outbound number may make multiple outbound calls, the interaction data obtained from different rounds of outbound calls will be different, and the interaction results obtained based on this interaction data will also be different. Therefore, merging these target interaction data based on account identifier, outbound number, or business name can solve the data distortion problem in multi-round outbound call scenarios and effectively improve the efficiency of data display and analysis.

[0108] In one optional embodiment, after determining the statistical parameters corresponding to each dimension value of the focus dimension, the method further includes: when an account identifier corresponds to multiple outbound call numbers, determining the outbound call number with the most recent outbound call time as the outbound call target number corresponding to the account identifier, and displaying the outbound call target numbers corresponding to each account identifier for each dimension value.

[0109] In cases where an account identifier corresponds to multiple outbound call numbers, the outbound call number with the most recent outbound call time can be determined from among the multiple outbound call numbers as the outbound call target number corresponding to the account identifier, and the target number corresponding to each account identifier for each dimension value can be displayed. In this way, when displaying, the target number corresponding to each account identifier for each dimension value is also displayed.

[0110] In the above embodiments, the target number corresponding to each account identifier for each dimension value is also displayed, providing multi-dimensional information for easy viewing by users.

[0111] In one optional embodiment, the performance element includes at least the performance date; the method further includes: periodically updating the interaction dataset; when the preset intent label slot value of the target interaction data is converted to a non-preset intent label slot value, when the performance date arrives, updating the remaining performance elements, and correcting the distribution data based on the updated performance elements.

[0112] Performance elements include the performance date and other performance elements, which include at least the account tag.

[0113] The interaction dataset is updated periodically. Within a preset statistical period, the intelligent chatbot can communicate with the corresponding account entity to obtain interaction data and store it in the interaction database. Based on the preset statistical period, the interaction data in the interaction database can be obtained, and the corresponding target interaction data can be obtained in the manner described above.

[0114] The preset intent tag slot values ​​of the target interaction data may be converted to non-preset intent tag slot values. In this way, when the performance date arrives, the remaining performance elements are updated, and the distribution data is corrected based on the updated performance elements.

[0115] The remaining performance elements include at least an account tag, which can be obtained based on the intent tag. Specifically, if the intent tag is a preset intent tag slot value, the account tag is the target account tag; if the intent tag is not a preset intent tag slot value, the account tag is a non-target account tag. Taking the financial sector as an example, the business data management module obtains the target interaction data mentioned above and tags customers (the tags are affected by the time dimension and are only effective within the statistical period). The account corresponding to the target interaction data is tagged with the target account tag, such as the PTP tag, while the account corresponding to the non-target interaction data is tagged with a non-target account tag, such as the Broken PTP tag.

[0116] Thus, within the statistical period, there may be multiple interaction data for an account identifier case id. The account tag obtained based on the target interaction data may be different each time. Therefore, it is necessary to determine the account tag corresponding to the account identifier to avoid data distortion caused by different results from multiple rounds of interaction data.

[0117] If the slot value of the intent tag in the most recent interaction data corresponding to a certain account identifier is not the target intent tag, then the historical interaction data of that account identifier within a preset statistical period is obtained as the interaction data to be processed. Then, it is checked whether there exists an intent tag slot valued at the target intent tag in the interaction data to be processed, and the corresponding intent behavior time slot is an interaction data to be processed with a valid value. If so, when the intent behavior time (i.e., the fulfillment time) represented by the valid value is reached, the account tag corresponding to the user identifier is modified to a non-target account tag. Simultaneously, the total number of target intent tags (PTP Amount) for this account identifier (caseid) is transferred to the corresponding total number of non-target intent tags (Broken PTPAmount).

[0118] For example, the interaction data corresponding to an account identifier case ID includes: a phone call was made on June 1st, and the customer replied that the value transfer would take place on June 7th; then on June 7th, the target account tag, i.e., the number of PTP accounts, is 1, and the non-target account tag, i.e., the number of Broken PTP accounts, is 0; when another call is made on June 7th to urge the value transfer to proceed, the result is not the A tag, so the data for June 7th will change to the number of PTP accounts corresponding to the target account tag being 0, and the number of non-target account tags, i.e., the number of Broken PTP accounts, being 1. At the same time, the total number of target intent tags, PTP Amount, will be transferred to the total number of non-target intent tags, Broken PTP Amount. That is, originally June 7th was the target intent tag with a total of 1 target intent tag, but now June 7th is changed to a non-target intent tag with a total of 0 target intent tags and a total of 1 non-target intent tag.

