CN116150329A

CN116150329A - Response method, response device, electronic equipment and storage medium

Info

Publication number: CN116150329A
Application number: CN202211150844.5A
Authority: CN
Inventors: 冉瑞房; 蒋宁; 吴海英; 罗仕杰; 赵飞
Original assignee: Mashang Xiaofei Finance Co Ltd
Current assignee: Mashang Xiaofei Finance Co Ltd
Priority date: 2022-09-21
Filing date: 2022-09-21
Publication date: 2023-05-23

Abstract

The embodiment of the specification provides a response method, a device, an electronic device and a storage medium, wherein the response method comprises the following steps: acquiring a response access request of a target user; the response access request comprises a first request identifier and a second request identifier; the first request identification includes at least one of a dialogue identification and a target user identification of the target user; determining risk judgment data corresponding to the response access request according to at least one of the dialogue identifier and the target user identifier; determining the risk user type of the target user according to the risk judgment data; placing a second request identifier for responding to the access request in a target queuing queue corresponding to the risk user type; each second request identifier in the plurality of second request identifiers included in the target queuing queue is sequentially arranged, and the arrangement sequence of the second request identifiers is used for representing the expected response sequence of the pending access request corresponding to the second request identifier. Therefore, the satisfaction degree of the user in the process of waiting for accessing the manual seat is improved.

Description

Response method, response device, electronic equipment and storage medium

Technical Field

The present disclosure relates to the field of artificial intelligence technologies, and in particular, to a response method, a device, an electronic device, and a storage medium.

Background

With the development of electronic technology, the application of artificial intelligence is becoming more and more popular. By adopting the artificial intelligence technology, consultation service, complaint acceptance service and other interactive services are provided for the user, so that the manpower can be greatly saved, and the problem can be effectively solved for the user. However, intelligent agents that employ artificial intelligence techniques have not yet completely replaced artificial agents, and users tend to switch from intelligent agents to artificial agents when the intelligent agents are unable to solve the user's problem.

If a plurality of users have the requirement of transferring the artificial agents, but the quantity and the reception capacity of the artificial agents are limited, the agent service system needs to queue the users and sequentially distributes the artificial agents for the users according to the queuing sequence. However, the emergency degree of the service provided by the manual agents is different from user to user, so that one part of users may have higher tolerance to longer queuing time, and the other part of users may be more likely to be discontented due to the overlong queuing time. If the agent service system queues each user only based on the time when the user sends out the request for accessing the artificial agent, the waiting time of some special users with urgent needs is possibly too long, and the satisfaction degree of the users waiting for accessing the artificial agent is reduced.

Disclosure of Invention

The embodiment of the application provides a response method, a response device, electronic equipment and a storage medium, so as to improve satisfaction of users waiting for accessing to a manual seat.

In a first aspect, an embodiment of the present application provides a response method, including:

acquiring a response access request of a target user; the response access request carries a first request identifier and a second request identifier; the first request identification includes at least one of a dialogue identification and a target user identification of the target user; the second request identifier is used for uniquely identifying the response access request;

determining risk judgment data corresponding to the response access request according to at least one of the dialogue identifier and the target user identifier;

determining the risk user type of the target user according to the risk judgment data;

placing a second request identifier of the response access request in a target queuing queue corresponding to the risk user type; each second request identifier in the plurality of second request identifiers included in the target queuing queue is sequentially arranged, and the arrangement sequence of the second request identifiers is used for representing the expected response sequence of the to-be-processed access request corresponding to the second request identifier.

In a second aspect, an embodiment of the present application provides a response device, including:

the acquisition unit is used for acquiring the response access request of the target user; the response access request carries a first request identifier and a second request identifier; the first request identification includes at least one of a dialogue identification and a target user identification of the target user; the second request identifier is used for uniquely identifying the response access request;

a first determining unit, configured to determine risk judgment data corresponding to the response access request according to at least one of the session identifier and the target user identifier;

the second determining unit is used for determining the risk user type of the target user according to the risk judging data;

the placing unit is used for placing the second request identifier of the response access request in a target queuing queue corresponding to the risk user type; each second request identifier in the plurality of second request identifiers included in the target queuing queue is sequentially arranged, and the arrangement sequence of the second request identifiers is used for representing the expected response sequence of the to-be-processed access request corresponding to the second request identifier.

In a third aspect, an embodiment of the present application provides an electronic device, including: a processor; and a memory configured to store computer-executable instructions that, when executed, cause the processor to perform the answering method as described in the first aspect.

In a fourth aspect, embodiments of the present application provide a computer-readable storage medium storing computer-executable instructions that, when executed by a processor, implement the answering method according to the first aspect.

It can be seen that in the embodiment of the present application, first, a response access request of a target user is obtained; the response access request carries a first request identifier and a second request identifier; the first request identification includes at least one of a dialogue identification and a target user identification of the target user; the second request identifier is used for uniquely identifying and responding to the access request; secondly, determining risk judgment data corresponding to the response access request according to at least one of the dialogue identifier and the target user identifier; then, determining the risk user type of the target user according to the risk judgment data; finally, a second request mark for answering the access request is placed in a target queuing queue corresponding to the risk user type; each second request identifier in the plurality of second request identifiers included in the target queuing queue is sequentially arranged, and the arrangement sequence of the second request identifiers is used for representing the expected response sequence of the pending access request corresponding to the second request identifier. In this way, through at least one of the dialogue identifier and the target user identifier carried in the response access request of the target user, the risk judgment data corresponding to the response access request can be determined, and further, the risk user types of the target user can be determined based on the risk judgment data, wherein each risk user type corresponds to a different queuing queue respectively, then by placing the second request identifier carried in the response access request of the target user in the target queuing queue corresponding to the risk user type, the target users of different risk user types can be pre-classified, the number of to-be-processed access requests corresponding to each queuing queue is indirectly reduced, the waiting time of the response access request is shortened, and the satisfaction degree of the user waiting for access to the manual seat is improved.

Drawings

For a clearer description of embodiments of the present application or of the solutions of the prior art, the drawings that are required to be used in the description of the embodiments or of the prior art will be briefly described, it being obvious that the drawings in the description below are only some of the embodiments described in the present specification, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art;

fig. 1 is a process flow diagram of a response method provided in an embodiment of the present application;

fig. 2 is a frame structure diagram of an agent service system according to an embodiment of the present application;

fig. 3 is a block diagram of an answer method according to an embodiment of the present application;

fig. 4 is a process flow chart of another response method provided in an embodiment of the present application:

FIG. 5 is a process flow diagram of yet another response method provided in an embodiment of the present application;

fig. 6 is a schematic diagram of a response device according to an embodiment of the present application;

fig. 7 is a schematic structural diagram of an electronic device according to an embodiment of the present application.

Detailed Description

In order to better understand the technical solutions in the embodiments of the present application, the following description will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only some embodiments of the present specification, not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments of the present application without making any inventive effort, shall fall within the scope of the present application.

The agents can be classified into artificial agents and intelligent agents, and the artificial agents include, but are not limited to, text agents, video agents, and voice agents. The text seat refers to a seat which mainly provides services for users in a typing chat mode; the video customer service refers to an agent which mainly provides services for users in the form of voice and video; voice customer service refers to an agent that serves a user, primarily in the form of a mobile phone. In order to reduce manpower and material resources, in general, an agent service system preferentially provides intelligent agents for users, namely, robots adopting artificial intelligence technology meet the needs of most customers. However, in a financial scenario or other business scenario, there are some special businesses, and it is difficult to meet the user demand only by the intelligent agent. Under the condition that the intelligent agent is difficult to solve the problem of the user, the user needs to transfer the manual agent.

