CN115796816A

CN115796816A - Method for realizing mail customer service system and device, equipment, medium and product thereof

Info

Publication number: CN115796816A
Application number: CN202211678952.XA
Authority: CN
Inventors: 张振伟; 吴文龙
Original assignee: Shenzhen Zhiyan Technology Co Ltd; Shenzhen Qianyan Technology Co Ltd
Current assignee: Shenzhen Zhiyan Technology Co Ltd; Shenzhen Qianyan Technology Co Ltd
Priority date: 2022-12-26
Filing date: 2022-12-26
Publication date: 2023-03-14

Abstract

The application relates to a mail customer service system implementation method, a device, equipment, a medium and a product thereof, wherein the method comprises the following steps: pulling a newly added mail to a customer service mailbox corresponding to each channel according to mailbox configuration information of a plurality of online shop channels; predicting a target intention corresponding to the newly added mail by adopting a preset problem analysis model, and matching a mail template corresponding to the target intention; creating a work order based on the newly added mails, wherein the work order comprises a channel identifier of a source channel of the newly added mails, a commodity identifier of a commodity item pointed by the mail content of the newly added mails and the target intention, and the work order is correspondingly associated with the mail template; and matching corresponding agent users according to the channel identifications and the commodity identifications, and adding the work order to a work list of the agent users. The application realizes the customer service system based on the e-mails, can realize centralized management, and improves the customer service efficiency.

Description

Method for realizing mail customer service system and device, equipment, medium and product thereof

Technical Field

The present application relates to the field of intelligent customer service, and in particular, to a method, an apparatus, a device, a medium, and a product for implementing a mail customer service system.

Background

A merchant of a cross-border e-commerce usually operates a plurality of online shops through a plurality of e-commerce platforms, each e-commerce platform can usually provide an intelligent customer service robot to support customer service requirements of the corresponding online shop, but it is not practical for merchants correspondingly owning the plurality of online shops on the plurality of e-commerce platforms to realize centralized management and control of customer service data of the plurality of e-commerce platforms.

An alternative method is that customer service inlets are arranged in all online stores for users to feed back problems, in order to realize centralized management and control, an electronic mailbox is usually adopted to collect customer problems, a mailbox address of the electronic mailbox is configured in the corresponding online store so as to receive corresponding electronic mails, the customer service problems submitted by the online stores of the users are sent to corresponding mailbox servers in the form of the electronic mails, and customer service personnel can solve the customer service problems by checking and receiving the mails.

It is easy to see that, in such a scenario, although the customer service staff of the merchant can process the emails generated by the online shops in a centralized manner, the customer service staff can only process massive emails in a manual manner, and the processing manner is inefficient, cannot realize various analysis functions, and cannot embody the advantage of centralized management and control.

Theoretically, efficiency can be improved by automatically analyzing and replying each email through the customer service robot, however, the customer service robot cannot effectively provide effective reply texts due to the fact that the customer service robot often has the situations of incomplete information, unclear information and the like when a user reflects a problem, and therefore service quality is affected and expectations are not met.

In view of the above, how to efficiently handle the customer service problem based on email feedback is worthy of exploration.

Disclosure of Invention

The present application aims to solve the above problems and provide a mail service system implementation method, and a corresponding device, apparatus, non-volatile readable storage medium, and computer program product.

According to one aspect of the present application, a method for implementing a mail service system is provided, which includes the following steps:

pulling a newly added mail to a customer service mailbox corresponding to each channel according to mailbox configuration information of a plurality of online shop channels;

predicting a target intention corresponding to the newly added mail by adopting a preset problem analysis model, and matching a mail template corresponding to the target intention;

creating a work order based on the newly added mails, wherein the work order comprises a channel identifier of a source channel of the newly added mails, a commodity identifier of a commodity item pointed by the mail content of the newly added mails and the target intention, and the work order is correspondingly associated with the mail template;

and matching corresponding agent users according to the channel identifications and the commodity identifications, and adding the work order to a work list of the agent users.