[0119] In the above embodiments, by processing the multi-round interaction data, the account tag corresponding to the account is determined, thus avoiding data distortion caused by different results of multi-round interaction data.

[0120] In one optional embodiment, the method further includes: based on the current user's data retrieval request for the account subject to be queried, filtering out target interaction data and non-target interaction data corresponding to the account subject to be queried within a preset statistical period; non-target interaction data includes interaction data that does not conform to the preset intent tag slot value; displaying the fulfillment elements of the target interaction data, and / or, the fulfillment elements of the non-target interaction data.

[0121] In one optional embodiment, the method further includes: when no target interaction data or non-target interaction data corresponding to the account subject to be queried is found within a preset statistical period, displaying the performance elements in the interaction data of the most recent interaction time corresponding to the account subject to be queried within the preset statistical period.

[0122] The purpose of a data acquisition request is to obtain the corresponding interaction data. After receiving the data acquisition request, the system can collect the target interaction data within the preset statistical period, which is determined to be the target account tag, or the target interaction data which is determined to be the non-target account tag, and then display the corresponding target interaction data and non-target interaction data.

[0123] Non-target interaction data includes interaction data that does not conform to the preset intent tag slot value. Specifically, non-target account tag non-target interaction data generally includes at least two interaction data: one is a PTP indicating that the account is a target account, and the other is a Broken PTP indicating that the account is a non-target account. If, within the filtering period, the account is first determined to be a PTP indicating that it is a target account, then a Broken PTP indicating that it is a non-target account, then a PTP indicating that it is a target account, and then a Broken PTP indicating that it is a non-target account, then the same process is repeated, then all four interaction data need to be displayed.

[0124] Specifically, the business data management module displays the fulfillment elements of the target interaction data and / or the fulfillment elements of non-target interaction data, such as displaying the intent behavior time (PTP data) and intent tag (PTP Amount parameter). The latest PTP Amount field value is obtained based on the account tag and the case ID of each account. If the account tag is the target account tag (i.e., a PTP tag), the intent tag uploaded for the interaction data generated by the target account tag PTP tag is represented by the PTPAmount field value. If the account tag is a non-target account tag (i.e., a Broken PTP tag), the non-intent tag uploaded for the interaction data generated by the case ID (i.e., a Broken PTP tag) is represented by the PTPAmount field value. If there is no account tag, the intent tag uploaded for the most recently dialed interaction data is represented by the PTP Amount field value.

[0125] In other embodiments, the content displayed in the target interaction data may also include detailed data and NER slot values. The detailed data includes the task name, chatbot model, outbound number, call time, and user intent. For each outbound task, data such as call duration, ringing duration, ringing category, whether to transfer to a human operator, number of dialogue rounds, total number of calls, hang-up type, and ASR result may be displayed. NER slot values ​​include, but are not limited to, payment time, payment method, reason, PTP time, and PTP amount. In other embodiments, NER slot values ​​may also include other slot values, which are not specifically limited here.

[0126] In the above embodiments, the business data management module can display detailed data across multiple dimensions within the dimensions of interest as needed, ensuring data integrity.

[0127] In one optional embodiment, the method further includes: adding a dialogue summary field to the distributed data based on user configuration information; the user configuration information includes NER slot descriptions and event tags; the dialogue summary field is generated based on the intent tag, the hit NER slot description, the hit NER slot value, and the hit event tag in each interaction data in the target dataset.

[0128] The dialogue summary field is added to the detailed data of the interaction data in the business data management module based on enterprise user configuration.

[0129] The dialogue summary field is generated based on the intent label, NER slot description and slot value, and event label of each interaction in the target dataset. Optionally, the dialogue summary field consists of three sentences and is displayed in the detailed data of the interaction data.