In the case where there is a need to transfer an artificial agent for a large number of users, the agent service system needs to queue users who have issued requests to access the artificial agent due to the limited number of artificial agents and the limited reception capacity. Typically, the agent service system may queue the expected response order of the individual users to answer access requests based on the time order in which the users issued the answer access requests for the human agents. However, the demands of different users vary with respect to urgency. Some special users with higher emergency demands are sensitive to queuing time, and the satisfaction degree of the special users can be greatly reduced due to overlong waiting time; another part of the common users is tolerant to queuing time. If the special user and the common user are placed together for queuing, the time for waiting for the special user to access the manual seat is excessively long, and the special user is caused to be seriously discontented.

In order to solve the above problems, embodiments of the present application provide a response method.

The answer method provided by the embodiment of the application can be executed by the electronic equipment, and particularly can be executed by a processor in the electronic equipment. The electronic device mentioned herein may be a terminal device such as a smart phone, a tablet computer, a desktop computer, a smart voice interaction device, a wearable device, a robot, a car terminal, etc.; alternatively, the electronic device may also be a server, such as a stand-alone physical server, a server cluster composed of a plurality of servers, or a cloud server capable of cloud computing.

Fig. 1 is a process flow diagram of a response method provided in an embodiment of the present application. Fig. 2 is a frame structure diagram of an agent service system according to an embodiment of the present application.

The response method as shown in fig. 1 may be applied to an agent service system. The structure of the seat service system may refer to fig. 2.

The agent service system may be a system for providing a service to a user through at least one of an intelligent agent and a manual agent, for example, a customer service system. The "agent" appearing in the present specification may be a serviceman who provides a service in TSR (Telephone Service Representative, call center agent). The TSR is typically comprised of a seat computer, seat software, seat headset, attendant, etc. The TSR realizes related control functions through seat software and hardware equipment so as to achieve the purpose of customer service, and belongs to the category of customer service.

The intelligent agent may be a robot capable of providing services to the user, and the artificial agent may be a customer service, a volunteer attendant, or other types of attendant, for example.

As shown in fig. 2, the agent service system includes a presentation layer 201, a service layer 202, a middleware layer 203, and a storage layer 204. The presentation layer may be a user interface presented to the user by the agent service system, and includes an intelligent agent end 2011 and an artificial agent end 2012. The intelligent agent terminal 2011 refers to a first user interface for providing an intelligent agent service to a user, and the manual agent terminal 2012 refers to a second user interface for providing a manual agent service to a user.

The numbers of "first", "second", "third" … … and the like in the embodiments of the present application are merely for convenience of distinguishing features that are different in multiple applications and similar in content, and are not actually meant, and are not described in detail below.

The service layer 202 may include various system components for performing the reply method, including, but not limited to: a response system DM (dialog Manager) 2021, NLP (Natural Language Processing ) intention recognition 2022, a robot knowledge base 2023, a multi-round dialog engine 2025, a manual agent interface 2026, and a Session management service 2027.

Session management service 2027 may be used to manage conversations that occur between intelligent agents/manual agents and users.

The middle tier 203 includes frame components and connection components of the seat service system that can be used for information transfer in the seat service system. The intermediate layer 203 may include: a kafka component 2031, a redis component 2032, and a message bus 2033.

The storage layer may be used to store data required by the agent service system. The storage layer may include: mysq1 database, hbase database, and ES database.

It should be noted that the configuration of the agent service system shown in fig. 2 is merely exemplary, and the agent service system may have relatively large differences due to different configurations or performances.

In the practical application of the response method, the agent service system may be replaced by another system capable of providing the agent service, and in this specification, the application of the response method to the agent service system is mainly described as an example, and when the response method is applied to another system capable of providing the agent service, reference may be made to the corresponding description part of the embodiment of the response method applied to the agent service system.

Next, the response method shown in fig. 1 will be specifically described.

Step S102, obtaining a response access request of a target user; the response access request carries a first request identifier and a second request identifier; the first request identification includes at least one of a dialogue identification and a target user identification of the target user; the second request identity is used to uniquely identify the response access request.

The target user may be a user of an agent service system that has access to an artificial agent requirement. The target user may be a registered user of the agent service system or may be a temporary user of the agent service system. The temporary user may be a user who has not registered personal information in the agent service system and is assigned with a temporary user identifier by the agent service system.

In step S102, the number of target users may be one or a plurality of target users. It should be noted that the number of target users is plural, which means that the response access request of each target user in the plurality of target users is acquired at the same time point. If the times for obtaining the plurality of response access requests are different, the response method shown in the embodiment of fig. 1 may be separately executed for each target user.

The answer access request may be an answer access request transferred from the intelligent agent to the artificial agent, and in the case that the artificial agent responds to the answer access request, the artificial agent responding to the answer access request replaces the intelligent agent to provide service for the target user.

The response access request may be a response access request directly sent to the manual agent, and the manual agent responding to the response access request provides a service for the target user.

The services that the intelligent agent or the manual agent may provide to the target user include, but are not limited to: answer the consulting questions of the target user, provide shopping links, provide preferential activity information, reply the inquiry result of points, recommend services suitable for the requirements of the target user, provide an activity participation interface, pacify the target user with complaint will, and the like. The present specification does not impose particular restrictions on the services provided by the intelligent agent or the manual agent.

The reply access request may carry a first request identification and a second request identification.

The first request identifier may include only the session identifier, may include only the target user identifier of the target user, and may include both the session identifier and the target user identifier.

The dialogue identifier may be a character string identifier or other preset identifiers. The session identifier may be used to uniquely identify a session between the target user and the intelligent agent, which session may occur before step S102 is performed, i.e. before obtaining a response access request from the target user.

The target user identification of the target user may be a user identification for uniquely identifying the target user. The target user identification may be a formal user identification assigned by the agent service system for the registered user. The target user identification may be used to query historical service data of the target user in the agent service system. Historical service data includes, but is not limited to: historical dialogue data stored after the target user dialogues with at least one of the intelligent agent and the artificial agent, and historical response access requests and response records of the historical response access requests sent by the target user, and the like.

The second request identity may be used to uniquely identify the reply access request. The second request identifier may be in the form of a string or other preset form. It should be noted that, for each target user, if the target user sends out a response access request at different time points, the response access request sent out at each time point corresponds to a second request identifier.

The target user may send out a response access request transferred to the artificial agent after performing a dialogue with the intelligent agent, or may send out a response access request to the artificial agent without performing a dialogue with the intelligent agent.

In the case where the target user issues a transition to the manual agent to answer the access request after having first passed the dialogue with the intelligent agent, the first request identifier may include a dialogue identifier of the dialogue with the intelligent agent.

In the case where the target user directly sends a response access request to the human agent without having undergone a dialogue with the intelligent agent, the first request identifier may not include the dialogue identifier.

In the case where the target user is a registered user of the agent service system, the first request identification may include a target user identification of the target user.

In the case that the target user is a temporary user of the agent service system, the user identifier of the temporary user is a temporary user identifier, and cannot be used for querying historical service data in the agent service system, and the first request identifier of the artificial agent may not include the target user identifier of the target user.

The method for acquiring the response access request of the target user can be that acquiring a designated text input by the target user in a text input box, for example, a 'manual seat', or acquiring a click operation of a preset response access request control by the target user, or presetting other modes for triggering the response access request, receiving user input corresponding to the modes for triggering the response access request, triggering the response access request according to the user input, and the like.

Optionally, before obtaining the response access request of the target user, the method further includes: acquiring dialogue data between a target user and an intelligent agent; determining whether at least one target word matched with the sensitive words is included in the dialogue data based on a preset sensitive word set comprising a plurality of sensitive words; if the dialogue data comprises at least one target word, performing data embedding processing based on the target word to obtain embedded data.

In specific implementation, a sensitive word set may be preset in the agent service system, where the sensitive word set may include a plurality of sensitive words, such as complaints and the like. Further, under different service scenarios, the corresponding sensitive word set can be configured according to the actual service.