In an optional embodiment, predicting a target intention corresponding to the newly added email by using a preset problem analysis model, and matching an email template corresponding to the target intention, includes:

coding the mail content of the newly added mail by referring to a preset mapping word list to obtain coding information;

inputting the coded information into a preset problem analysis model, and obtaining target intention and session information predicted by the problem analysis model according to deep semantic information of the coded information, wherein the session information comprises at least one element information for describing customer service problems;

and matching the mail template from the template library according to the target intention and the conversation information.

In an optional embodiment, matching the mail template from the template library according to the target intention and the session information includes:

determining a corresponding problem tree according to the target intention, wherein the problem tree comprises a plurality of problem units pre-associated according to a tree structure, and each problem unit correspondingly provides the mail template;

according to matching of element information contained in the session information and element configuration of each problem unit in the problem tree, determining the problem unit pointed by the session information, wherein the element configuration comprises a plurality of element labels, and each element label is used for indicating one type of element information;

and taking the mail template corresponding to the question unit pointed by the session information as a matched mail template.

In an optional embodiment, matching a corresponding agent user according to the channel identifier and the commodity identifier, and adding the work order to the work list of the agent user includes:

responding to the work order query request, and correspondingly pushing a work order statistical page;

responding to a list loading request automatically triggered by the work order counting page, counting the total work order quantity corresponding to each commodity item in a work list of the agent user, and sequencing the work orders in the work list according to the total work order quantity;

and pushing the sorted work list to the work order counting page for display, and associating a problem counting control for each commodity item in the work order counting page.

In an optional embodiment, after pushing the sorted work list to the work order statistics page for display, the method includes:

responding to a problem detail browsing request triggered by a problem counting control acting on a target commodity item in the working list, and counting the total work order quantity corresponding to each target intention of the target commodity item;

and generating a data chart according to each target intention and the corresponding work order total amount, and pushing the data chart to the terminal equipment of the agent user for display.

acquiring invitation configuration information, wherein the invitation configuration information comprises object specifying information and invitation description information, and the object specifying information is used for confirming a specific shop identifier and/or a specific commodity identifier;

determining a historical work order according to the object designated information in the invitation configuration information, and extracting a sender mailbox address of a newly added mail corresponding to the historical work order;

and sending an invited comment mail to the sender mailbox address, wherein the mail body of the invited comment mail contains the invited description information.

responding to the customer service bulletin board browsing request, and correspondingly pushing a customer service bulletin board page;

responding to a customer service effect data loading request automatically triggered by the customer service billboard page, and counting customer service effect data based on the work order processing result data of the seat user, wherein the customer service effect data comprises a plurality of statistical data dimensions;

and generating a data chart of the customer service effect data of each statistical data dimension, and pushing each data chart to the customer service billboard page for display.

responding to a mail sending request of the agent user for confirming sending of the mail template, and sending the mail template to a receiver mailbox address of the newly added mail;

monitoring a reply email replied by the sender of the newly added email based on the email template, and determining an updating target intention corresponding to the email content of the reply email;

and determining a reply text mapped with the updating target intention according to the updating target intention, and answering the reply mail with the reply text.

According to another aspect of the present application, there is provided a mail service system implementation apparatus, including:

the mail acquisition module is set to pull the newly added mails to the service mailboxes corresponding to the channels according to the mailbox configuration information of a plurality of online shop channels;

the problem analysis module is used for predicting a target intention corresponding to the newly added mail by adopting a preset problem analysis model and matching a mail template corresponding to the target intention;

the work order creating module is used for creating a work order based on the newly-added mails, the work order comprises channel identification of a source channel of the newly-added mails, commodity identification of a commodity item pointed by the mail content of the newly-added mails and the target intention, and the work order is correspondingly associated with the mail template;

and the work order distribution module is used for matching corresponding agent users according to the channel identification and the commodity identification and adding the work orders to a work list of the agent users.