[0130] For example, if the intent tag is "Confirm Repayment", the matched slot description is "PTP date", the slot value is "2024-07-12", and the matched event tag description is "No_serious_repayment_intent", then the dialogue summary field would be: "Willing to repay with a specific repayment date. PTP date: 2024-07-12. No serious repayment intent."

[0131] In the above embodiments, the dialogue summary field can generate a summary of each interaction data based on scenario requirements, including intent tags, NER slot descriptions and slot values, event tag descriptions, etc., to help enterprise users quickly understand the core content of the call and improve the efficiency of subsequent business decisions.

[0132] In one exemplary embodiment, such as Figure 5 As shown, an interaction method for an intelligent chatbot is provided, which can be applied to... Figure 1 Taking the intelligent chatbot platform as an example, the explanation includes the following steps S502 to S510.

[0133] S502: Request to display distributed data configured by the current user; the distributed data display request includes the preset statistical period, the performance elements to be displayed, the preset intent tag slot values, the statistical parameters to be displayed, and the preset attention dimensions.

[0134] S504: Based on the statistical period, filter the interaction dataset corresponding to the current user from the interaction database; each interaction data in the interaction dataset includes at least one intent tag and at least one fulfillment element.

[0135] S506: Determine the target dataset from the interaction dataset based on the preset intent tag slot values; the target dataset includes all interaction data under the account subject corresponding to the target interaction data; the target interaction data includes interaction data that conforms to the preset intent tag slot values.

[0136] S508: Calculate the statistical parameters of each performance element in the target dataset.

[0137] S510: Generate and display the distribution data of statistical parameters based on preset focus dimensions; preset focus dimensions include subject dimension or time dimension.

[0138] The difference between this embodiment and the one described above lies only in step S502. Step S502 is an interaction step with the user, which can obtain the distribution data display request configured by the current user. The distribution data display request includes a preset statistical period, the performance elements to be displayed, the preset intent tag slot values, the statistical parameters to be displayed, and the preset attention dimensions. Subsequently, the intelligent chatbot platform can process the data based on the preset statistical period, the performance elements to be displayed, the preset intent tag slot values, the statistical parameters to be displayed, and the preset attention dimensions to generate and display the distribution data of the statistical parameters based on the preset attention dimensions.

[0139] In one optional embodiment, the method further includes: obtaining the deduplication rule configured by the current user; when the deduplication rule is an account identifier dimension deduplication rule, merging the interaction data in the target dataset based on the account identifier; and determining the statistical parameter corresponding to each dimension value of the focus dimension based on the interaction data in the merged target dataset.

[0140] In one optional embodiment, the method further includes: obtaining a data acquisition request configured by the current user, the data acquisition request including the account subject to be queried; based on the data acquisition request, filtering out target interaction data and non-target interaction data corresponding to the account subject to be queried within a preset statistical period; the non-target interaction data includes interaction data that does not conform to the preset intent tag slot value; displaying the performance elements of the target interaction data, and / or the performance elements in the non-target interaction data.

[0141] In one optional embodiment, the method further includes: obtaining user configuration information; the user configuration information includes NER slot descriptions and event tags; displaying a dialogue summary field in the distribution data based on the user configuration information; the dialogue summary field is generated based on the intent tag, the hit NER slot description, the hit NER slot value, and the hit event tag in each interaction data in the target dataset.

[0142] In all the above embodiments, the user configuration information is obtained through interaction with the user. For specific limitations of other parts of the above embodiments, please refer to the above text, and will not be repeated here.

[0143] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0144] Based on the same inventive concept, this application also provides an interactive data processing device for intelligent chatbots to implement the above-described interactive data processing method for intelligent chatbots, and an intelligent chatbot dialogue device corresponding to the intelligent chatbot dialogue method. The solution provided by this device is similar to the solution described in the above method. Therefore, the specific limitations of the one or more interactive data processing devices and intelligent chatbot dialogue devices provided below can be found in the above-described limitations of the interactive data processing method and intelligent chatbot dialogue method for intelligent chatbots, and will not be repeated here.