When the target user and the intelligent agent are in a conversation, conversation data between the target user and the intelligent agent can be acquired, wherein the conversation data comprises chat contents of each sentence of the target user and the intelligent agent.

Based on the sensitive word set, hit judgment of sensitive words can be carried out on each chat content of the target user and the robot, so as to determine whether each chat content comprises at least one target word matched with the sensitive word.

The hit judgment of the sensitive word can comprise the following steps:

(1) And performing word segmentation processing on each sentence of chat content to obtain word segmentation results, wherein the word segmentation results comprise a plurality of words corresponding to the sentence of chat content. For example, a sentence of chat content is: i want to complain about XXX, with his service attitudes being very poor. After the sentence chat content is subjected to word segmentation, the word segmentation result comprises the following words: "I", "want", "complaint", "XXX", "he", "service", "attitude", "pole" and "bad".

(2.1) in one embodiment, for each term in the plurality of terms, a similarity between each sensitive term in the set of sensitive terms and the term may be calculated, and if the similarity is greater than or equal to a preset threshold, it is determined that the term matches the sensitive term.

For example, the sensitive word set includes: A. b, etc. The word segmentation result of the chat content includes a plurality of words: x1, x2, x3, x4, x5, x6. Calculating the similarity between a sensitive word A in a sensitive word set and each word X1, X2, X3, X4, X5 and X6 in the word segmentation result, if the similarity is larger than a preset threshold value, determining that the word A is matched with the sensitive word A, and if the similarity is smaller than or equal to the preset threshold value, determining that the word A is not matched with the sensitive word A; and respectively calculating the similarity between the sensitive word 'B' in the sensitive word set and each word x1, x2, x3, x4, x5 and x6 in the word segmentation result, if the similarity is larger than a preset threshold value, determining that the word is matched with the sensitive word 'B', if the similarity is smaller than or equal to the preset threshold value, determining that the word is not matched with the sensitive word 'B', and the like.

(2.2) in another embodiment, a set of paraphrasing terms for each sensitive term may be preconfigured, and for each term in the plurality of terms, it may be queried whether the term belongs to the set of paraphrasing terms for the respective sensitive term, and if so, it is determined that the term matches the sensitive term.

For example, the sensitive word set includes: "A", "B", etc. A set of paraphraseology for the sensitive word "a", set 1, comprising: al, A2, A3, A4. A set of paraphraseology for the sensitive word "B", set 2, comprising: b1, B2, B3, B4, B5. The word segmentation result of the chat content includes a plurality of words: x1, x2, x3, x4, x5, x6. For x1, inquiring whether the word belongs to the set 1, if so, determining that x1 is matched with the sensitive word A, and if not, determining that x1 is not matched with the sensitive word A; and inquiring whether the word belongs to the set 2, if so, determining that x1 is matched with the sensitive word B, and if not, determining that x1 is not matched with the sensitive word B. x2, x3, x4, x5, x6 are similar to x1, and reference is made to the description of x 1.

And carrying out data embedding processing based on the target words to obtain embedded point data, wherein the number of the target words can be determined as the embedded point data, and sentences to which the target words belong can be determined as the embedded point data. The buried point data may be stored in the db database and redis component.

Step S104, determining risk judgment data corresponding to the response access request according to at least one of the dialogue identifier and the target user identifier.

Each session identification corresponds to a session between a user and the intelligent agent. Each user may have multiple conversations with the intelligent agent. The target user's target user identification may correspond to one or more dialog identifications.

The session identifier included in the response access request may be a session identifier of a last session between the target user and the intelligent agent before issuing the response access request transferred from the intelligent agent to the artificial agent.

Based on the dialogue identification, dialogue content of the last dialogue between the target user and the intelligent agent can be obtained, and further, risk judgment data of the access request can be determined according to the dialogue content.

Based on the target user identification, historical dialogue data, historical burial data and other types of historical data of the target user can be queried, so that risk judgment data corresponding to the access request is determined based on the historical data of each type.

The risk determination data may be a parameter for determining whether the target user is a risk user.

And S106, determining the risk user type of the target user according to the risk judgment data.

The risk user types may include risk users and non-risk users, may include multiple risk levels, and may be other pre-configured risk user types.

Optionally, determining risk judgment data corresponding to the response access request according to the session identifier includes: inquiring buried point data corresponding to the dialogue identifier in the historical buried point data based on the dialogue identifier to obtain first buried point data; reading the number of target words from the first buried point data to obtain a first number, and determining the first number as risk judgment data; determining the risk user type of the target user according to the risk judgment data, including: and if the first number is larger than a first preset number threshold, determining the risk user type of the target user as a first risk type.

Under the condition that the target user transfers the artificial agent from the intelligent agent, risk judgment data corresponding to the response access request can be determined according to the dialogue identifier.

The historical buried point data may include a pre-generated correspondence of session identification to buried point data. The generation mode of the buried point data corresponding to each dialogue identifier in the historical buried point data can refer to the previous text for acquiring dialogue data between a target user and an intelligent agent; determining whether at least one target word matched with the sensitive words is included in the dialogue data based on a preset sensitive word set comprising a plurality of sensitive words; if the dialogue data comprises at least one target word, performing data embedding processing based on the target word to obtain an explanation part of embedded data'.

In a multi-round dialogue flow of the seat service system, the seat service system can judge whether a dialogue corresponding to a dialogue identifier carried by a response access request of a target user hits sensitive words or not, specifically, the accumulated hit times A of the dialogue are inquired out from a redis component or a db database according to the dialogue identifier sessionid, namely buried point data corresponding to the dialogue identifier are inquired, and first buried point data are obtained; and reading the number of the target words from the first buried point data to obtain a first number.

The first preset number threshold may be a sensitive word hit threshold X preconfigured by the agent service system, and if the accumulated hit number a is greater than X, the session is indicated to trigger a sensitive word, so that the risk user type of the target user may be determined as a first risk type.

Optionally, determining risk judgment data corresponding to the response access request according to the target user identifier includes: inquiring buried point data corresponding to the target user identifier in the historical buried point data based on the preset time length and the target user identifier to obtain second buried point data; reading the number of target words from the second buried data to obtain a second number, and determining the second number as risk judgment data; determining the risk user type of the target user according to the risk judgment data, including: and if the second number is larger than a second preset number threshold, determining the risk user type of the target user as a second risk type.

First, counting the times A of hit sensitive words from all dialogues of a target user in the period of inquiring X days in the hit sensitive word buried point data table, wherein X can be used for representing the preset time length. The second preset number threshold may be a history session sensitive word configuration threshold N. And acquiring a history session sensitive word configuration threshold N from the seat service system configuration table, and determining the risk user type of the target user as a second risk type if A is more than N and represents the history session trigger sensitive word.

The second preset number threshold may be a value equal to or greater than the first preset number threshold.

Specifically, the target time range may be determined based on the preset time length and the current time point; based on the target time range and the target user identification, buried point data corresponding to the target user identification can be queried from the historical buried point data, and second buried point data can be obtained.

The number of dialogs corresponding to the target user identification may be one, may not exist, or may be plural within the target time range. The dialogue identification corresponding to the target user identification can be inquired from the history service data, and further, the buried point data corresponding to each dialogue identification can be inquired from the history buried point data under the condition that the number of the dialogue identifications is more than or equal to one. In the case where the number of session identifiers is zero, there is no buried point data corresponding to the target user identifier in the history buried point data, and the second buried point data may be represented by a preset null value, for example, "0".

In another embodiment, the second buried point data may be obtained by querying buried point data corresponding to the target user identifier in the historical buried point data based on the preset time range and the target user identifier; reading the number of target words from the second buried data to obtain a second number, and determining the second number as risk judgment data; determining the risk user type of the target user according to the risk judgment data, including: and if the second number is larger than a second preset number threshold value, determining the risk user type of the target user as a second risk type.