[0145] In one exemplary embodiment, such as Figure 6 As shown, an interactive data processing device for intelligent chatbots is provided, comprising: a first data acquisition module 601, a first filtering module 602, a first statistical parameter determination module 603, and a first display module 604, wherein:

[0146] The first data acquisition module 601 is used to filter the interaction dataset corresponding to the current user from the interaction database based on a preset statistical period; each piece of interaction data in the interaction dataset includes at least one intent tag and at least one fulfillment element;

[0147] The first filtering module 602 is used to determine the target dataset from the interaction dataset based on the preset intent tag slot value; the target dataset includes all interaction data under the account subject corresponding to the target interaction data; the target interaction data includes interaction data that meets the preset intent tag slot value.

[0148] The first statistical parameter determination module 603 is used to calculate the statistical parameters of each performance element in the target dataset;

[0149] The first display module 604 is used to generate and display the distribution data of statistical parameters based on preset attention dimensions; the preset attention dimensions include subject dimension or time dimension.

[0150] In one optional embodiment, the above apparatus further includes: a first data merging module, configured to merge interactive data in the target dataset based on account identifiers when a deduplication rule for the account identifier dimension is configured; and to determine the statistical parameters corresponding to each dimension value of the focus dimension based on the interactive data in the merged target dataset.

[0151] In one optional embodiment, the first data merging module is further configured to merge the interaction data in the target dataset based on the outbound call number and / or service name when no deduplication rule for the account identifier dimension is configured, or when the account identifier dimension of the interaction data in the target dataset is empty.

[0152] In one optional embodiment, the first display module 604 is further configured to determine the outbound number with the closest outbound time as the outbound target number corresponding to the account identifier when the account identifier corresponds to multiple outbound numbers, and display the outbound target number corresponding to each account identifier for each dimension value.

[0153] In one optional embodiment, the performance element includes at least the performance date; the apparatus further includes: an update module for periodically updating the interaction dataset; when the preset intent tag slot value of the target interaction data is converted to a non-preset intent tag slot value, and when the performance date arrives, updating the remaining performance elements and correcting the distribution data based on the updated performance elements.

[0154] In one optional embodiment, the above-mentioned device further includes: a first query module, configured to filter out target interaction data and non-target interaction data corresponding to the account subject to be queried within a preset statistical period based on the current user's data acquisition request for the account subject to be queried; the non-target interaction data includes interaction data that does not conform to the preset intent tag slot value; and display the fulfillment elements of the target interaction data, and / or the fulfillment elements of the non-target interaction data.

[0155] In one optional embodiment, the query module is further configured to display the performance elements in the interaction data of the most recent interaction time of the account subject to be queried within the preset statistical period when no target interaction data or non-target interaction data corresponding to the account subject to be queried is found within the preset statistical period.

[0156] In one optional embodiment, the apparatus further includes: a first dialogue summary module, configured to add a dialogue summary field to the distributed data based on user configuration information; the user configuration information includes NER slot descriptions and event tags; the dialogue summary field is generated based on the intent tag, the hit NER slot description, the hit NER slot value, and the hit event tag in each interaction data in the target dataset.

[0157] In one exemplary embodiment, such as Figure 7 As shown, an intelligent dialogue robot interaction device is provided, including: a request acquisition module 701, a second data acquisition module 702, a second filtering module 703, a second statistical parameter determination module 704, and a second display module 705, wherein:

[0158] The request acquisition module 701 is used to acquire the distribution data display request configured by the current user; the distribution data display request includes the preset statistical period, the performance elements to be displayed, the preset intent tag slot values, the statistical parameters to be displayed, and the preset attention dimensions;

[0159] The second data acquisition module 702 is used to filter the interaction dataset corresponding to the current user from the interaction database based on a statistical period; each piece of interaction data in the interaction dataset includes at least one intent tag and at least one fulfillment element.

[0160] The second filtering module 703 is used to determine the target dataset from the interaction dataset based on the preset intent tag slot value; the target dataset includes all interaction data under the account subject corresponding to the target interaction data; the target interaction data includes interaction data that meets the preset intent tag slot value.

[0161] The second statistical parameter determination module 704 is used to calculate the statistical parameters of each performance element in the target dataset;

[0162] The second display module 705 is used to generate and display the distribution data of statistical parameters based on preset attention dimensions; the preset attention dimensions include subject dimension or time dimension.

[0163] In one optional embodiment, the above apparatus further includes: a second data merging module, configured to obtain the deduplication rule configured by the current user; when the deduplication rule is an account identifier dimension deduplication rule, merge the interaction data in the target dataset based on the account identifier; and determine the statistical parameter corresponding to each dimension value of the focus dimension based on the interaction data in the merged target dataset.

[0164] In one optional embodiment, the above-mentioned device further includes: a second query module, configured to obtain a data acquisition request configured by the current user, the data acquisition request including the account subject to be queried; based on the data acquisition request, filter out target interaction data and non-target interaction data corresponding to the account subject to be queried within a preset statistical period; the non-target interaction data includes interaction data that does not conform to the preset intent tag slot value; display the performance elements of the target interaction data, and / or the performance elements in the non-target interaction data.

[0165] In one optional embodiment, the apparatus further includes: a second dialogue summary module, configured to acquire user configuration information; the user configuration information includes NER slot descriptions and event tags; a dialogue summary field is displayed in the distribution data based on the user configuration information; the dialogue summary field is generated based on the intent tag, the hit NER slot description, the hit NER slot value, and the hit event tag in each interaction data in the target dataset.

[0166] The aforementioned interactive data processing device for intelligent chatbots and the various modules within the intelligent chatbot interaction device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0167] In one exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 8As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores interactive data. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network. When the computer program is executed by the processor, it implements an interactive data processing method and an interactive method for intelligent chatbots.

[0168] Those skilled in the art will understand that Figure 8 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0169] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0170] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0171] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0172] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0173] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0174] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0175] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. An interaction data processing method applied to an intelligent dialogue robot, characterized in that, The method comprises: Screening an interaction data set corresponding to a current user from an interaction database based on a preset statistical period; each piece of interaction data in the interaction data set comprises at least one intent label and at least one performance element; Determining a target data set from the interaction data set based on a preset intent label slot value; the target data set comprises all interaction data under a target account subject corresponding to target interaction data; the target interaction data comprises the interaction data meeting the preset intent label slot value; Calculating statistical parameters of each performance element in the target data set; Generating and displaying distribution data of the statistical parameters based on a preset attention dimension; the preset attention dimension comprises a subject dimension or a time dimension; After determining the target data set from the interaction data set, before generating and displaying the distribution data of the statistical parameters based on the preset attention dimension, the method further comprises: In the case where the account identifier dimension deduplication rule is configured, merging the target interaction data in the target data set based on the account identifier; Based on the target interaction data in the merged target data set, determining the statistical parameters corresponding to each dimension value of the attention dimension; After determining the statistical parameters corresponding to each dimension value of the attention dimension, the method further comprises: In the case where the account identifier corresponds to multiple outbound call numbers, determining the outbound call number with the most recent outbound call time as the outbound target number corresponding to the account identifier, and displaying the outbound target number corresponding to each account identifier corresponding to each dimension value; The performance element at least comprises a performance date; the method further comprises: Periodically updating the interaction data set; In the case where the preset intent label slot value of the target interaction data is converted into a non-preset intent label slot value, updating the remaining performance elements in the case where the performance date arrives, and correcting the distribution data based on the updated performance elements; Based on the data acquisition request of the current user for a to-be-queried account subject, screening the target interaction data and non-target interaction data corresponding to the to-be-queried account subject in the preset statistical period; the non-target interaction data comprises the interaction data not meeting the preset intent label slot value; Displaying the performance elements of the target interaction data and / or the performance elements of the non-target interaction data; Based on user configuration information, adding a conversation summary field in the distribution data; the user configuration information comprises NER slot description and event label; The conversation summary field is generated based on the intent label, the hit NER slot description, the hit NER slot value and the hit event label in each piece of interaction data in the target data set.