Optionally, determining risk judgment data corresponding to the response access request according to the target user identifier includes: inquiring whether at least one target complaint work order exists in a preset storage space according to the target user identification, obtaining a work order inquiring result, and determining the work order inquiring result as risk judging data; the target complaint worksheet meets the preset worksheet screening conditions; the corresponding relation between the user identification and the complaint work orders is stored in the preset storage space; determining the risk user type of the target user according to the risk judgment data, including: and if the work order query result is used for indicating that at least one target complaint work order exists, determining the risk user type of the target user as a third risk type.

The preset work order screening conditions include, but are not limited to, one or more of customer source screening conditions, status screening conditions, time screening conditions, work order tag screening conditions.

The preset storage space may be a CRM (Customer Relationship Management ) system.

The third risk type may include a plurality of risk subtypes, and the risk subtypes may be determined based on preset work order screening conditions. Several risk subtypes of the third risk type are listed below in connection with the examples.

Under the condition that the preset work order screening conditions comprise customer source screening conditions and state screening conditions, the current day work order data in the CRM system can be queried through target user identification, if the state of the work order data is one of processing, disconnection, suspension, distribution and junction follow-up, and the customer source is one of a network platform and media, the fact that the target user has a complaint work order which is not processed yet in the current day is indicated, and the risk user type of the target user can be determined to be the first subtype in the third risk type.

Under the condition that the preset worksheet screening conditions comprise time screening conditions, customer source screening conditions and state screening conditions, acquiring worksheets of the last X days (service personnel are configured to a system service configuration table) from the CRM system according to the target user identification, if one of a network platform and a media exists as a customer source and the worksheet of the state is a statement, indicating that the target user has a processed historical complaint worksheet, and determining the risk user type of the target user as a second subtype in a third risk type.

Under the condition that the preset work order screening conditions comprise work order label screening conditions and state screening conditions, acquiring the current day work order data from the CRM system according to the target user identification, and if the existence state is one of processing, disconnection, suspension, to-be-allocated and handling and to-be-followed, and the work order label is the work order data of one of consultation class, 1-level complaint and 2-level complaint, indicating that the target user has the complaint work order which is not processed yet in the current day, determining the risk user type of the target user as a third subtype in the third risk type.

Under the condition that the preset worksheet screening conditions comprise time screening conditions, worksheet label screening conditions and state screening conditions, acquiring the latest X-day worksheet data from the CRM system according to the target user identification, and if the existence state is one of processing, disconnecting, suspending, distributing and handling, and the worksheet label is worksheet data of one of consultation class, 1-level complaint and 2-level complaint, indicating that the target user has a history complaint worksheet which is not processed yet, determining the risk user type of the target user as a fourth subtype in the third risk type. The first, second, third, and fourth subtypes described above may be four different types of risk subtypes that are assigned to the third risk type.

Optionally, after determining whether the dialogue data includes at least one target word matching the sensitive word based on the preconfigured sensitive word set including the plurality of sensitive words, the method further includes: if the number of the target words included in the dialogue data is larger than a third preset number threshold, triggering a manual seat access request; if the artificial seat is successfully accessed through the artificial seat access request, generating a target behavior buried point record; according to the target user identification, determining risk judgment data corresponding to the access request, including: inquiring the number of target behavior buried point records corresponding to the target user identification from the historical buried point data according to the target user identification to obtain a third number, and determining the third number as risk judgment data; determining the risk user type of the target user according to the risk judgment data, including: and if the third quantity is larger than the fourth quantity threshold value, determining the risk user type of the target user as a fourth risk type.

After the target user and the intelligent agent have a dialogue, the automatic transfer agent is triggered possibly because the dialogue comprises a plurality of sensitive words, and after each time of the transfer agent is triggered successfully by the user, the transfer agent behavior data are buried in the redis component and the db database.

When the multi-round dialogue flow is manually judged, the manual conversion times A of the target user can be obtained from redis or db based on the target user identification. The fourth preset number threshold may be the number of times X of manual transfer configured by the agent service system configuration table, and if a > X, it indicates that the target user has repeated manual transfer before, and the risk user type of the target user is determined as the fourth risk type.

Optionally, determining risk judgment data corresponding to the response access request according to the target user identifier includes: inquiring the number of dialogue data corresponding to the target user identifier from the historical dialogue data according to the target user identifier and a preset time range to obtain a fourth number; and inquiring the number of the interactive voice response IVR (Interactive Voice Response, i.e. interactive voice response) records corresponding to the target user identification from the historical call records according to the target user identification and the preset time range to obtain a fifth number; determining the fourth number and the fifth number as risk judgment data; determining the risk user type of the target user according to the risk judgment data, including: if the fourth number is greater than the fifth number threshold and the fifth number is greater than the sixth number threshold, determining the risk user type of the target user as a fifth risk type.

The risk user type of the target user being a fifth risk type may be used to indicate that the target user was repeatedly incoming. Repeated incoming lines can be divided into two cases: 1. repeating incoming wires of the intelligent seat; 2. IVR voice repeats incoming.

The definition of the intelligent agent repeated incoming line is as follows: and the target user enters a multi-round dialogue in the timeA, counts the number of times C1 of accessing the intelligent agent in the current day time range before the timeA, acquires the incoming line number threshold C2 configured by service personnel from the service configuration table, and if C1 is more than C2, indicates that the intelligent agent repeatedly enters the line.

The definition of IVR voice repeat incoming line is as follows: and the target user enters a plurality of rounds of conversations at the timeA, counts the number A1 of active dialing of the seat telephone by the target user in the time of the day before the timeA moment, and simultaneously acquires the threshold value B1 of dialing the seat telephone from the system service configuration table, and if A1 is more than B1, the IVR voice repeated incoming line is indicated.

If the intelligent agent repeated incoming line and the IVR voice repeated incoming line are repeated simultaneously, the fact that the target user meets the preset repeated incoming line condition on the same day is indicated, and then the risk user type of the target user can be determined to be a fifth risk type.

In another embodiment, whether the target user belongs to a member of the agent service system may be further queried according to the target user identifier, if so, the member class may be further queried, and if the target user belongs to the member, the risk user type of the target user may be determined as the sixth risk type. Further, the subtype corresponding to the target user in the sixth risk type can be determined according to the member level, so that different levels of queuing services can be provided for members of different levels.

It should be noted that, the above embodiments for determining the risk user type of the target user according to the risk determination data may be performed simultaneously, or may be performed sequentially according to a preset sequence, or may be performed in one or more embodiments, or may be performed in all embodiments.

Under the condition that multiple embodiments for determining the risk user types of the target user are executed at the same time, if the number of the determined risk user types of the target user is greater than one, the risk user type with the highest priority can be determined as the target risk user type of the target user according to the pre-configured priority order, and then the target queuing queue corresponding to the target risk user type can be determined in the subsequent step S108.

Illustratively, multiple embodiments of determining the risk user type of the target user may be performed simultaneously at point in time T1 and the process flow and results for each embodiment are as follows:

(a1) Inquiring buried point data corresponding to the dialogue identifier in the historical buried point data based on the dialogue identifier to obtain first buried point data; and reading the number of the target words from the first buried point data to obtain a first number, wherein if the first number is smaller than a first preset number threshold value, the risk user type of the target user can be determined to be not the first risk type.

(a2) Inquiring buried point data corresponding to the target user identifier in the historical buried point data based on the preset time length and the target user identifier to obtain second buried point data; reading the number of target words from the second buried data to obtain a second number; and if the second number is larger than a second preset number threshold, determining the risk user type of the target user as a second risk type.