2. The method of claim 1, wherein, Further comprising: In the case where the account identifier dimension deduplication rule is not configured, or the account identifier dimension of the target interaction data in the target data set is empty, merging the target interaction data in the target data set based on the outbound call number and / or the business name.

3. The method of claim 1, wherein, The method further comprises: When the target interaction data or non-target interaction data corresponding to the account subject to be queried is not screened out in the preset statistical period, display the performance elements in the interaction data of the latest interaction time corresponding to the account subject to be queried in the preset statistical period.

4. An intelligent conversational robot interaction method, characterized by, The method comprises: obtaining a distribution data display request configured by a current user; the distribution data display request comprises a preset statistical period, a performance element to be displayed, a preset intent label slot value, a statistical parameter to be displayed, and a preset attention dimension; Based on the statistical period, screen the interaction data set corresponding to the current user from the interaction database; each piece of interaction data in the interaction data set comprises at least one intent label and at least one performance element; Based on the preset intent label slot value, determine a target data set from the interaction data set; the target data set comprises all interaction data under the account subject corresponding to the target interaction data; the target interaction data comprises the interaction data meeting the preset intent label slot value; Calculate the statistical parameter of each performance element in the target data set; Generate and display the distribution data of the statistical parameter based on the preset attention dimension; the preset attention dimension comprises a subject dimension or a time dimension; Obtain a deduplication rule configured by the current user; When the deduplication rule is an account identifier dimension deduplication rule, merge the interaction data in the target data set based on the account identifier; Based on the merged interaction data in the target data set, determine the statistical parameter corresponding to each dimension value of the attention dimension; In the case that the account identifier corresponds to multiple outbound call numbers, determine the outbound call number with the latest outbound call time as the outbound target number corresponding to the account identifier, and display the outbound target number corresponding to each account identifier corresponding to each dimension value; The performance elements at least comprise performance dates; the method further comprises: Periodically update the interaction data set; In the case that the preset intent label slot value of the target interaction data is converted into a non-pre-set intent label slot value, update the remaining performance elements when the performance date arrives, and correct the distribution data based on the updated performance elements; obtain a data acquisition request configured by the current user, the data acquisition request comprising an account subject to be queried; Based on the data acquisition request, screen out the target interaction data and non-target interaction data corresponding to the account subject to be queried in the preset statistical period; the non-target interaction data comprises the interaction data not meeting the preset intent label slot value; Display the performance elements of the target interaction data and / or the performance elements in the non-target interaction data; Obtain user configuration information; the user configuration information comprises NER slot description and event label; Based on the user configuration information, display a conversation summary field in the distribution data; the conversation summary field is generated based on the intent label in each piece of interaction data in the target data set, the hit NER slot description, the hit NER slot value, and the hit event label.

5. An interactive data processing apparatus applied to an intelligent dialog robot, characterized by, The device comprises: A first data acquisition module is configured to filter an interaction data set corresponding to a current user from an interaction database based on a preset statistical period; each piece of interaction data in the interaction data set comprises at least one intent label and at least one performance element; A first screening module is configured to determine a target data set from the interaction data set based on a preset intent label slot value; the target data set comprises all the interaction data under a target account subject corresponding to target interaction data; the target interaction data comprises the interaction data meeting the preset intent label slot value; A first statistical parameter determination module is configured to calculate statistical parameters of each performance element in the target data set; A first display module is configured to generate and display distribution data of the statistical parameters based on a preset attention dimension; the preset attention dimension comprises a subject dimension or a time dimension; A first data merging module is configured to merge the target interaction data in the target data set based on an account identifier in a case where an account identifier dimension deduplication rule is configured; and determine the statistical parameters corresponding to each dimension value of the attention dimension based on the target interaction data in the target data set after merging; The first data merging module is further configured to determine an outbound call number with the most recent outbound call time as an outbound target number corresponding to the account identifier in a case where the account identifier corresponds to multiple outbound call numbers, and display the outbound target number corresponding to each account identifier corresponding to each dimension value; The performance element at least comprises a performance date; the device further comprises: A first updating module is configured to periodically update the interaction data set; in a case where the preset intent label slot value of the target interaction data is converted into a non-pre-set intent label slot value, update the remaining performance elements in a case where the performance date arrives, and correct the distribution data based on the updated performance elements; A first query module is configured to filter the target interaction data and non-target interaction data corresponding to a to-be-queried account subject in the preset statistical period based on a data acquisition request of the current user for the to-be-queried account subject; the non-target interaction data comprises the interaction data not meeting the preset intent label slot value; display the performance elements of the target interaction data and / or the performance elements of the non-target interaction data; A first dialogue summary module is configured to add a dialogue summary field in the distribution data based on user configuration information; the user configuration information comprises NER slot description and event label; the dialogue summary field is generated based on the intent label, the hit NER slot description, the hit NER slot value and the hit event label in each piece of interaction data in the target data set.

6. The apparatus of claim 5, wherein, The first data merging module is further configured to merge the target interaction data in the target data set based on an outbound call number and / or a business name in a case where the account identifier dimension deduplication rule is not configured or the account identifier dimension of the target interaction data in the target data set is empty.

7. The apparatus of claim 5, wherein, The query module is further configured to display the performance element in the interaction data of the latest interaction time corresponding to the to-be-queried account principal in the preset statistical period when the target interaction data or non-target interaction data corresponding to the to-be-queried account principal is not screened out in the preset statistical period.

8. An intelligent conversational robot interaction apparatus, characterized by, The device comprises: The request acquisition module is configured to acquire a distribution data display request configured by a current user; the distribution data display request comprises a preset statistical period, a performance element to be displayed, a preset intent label slot value, a statistical parameter to be displayed, and a preset attention dimension; The second data acquisition module is configured to screen an interaction data set corresponding to the current user from an interaction database based on the statistical period; each piece of interaction data in the interaction data set comprises at least one intent label and at least one performance element; The second screening module is configured to determine a target data set from the interaction data set based on the preset intent label slot value; the target data set comprises all the interaction data under an account principal corresponding to target interaction data; the target interaction data comprises the interaction data meeting the preset intent label slot value; The second statistical parameter determination module is configured to calculate the statistical parameter of each performance element in the target data set; The second display module is configured to generate and display distribution data of the statistical parameter based on the preset attention dimension; the preset attention dimension comprises a principal dimension or a time dimension; The second data merging module is configured to acquire a deduplication rule configured by the current user; when the deduplication rule is an account identifier dimension deduplication rule, the interaction data in the target data set is merged based on an account identifier; the statistical parameter corresponding to each dimension value of the attention dimension is determined based on the interaction data in the target data set after merging; when the account identifier corresponds to multiple outbound call numbers, an outbound call number with the latest outbound call time is determined as an outbound target number corresponding to the account identifier, and the outbound target number corresponding to each account identifier corresponding to each dimension value is displayed; The performance element at least comprises a performance date; The device further comprises: The second update module is configured to periodically update the interaction data set; when the preset intent label slot value of the target interaction data is converted into a non-pre-set intent label slot value, the remaining performance elements are updated when the performance date arrives, and the distribution data is corrected based on the updated performance elements; The second query module is configured to acquire a data acquisition request configured by the current user, the data acquisition request comprising a to-be-queried account principal; based on the data acquisition request, the target interaction data and non-target interaction data corresponding to the to-be-queried account principal in the preset statistical period are screened out; the non-target interaction data comprises the interaction data not meeting the preset intent label slot value; the performance element of the target interaction data and / or the performance element in the non-target interaction data is displayed. A second dialogue summary module is configured to obtain user configuration information, wherein the user configuration information comprises NER slot descriptions and event labels; and generate a dialogue summary field in the display of the distribution data based on the user configuration information, wherein the dialogue summary field is generated based on the intent labels, the hit NER slot descriptions, the hit NER slot values, and the hit event labels in each piece of the interaction data in the target data set. 9.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-8 when the computer program is executed by the processor. The computer program is executed by the processor to implement the steps of the method of any one of claims 1 to 4.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the method of any one of claims 1 to 4.

11. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the method of any one of claims 1 to 4. The computer program is executed by the processor to implement the steps of the method of any one of claims 1 to 4.

Images