(a3) Inquiring the number of dialogue data corresponding to the target user identifier from the historical dialogue data according to the target user identifier and a preset time range to obtain a fourth number; inquiring the number of Interactive Voice Response (IVR) records corresponding to the target user identification from the historical call records according to the target user identification and a preset time range to obtain a fifth number; and if the fourth number is greater than the fifth number threshold and the fifth number is greater than the sixth number threshold, determining the risk user type of the target user as a fifth risk type.

(a4) And inquiring whether the target user belongs to a member of the seat service system according to the target user identification to obtain an inquiry result. And determining that the target user belongs to the member according to the query result, and determining the risk user type of the target user as a sixth risk type.

In summary, it may be determined that the risk user types of the target user include a second risk type, a fifth risk type, and a sixth risk type. The priority order of the first risk user type r1, the second risk user type r2, the fifth risk user type r5 and the sixth risk user type r6 is pre-configured from high to low: r1, r2, r5, r6. It may be determined that the final risk user type of the target user is the second risk type in the order of priority. Further, a target queuing queue corresponding to the second risk type may be determined in a subsequent step S108.

Under the condition that multiple embodiments for determining the risk user type of the target user are sequentially executed according to the preset sequence, the embodiment for determining the risk user type of the target user according to the risk judgment data, which is not executed, can be skipped after one risk user type is obtained, the currently determined risk user type is determined as the target risk user type of the target user, and then a target queuing queue corresponding to the target risk user type can be determined in the subsequent step S108.

For example, the preset order may be used to represent whether the risk user type of the target user is sequentially determined to be the first risk type, the second risk type, the fifth risk type, and the sixth risk type. The various embodiments for determining the risk user type of the target user are sequentially executed according to the preset sequence, and the processing flow and the result of each embodiment are as follows:

(b1) Firstly, inquiring buried point data corresponding to a dialogue identifier in historical buried point data based on the dialogue identifier to obtain first buried point data; and reading the number of the target words from the first buried point data to obtain a first number, wherein if the first number is smaller than a first preset number threshold value, the risk user type of the target user can be determined to be not the first risk type.

(b2) Secondly, inquiring buried point data corresponding to the target user identification in the historical buried point data based on the preset time length and the target user identification to obtain second buried point data; reading the number of target words from the second buried data to obtain a second number; if the second number is less than the second preset number threshold, it may be determined that the risk user type of the target user is not the second risk type.

(b3) Then, according to the target user identification and the preset time range, inquiring the number of dialogue data corresponding to the target user identification from the historical dialogue data to obtain a fourth number; inquiring the number of Interactive Voice Response (IVR) records corresponding to the target user identification from the historical call records according to the target user identification and a preset time range to obtain a fifth number; and if the fourth number is greater than the fifth number threshold and the fifth number is greater than the sixth number threshold, determining the risk user type of the target user as a fifth risk type.

After the risk user type of the target user is determined to be the fifth risk type, since one risk user type is already obtained, the query result can be obtained by skipping the query whether the target user belongs to a member of the agent service system according to the target user identification, which is not yet executed. And determining whether the target user belongs to a member according to the query result, and if so, determining the risk user type of the target user as a sixth risk type. The currently determined risk user type is a fifth risk type, and the fifth risk type is determined as the target risk user type of the target user, so that a target queuing queue corresponding to the fifth risk type may be determined in the subsequent step S108.

Step S108, a second request mark for responding to the access request is placed in a target queuing queue corresponding to the risk user type; each second request identifier in the plurality of second request identifiers included in the target queuing queue is sequentially arranged, and the arrangement sequence of the second request identifiers is used for representing the expected response sequence of the pending access request corresponding to the second request identifier.

The correspondence between risk user types and queuing queues may be one-to-one, one-to-many, many-to-one, or many-to-many. For example, each risk user type may correspond to one queuing queue, or may correspond to K queuing queues. Alternatively, each queuing queue may correspond to L risk user types.

Under the condition that each risk user type corresponds to a plurality of queuing queues, a plurality of queuing queues corresponding to the risk user type of the target user are obtained through inquiry, then the queue length of each queuing queue can be inquired, and the queuing queue with the shortest queue length is determined to be the target queuing queue.

In another embodiment, in the case that each risk user type corresponds to a plurality of queuing queues, based on the risk user type of the target user, the corresponding plurality of queuing queues are obtained by first querying, and then, the queuing queue with the highest queuing priority order may be determined as the target queuing queue according to the preset queuing priority order of the plurality of queuing queues.

Based on the correspondence between the pre-configured risk user type and the queuing, the target queuing may be obtained according to the risk user type query of the target user determined in step S106, and in the case of executing multiple embodiments for determining the risk user type of the target user at the same time, the target queuing may also be obtained according to the final risk user type query of the target user determined in step S106.

In the case where the risk user type includes a risk user and a non-risk user, a correspondence between the risk tag and the first queuing queue and a correspondence between the risk-free tag and the second queuing queue may be preconfigured. If the risk user type of the target user is determined to be the risk user according to the risk judgment data, a risk tag may be added to the target user, so that the first queuing queue corresponding to the risk tag is determined to be the target queuing queue corresponding to the risk user type when step S108 is executed. If the risk user type of the target user is determined to be the risk-free user according to the risk judgment data, a risk-free label can be added to the target user, so that a second queuing queue corresponding to the risk-free label is determined to be a target queuing queue corresponding to the risk user type in a subsequent step.

In the case that the risk user type includes a risk user and a non-risk user, a corresponding relationship between the carried risk tag and the first queuing queue and a corresponding relationship between the non-carried risk tag and the second queuing queue may also be preconfigured.

The first and second queues may be two different queues. The first queuing queue may include a plurality of second request identifiers, each of the second request identifiers being sequentially arranged, the arrangement order of the second request identifiers being used to characterize a response order of the pending access requests corresponding to the second request identifiers. The second queuing queue may include a plurality of second request identifiers, each of the second request identifiers being sequentially arranged, the arrangement order of the second request identifiers being used to characterize a response order of the pending access requests corresponding to the second request identifiers.

Embodiments in which the risk user type includes multiple risk levels, and embodiments in which the risk user type includes other pre-configured risk user types, reference may be made to corresponding description portions in which the risk user type includes both risk users and non-risk users.

The second request identifier of the response access request is placed in the target queuing queue corresponding to the risk user type, and the second request identifier of the response access request may be inserted into the middle of the target queuing queue, or may be inserted after the last second request identifier of the target queuing queue.

Typically, the target queuing queue may include at least one second request identification therein. Each second request identifier in the at least one second request identifier may be a second request identifier carried by the response access request of a user other than the target user before performing step S108. The process flow of placing the second request identifier carried by the response access request of the other user into the target queuing queue may refer to steps S102-S108 of the response method provided in the embodiment of fig. 1.

In order to facilitate distinguishing the second request identifier included in the target queuing queue from the second request identifier carried by the response access request of the target user before executing step S108, the second request identifier carried by the response access request of the target user may also be referred to as a target request identifier.

The second request identifier for answering the access request is placed in the target queuing queue corresponding to the risk user type, and the target request identifier may be placed in the target queuing queue corresponding to the risk user type.

Before step S108 is performed, each of the plurality of second request identifiers included in the target queue is sequentially arranged, and the arrangement order of the second request identifiers is used to characterize the expected response order of the pending access requests corresponding to the second request identifiers. The pending access request corresponding to the second request identifier may be a response access request of a user other than the target user.

After step S108 is performed, the plurality of second request identifiers included in the target queue include both the second request identifier carried by the response access request of the other user and the target request identifier. Each second request identifier is sequentially arranged, and the arrangement sequence of the second request identifiers is used for representing the expected response sequence of the pending access request corresponding to the second request identifier. The to-be-processed access request corresponding to the second request identifier may be a response access request of a user other than the target user, or may be a response access request of the target user.

Optionally, placing the response access request in a target queuing queue corresponding to the risk user type, including: placing a second request identifier for answering the access request in the target queuing queue, so that the second request identifier for answering the access request is positioned after the last second request identifier included in the target queuing queue; the target queuing queue comprises a plurality of second request identifications which are sequentially arranged according to the sequence of the placement time; each second request identifier in the plurality of second request identifiers corresponds to a to-be-processed access request of a to-be-accessed target artificial agent; the target artificial agent is an artificial agent corresponding to the type of the risk user.

In particular, each queuing queue may correspond to one risk user type or may correspond to multiple risk user types. The target queuing queue may correspond to one risk user type or may correspond to multiple risk user types. In addition, one risk user type may also correspond to multiple queuing queues.

In the case that each risk user type corresponds to a plurality of queuing queues, a target queuing queue to be placed with a second request identification of a response access request may be determined based on the priority order of the plurality of queuing queues and the generation time of the response access request, and the second request identification of the response access request may be placed in the target queuing queue.

Each of the plurality of second request identifiers corresponds to a pending access request waiting for accessing the target manual agent, and the pending access request may be a response access request of a user other than the target user. The second request identifier of the pending access request is placed in the target queuing queue before the execution of step S108, and the specific implementation of the method from the step S102 to the step S108 of obtaining the pending access request to the step S108 of placing the second request identifier of the pending access request in the target queuing queue may refer to the answer method provided in the embodiment of fig. 1.

For each risk user type, the seat number of the artificial seat corresponding to the risk user type can be configured, so that targeted artificial services are provided for different risk user types, and the satisfaction degree of the target user is improved.

Optionally, placing the response access request in a target queuing queue corresponding to the risk user type, including: determining the insertion position of a second request identifier for responding to the access request in the target queuing queue according to the risk judging data; based on the insertion location, a second request identification that answers the access request is placed in the target queuing queue.

The determining the insertion position of the access request in the target queuing queue according to the risk judgment data may be that, when a plurality of response access requests are acquired at the same time, the risk judgment data corresponding to each response access request is ordered, and the insertion position of each response access request in the target queuing queue is determined based on the ordering result. Based on the insertion location, the access request is placed in a target queuing queue.

For example, the response access request 1 of the target user 1, the response access request 2 of the target user 2, and the response access request 3 of the target user 3 are acquired at the same time. Next, the following steps are simultaneously performed:

(c1) Inquiring buried point data corresponding to the dialogue identifier in the history buried point data based on the dialogue identifier carried by the response access request 1 to obtain first buried point data; the number of target words is read from the first buried point data, the first number corresponding to the response access request 1 is obtained, and the first number corresponding to the response access request 1 is determined to be risk judgment data of the response access request 1.

(c2) Inquiring buried point data corresponding to the dialogue identifier in the history buried point data based on the dialogue identifier carried by the response access request 2 to obtain first buried point data; the number of target words is read from the first buried point data, the first number corresponding to the response access request 2 is obtained, and the first number corresponding to the response access request 2 is determined to be risk judgment data of the response access request 2.

(c3) Inquiring buried point data corresponding to the dialogue identifier in the history buried point data based on the dialogue identifier carried by the response access request 3 to obtain first buried point data; the number of target words is read from the first buried point data, the first number corresponding to the response access request 3 is obtained, and the first number corresponding to the response access request 1 is determined to be risk judgment data of the response access request 3.

Then, the first number corresponding to the response access request 1, the first number corresponding to the response access request 2 and the first number corresponding to the response access request 3 are ranked, and a ranking result is obtained: the first number corresponding to the response access request 1 is the largest, the first number corresponding to the response access request 3 is centered, and the first number corresponding to the response access request 2 is the smallest.

The target queue already includes 5 second request identifications, the insertion position of the response access request 1 in the target queue may be determined as a sixth bit, the insertion position of the response access request 3 in the target queue may be determined as a seventh bit, and the insertion position of the response access request 2 in the target queue may be determined as an eighth bit.

In another embodiment, the insertion position of the second request identifier for responding to the access request in the target queuing queue may be determined according to a comparison result between the risk judging data and a preset threshold; based on the insertion location, a second request identification that answers the access request is placed in the target queuing queue.

For example, the target queuing queue already includes 6 second request identifications. The risk judging data is the third number, that is, the number of target behavior buried point records corresponding to the target user identifier, and if the third number belongs to the preset threshold interval [ n1, n2], the insertion position of the second request identifier for responding to the access request in the target queuing queue can be determined as the middle position in the target queuing queue, that is, the fourth position in the target queuing queue. Based on the insertion position, the second request identification of the response access request is placed in the target queuing queue, so that the second request identification of the response access request is located in the fourth bit in the target queuing queue, and three second request identifications are arranged before the second request identification of the response access request, and three second request identifications are arranged after the second request identification. Among the 6 second request identifiers originally included in the target queuing queue, the last three second request identifiers are respectively shifted one bit after the step of 'placing the second request identifier responding to the access request in the target queuing queue based on the insertion position'.

In the embodiment of the response method shown in fig. 1, first, a response access request of a target user is obtained; the response access request carries a first request identifier and a second request identifier; the first request identification includes at least one of a dialogue identification and a target user identification of the target user; the second request identifier is used for uniquely identifying and responding to the access request; secondly, determining risk judgment data corresponding to the response access request according to at least one of the dialogue identifier and the target user identifier; then, determining the risk user type of the target user according to the risk judgment data; finally, a second request mark for answering the access request is placed in a target queuing queue corresponding to the risk user type; each second request identifier in the plurality of second request identifiers included in the target queuing queue is sequentially arranged, and the arrangement sequence of the second request identifiers is used for representing the expected response sequence of the pending access request corresponding to the second request identifier. In this way, through at least one of the dialogue identifier and the target user identifier carried in the response access request of the target user, the risk judgment data corresponding to the response access request can be determined, and further, the risk user types of the target user can be determined based on the risk judgment data, wherein each risk user type corresponds to a different queuing queue respectively, then by placing the second request identifier carried in the response access request of the target user in the target queuing queue corresponding to the risk user type, the target users of different risk user types can be pre-classified, the number of to-be-processed access requests corresponding to each queuing queue is indirectly reduced, the waiting time of the response access request is shortened, and the satisfaction degree of the user waiting for access to the manual seat is improved.

The embodiment of the application also provides an embodiment of a scheme block diagram of a response method, which is the same technical concept as the embodiment of the method. Fig. 3 is a block diagram of an answer method according to an embodiment of the present application.

As shown in fig. 3, the agent service system calculates whether a risk tag needs to be added to a response access request of a target user through response DM301, NLP intention recognition 302, and flow engine 303. The agent service system may execute manual queue priority segmentation in advance, where the queues of different segments correspond to different risk user types, or execute agent service line segmentation based on different service types. The agent service system can put the response access request of the target user into a corresponding queuing queue according to whether the response access request carries a risk tag and the risk user type corresponding to the risk tag so as to wait for the connection of an idle manual agent.

The present application also provides another embodiment of the response method, in view of the same technical concept as the previous embodiment of the method.

Fig. 4 is a process flow diagram of another response method according to an embodiment of the present application.

As shown in fig. 4, first, as shown, a target user inputs "manual customer service" 402, then, the agent service system transmits the user inputs to the response system 404, and then, the response system 404 invokes the NLP intent recognition 406 to perform intent recognition processing on the user inputs, and further, performs intent recognition classification 408, i.e., converts the user inputs into intelligent agent knowledge classification intent.

The response system invokes the multi-round dialog engine system through the NLP recognized intent classification-to-manual, and determines whether to start the process 410 through the process entry intent determination conditions.

If yes, returning a manual flow prompt 412 to the target user, and quickly entering 9 label node judgment of the risk user, wherein if any service point judgment is successful, the target user is indicated to be the risk user, and if all the 9 labels are failed, the target user is indicated to be a general user.

The 9 tag node judgment comprises the following steps: sensitive word check 414, turn manual check 416, historical conversation sensitive word check 418, whether incoming line 420 is repeated, whether member 422, whether there is a complaint in transit 1 (424), whether there is a complaint in transit 1 (426), whether there is a complaint in transit 2 (428), whether there is a complaint in transit 2 for x days (430).

And finally switching to the artificial seat through the label node judgment result of the target user, switching to the artificial special channel 436 if the target user is a risk user, and switching to the artificial normal channel 432 if the target user is a general user.

The manual transfer policy may be to determine whether the manual transfer channel identifier 438 is a normal identifier or a special identifier. If the transfer manual channel identifier 438 is a normal identifier, then the normal transfer manual is placed into the normal transfer manual queue segment 440. If the manual lane identification 438 is a special identification, then the user of the special lane is placed in the segment 442 with the higher queue priority. Thereby providing for seat allocation 444 for target users waiting for different segments.

The segments corresponding to the artificial special channels are used for queuing risk user services, when the artificial agents are configured, the special artificial agents for receiving the artificial special channels can be configured, and when the artificial agents of the artificial special channels are available, the agent service system can be automatically distributed to the idle agents corresponding to the queuing queues of the artificial special channels.

After the target user is linked to the manual seat, the problem of the target user can be solved through the professional knowledge, so that the professional business is solved by the professional seat.

After the manual agent meets the requirements of the current target user and determines that the service can be ended, the link with the current target user is terminated, so that the incoming line of other target users is distributed.

Because the technical conception is the same, the description in this embodiment is simpler, and the relevant parts only need to refer to the corresponding description of the embodiment of the response method provided above.

The present application embodiment also provides an embodiment of a further response method, in view of the same technical ideas as of the previous method embodiments. Fig. 5 is a process flow diagram of a method for forking a response according to an embodiment of the present application.

As shown in FIG. 5, a multi-turn dialog prompt 502 is entered, and then, a determination is made as to whether the sensitive word verification result 504, the turn count result 506, the history dialog triggers the sensitive word 508, whether the incoming line 510 is repeated, whether a member 512, whether a first type of on-way complaint 514 exists, whether a first type of history complaint 516 exists, whether a second type of on-way complaint 518 exists, whether a second type of history complaint 520 exists, respectively, and a risk user type 522 may be determined based on one or more of the determinations 504-520 described above.

In the above-described embodiments, a response method is provided, and a response device is provided corresponding to the response method, and the following description is made with reference to the accompanying drawings.

Fig. 6 is a schematic diagram of a response device according to an embodiment of the present application.

The present embodiment provides a response device, including:

an obtaining unit 601, configured to obtain a response access request of a target user; the response access request carries a first request identifier and a second request identifier; the first request identification includes at least one of a dialogue identification and a target user identification of the target user; the second request identifier is used for uniquely identifying and responding to the access request;

a first determining unit 602, configured to determine risk judgment data corresponding to the response access request according to at least one of the session identifier and the target user identifier;

a second determining unit 603, configured to determine a risk user type of the target user according to the risk judging data;

a placement unit 604, configured to place a second request identifier for answering the access request in a target queuing queue corresponding to the risk user type; each second request identifier in the plurality of second request identifiers included in the target queuing queue is sequentially arranged, and the arrangement sequence of the second request identifiers is used for representing the expected response sequence of the pending access request corresponding to the second request identifier.

Optionally, the answering device further includes:

the dialogue acquisition unit is used for acquiring dialogue data between the target user and the intelligent agent;

a third determining unit, configured to determine, based on a preconfigured set of sensitive words including a plurality of sensitive words, whether at least one target word matching the sensitive words is included in the dialogue data;

and the embedded point unit is used for carrying out data embedded point processing based on the target words if the dialogue data comprises at least one target word, so as to obtain embedded point data.

Optionally, the first determining unit 602 is specifically configured to:

inquiring buried point data corresponding to the dialogue identifier in the historical buried point data based on the dialogue identifier to obtain first buried point data;

reading the number of target words from the first buried point data to obtain a first number, and determining the first number as risk judgment data;

the second determining unit 603 is specifically configured to:

and if the first number is larger than a first preset number threshold, determining the risk user type of the target user as a first risk type.

Optionally, the first determining unit 602 is specifically configured to:

inquiring buried point data corresponding to the target user identifier in the historical buried point data based on the preset time length and the target user identifier to obtain second buried point data;

Reading the number of target words from the second buried data to obtain a second number, and determining the second number as risk judgment data;

the second determining unit 603 is specifically configured to:

and if the second number is larger than a second preset number threshold, determining the risk user type of the target user as a second risk type.

Optionally, the first determining unit 602 is specifically configured to:

inquiring whether at least one target complaint work order exists in a preset storage space according to the target user identification, obtaining a work order inquiring result, and determining the work order inquiring result as risk judging data; the target complaint worksheet meets the preset worksheet screening conditions; the corresponding relation between the user identification and the complaint work orders is stored in the preset storage space;

the second determining unit 603 is specifically configured to:

and if the work order query result is used for indicating that at least one target complaint work order exists, determining the risk user type of the target user as a third risk type.

Optionally, the answering device further includes:

the triggering unit is used for triggering the manual seat access request if the number of the target words included in the dialogue data is larger than a third preset number threshold;

the generation unit is used for generating a target behavior buried point record if the artificial seat is successfully accessed through the artificial seat access request;

The first determining unit 602 is specifically configured to:

inquiring the number of target behavior buried point records corresponding to the target user identification from the historical buried point data according to the target user identification to obtain a third number, and determining the third number as risk judgment data;

the second determining unit 603 is specifically configured to:

and if the third quantity is larger than the fourth quantity threshold value, determining the risk user type of the target user as a fourth risk type.

Optionally, the first determining unit 602 is specifically configured to:

inquiring the number of dialogue data corresponding to the target user identifier from the historical dialogue data according to the target user identifier and a preset time range to obtain a fourth number; inquiring the number of Interactive Voice Response (IVR) records corresponding to the target user identification from the historical call records according to the target user identification and a preset time range to obtain a fifth number; determining the fourth number and the fifth number as risk judgment data;

the second determining unit 603 is specifically configured to:

if the fourth number is greater than the fifth number threshold and the fifth number is greater than the sixth number threshold, determining the risk user type of the target user as a fifth risk type.

Optionally, the placement unit 604 is specifically configured to:

Placing a second request identifier for answering the access request in the target queuing queue, so that the second request identifier for answering the access request is positioned after the last second request identifier included in the target queuing queue; the target queuing queue comprises a plurality of second request identifications which are sequentially arranged according to the sequence of the placement time; each second request identifier in the plurality of second request identifiers corresponds to a to-be-processed access request of a to-be-accessed target artificial agent; the target artificial agent is an artificial agent corresponding to the type of the risk user.

Optionally, the placement unit 604 is specifically configured to:

determining the insertion position of a second request identifier for responding to the access request in the target queuing queue according to the risk judging data;

based on the insertion location, a second request identification that answers the access request is placed in the target queuing queue.

In the embodiment of the application, the response device comprises an acquisition unit, a first determination unit, a second determination unit and a placement unit; the system comprises an acquisition unit, a response access request acquisition unit and a response access request acquisition unit, wherein the acquisition unit is used for acquiring a response access request of a target user; the response access request carries a first request identifier and a second request identifier; the first request identification includes at least one of a dialogue identification and a target user identification of the target user; the second request identifier is used for uniquely identifying and responding to the access request; the first determining unit is used for determining risk judging data corresponding to the response access request according to at least one of the dialogue identifier and the target user identifier; the second determining unit is used for determining the risk user type of the target user according to the risk judging data; the placing unit is used for placing a second request identifier for responding to the access request in a target queuing queue corresponding to the risk user type; each second request identifier in the plurality of second request identifiers included in the target queuing queue is sequentially arranged, and the arrangement sequence of the second request identifiers is used for representing the expected response sequence of the pending access request corresponding to the second request identifier. In this way, through at least one of the dialogue identifier and the target user identifier carried in the response access request of the target user, the risk judgment data corresponding to the response access request can be determined, and further, the risk user types of the target user can be determined based on the risk judgment data, wherein each risk user type corresponds to a different queuing queue respectively, then by placing the second request identifier carried in the response access request of the target user in the target queuing queue corresponding to the risk user type, the target users of different risk user types can be pre-classified, the number of to-be-processed access requests corresponding to each queuing queue is indirectly reduced, the waiting time of the response access request is shortened, and the satisfaction degree of the user waiting for access to the manual seat is improved.

Corresponding to the above-described response method, based on the same technical concept, the embodiment of the present application further provides an electronic device, where the electronic device is configured to execute the above-provided response method, and fig. 7 is a schematic structural diagram of an electronic device provided in the embodiment of the present application.

As shown in fig. 7, the electronic device may have a relatively large difference due to different configurations or performances, and may include one or more processors 701 and a memory 702, where the memory 702 may store one or more storage applications or data. Wherein the memory 702 may be transient storage or persistent storage. The application programs stored in the memory 702 may include one or more modules (not shown), each of which may include a series of computer-executable instructions in the electronic device. Still further, the processor 701 may be arranged to communicate with the memory 702 and execute a series of computer executable instructions in the memory 702 on an electronic device. The electronic device may also include one or more power supplies 703, one or more wired or wireless network interfaces 704, one or more input/output interfaces 705, one or more keyboards 706, and the like.

In one particular embodiment, an electronic device includes a memory, and one or more programs, where the one or more programs are stored in the memory, and the one or more programs may include one or more modules, and each module may include a series of computer-executable instructions for the electronic device, and execution of the one or more programs by one or more processors includes instructions for:

acquiring a response access request of a target user; the response access request carries a first request identifier and a second request identifier; the first request identification includes at least one of a dialogue identification and a target user identification of the target user; the second request identifier is used for uniquely identifying and responding to the access request;

placing a second request identifier for responding to the access request in a target queuing queue corresponding to the risk user type; each second request identifier in the plurality of second request identifiers included in the target queuing queue is sequentially arranged, and the arrangement sequence of the second request identifiers is used for representing the expected response sequence of the pending access request corresponding to the second request identifier.

Corresponding to the above-described response method, the embodiments of the present application further provide a computer-readable storage medium based on the same technical concept.

The computer readable storage medium provided in this embodiment is configured to store computer executable instructions, where the computer executable instructions when executed by a processor implement the following procedures:

It should be noted that, in the present specification, the embodiments related to the computer readable storage medium and the embodiments related to the response method in the present specification are based on the same inventive concept, so the specific implementation of this embodiment may refer to the implementation of the foregoing corresponding method, and the repetition is omitted.

The foregoing describes specific embodiments of the present disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims can be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present embodiments may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present description can take the form of a computer program product on one or more computer-readable storage media (including, but not limited to, magnetic disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.

The present description is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the specification. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

In one typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

The memory may include volatile memory in a computer-readable medium, random Access Memory (RAM) and/or nonvolatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of computer-readable media.

Computer readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of storage media for a computer include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Disks (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium which can be used to store information that can be accessed by a computing device. Computer-readable media, as defined herein, does not include transitory computer-readable media (transmission media), such as modulated data signals and carrier waves.

It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises an element.

Embodiments of the application may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. One or more embodiments of the specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.

In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for system embodiments, since they are substantially similar to method embodiments, the description is relatively simple, as relevant to see a section of the description of method embodiments.

The foregoing description is by way of example only and is not intended to limit the present disclosure. Various modifications and changes may occur to those skilled in the art. Any modifications, equivalent substitutions, improvements, etc. that fall within the spirit and principles of the present document are intended to be included within the scope of the claims of the present document.

Claims

1. A method of responding, comprising:

2. The method of claim 1, wherein prior to obtaining the reply access request of the target user, further comprising:

acquiring dialogue data between the target user and the intelligent agent;

determining whether at least one target word matched with the sensitive words is included in the dialogue data or not based on a preset sensitive word set comprising a plurality of sensitive words;

and if the dialogue data comprises at least one target word, performing data embedding processing based on the target word to obtain embedded data.

3. The method according to claim 2, wherein determining risk judgment data corresponding to the response access request according to the session identifier includes:

reading the number of the target words from the first buried point data to obtain a first number, and determining the first number as the risk judgment data;

the determining the risk user type of the target user according to the risk judging data comprises the following steps:

4. The method according to claim 2, wherein determining risk judgment data corresponding to the response access request according to the target user identifier includes:

reading the number of the target words from the second buried data to obtain a second number, and determining the second number as the risk judgment data;

5. The method of claim 1, determining risk judgment data corresponding to the response access request according to the target user identifier, including:

inquiring whether at least one target complaint work order exists in a preset storage space according to the target user identification, obtaining a work order inquiring result, and determining the work order inquiring result as the risk judging data; the target complaint worksheet meets preset worksheet screening conditions; the corresponding relation between the user identification and the complaint work orders is stored in the preset storage space;

6. The method of claim 2, wherein after determining whether the dialogue data includes at least one target word matching the sensitive word based on a preconfigured set of sensitive words including a plurality of sensitive words, further comprising:

If the number of the target words included in the dialogue data is larger than a third preset number threshold, triggering a manual agent access request;

if the artificial seat is successfully accessed through the artificial seat access request, generating a target behavior buried point record;

according to the target user identifier, determining risk judgment data corresponding to the response access request, including:

inquiring the number of the target behavior buried point records corresponding to the target user identifier from historical buried point data according to the target user identifier to obtain a third number, and determining the third number as the risk judgment data;

and if the third quantity is larger than a fourth quantity threshold value, determining the risk user type of the target user as a fourth risk type.

7. The method according to claim 1, wherein determining risk judgment data corresponding to the response access request according to the target user identifier includes:

inquiring the number of dialogue data corresponding to the target user identifier from historical dialogue data according to the target user identifier and a preset time range to obtain a fourth number; inquiring the number of Interactive Voice Response (IVR) records corresponding to the target user identifier from a history call record according to the target user identifier and a preset time range to obtain a fifth number; determining the fourth number and the fifth number as the risk judgment data;

and if the fourth number is greater than a fifth number threshold and the fifth number is greater than a sixth number threshold, determining the risk user type of the target user as a fifth risk type.

8. The method of claim 1, wherein placing the second request identification of the reply access request in the target queuing queue corresponding to the risk user type comprises:

placing the second request identifier of the response access request in the target queuing queue, so that the second request identifier of the response access request is positioned after the last second request identifier included in the target queuing queue; the target queuing queue comprises a plurality of second request identifications which are sequentially arranged according to the sequence of the placement time; each second request identifier in the plurality of second request identifiers corresponds to a to-be-processed access request waiting for accessing a target artificial agent; the target artificial agent is an artificial agent corresponding to the risk user type.

9. The method of claim 1, wherein placing the second request identification of the reply access request in the target queuing queue corresponding to the risk user type comprises:

Determining the insertion position of the second request identifier of the response access request in the target queuing queue according to the risk judging data;

and inserting a second request identification of the response access request into the target queuing queue based on the insertion position.

10. A response device, comprising:

11. An electronic device, comprising:

a processor; and a memory configured to store computer-executable instructions that, when executed, cause the processor to perform the answering method of any one of claims 1-9.

12. A computer readable storage medium for storing computer executable instructions which, when executed by a processor, implement the answering method according to any one of claims 1 to 9.