According to another aspect of the present application, there is provided a mail service system implementation device, including a central processing unit and a memory, where the central processing unit is configured to invoke and run a computer program stored in the memory to execute the steps of the mail service system implementation method described in the present application.

According to another aspect of the present application, a non-volatile readable storage medium is provided, which stores a computer program implemented according to the mail service system implementation method in the form of computer readable instructions, and when the computer program is called by a computer, the steps included in the method are executed.

According to another aspect of the present application, there is provided a computer program product comprising computer programs/instructions which, when executed by a processor, implement the steps of the method described in any one of the embodiments of the present application.

Compared with the prior art, the method and the system have the advantages that the newly added mails submitted when customer service problems are fed back by consumer users in multiple online shop channels are pulled to the customer service mailbox, the target intentions of the newly added mails are analyzed by the aid of the preset problem analysis model, the corresponding mail templates are automatically matched according to the target intentions, the newly added mails are converted into work orders and are associated with the corresponding mail templates, the work orders are automatically distributed to seat users, centralized management of the customer service problems based on e-mail feedback is achieved through systematic management, reply effectiveness and reply efficiency are both considered between machine automation processing and manual seat answering, a merchant can conveniently conduct centralized automation preprocessing and automatic distribution on the customer service problems of all online shops scattered on different e-commerce platforms, centralized data management is achieved, customer service efficiency is embodied, and user experience is improved.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the description below are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings without creative efforts.

FIG. 1 is an architectural diagram of an exemplary network environment of the present application;

FIG. 2 is a schematic diagram of a network architecture of an intent analysis model as exemplary employed herein;

FIG. 3 is a schematic diagram of a network architecture of an exemplary named entity tagging model employed herein;

FIG. 4 is a schematic diagram of a network architecture of an exemplary problem analysis model employed herein;

FIG. 5 is a flow chart illustrating an embodiment of a method for implementing a mail service system of the present application;

FIG. 6 is a schematic flowchart illustrating a process of determining a mail template according to mail content in an embodiment of the present application;

FIG. 7 is a schematic flow chart illustrating the process of determining a mail template based on a question tree in the embodiment of the present application;

FIG. 8 is a schematic flow chart illustrating pushing a work order statistics page in an embodiment of the present application;

FIG. 9 is a schematic flow chart of sending an invitation mail in the embodiment of the application;

FIG. 10 is a schematic flow chart illustrating pushing a customer service bulletin board page in an embodiment of the present application;

FIG. 11 is a flowchart illustrating a reply text for providing a customer service question according to a reply email based on an email template reply in an embodiment of the present application;

FIG. 12 is a schematic block diagram of an apparatus for implementing the mail service system of the present application;

fig. 13 is a schematic structural diagram of an implementation apparatus of a mail service system according to the present application.

Detailed Description

The models cited or possibly cited in the application comprise a traditional machine learning model or a deep learning model, and unless specified in clear text, the models can be deployed in a remote server and remotely called at a client, and can also be deployed in a client with qualified equipment capability to be directly called.

Referring to fig. 1, a network architecture adopted in an exemplary application scenario in the present application includes a terminal device 80, a mailbox server 81, an application server 82, and a provider server 83, where the application server 82 may be used to deploy a computer program product implemented by programming according to the mail service system implementation method in the present application; thereby providing corresponding customer service system service, the mail server 81 can be used for storing the e-mail sent by the consumer user, and the application server 82 can pull or reply the corresponding e-mail; the terminal device 80 can be used for a consumer user and/or a manual agent user to log in various pages correspondingly provided by the customer service system service to complete various functions; the e-commerce server 83 is used for providing the customer user with access service of an independent station corresponding to the online shop.

In one embodiment, the mailbox server 81 may be multiple mailbox servers, each mailbox server is configured to provide a support service for a corresponding mailbox, the mailbox serves as a customer service mailbox, and the application server 82 is configured with corresponding mailbox configuration information in advance, such as login verification information of a mailbox address, so as to achieve access to the corresponding mailbox.

In one embodiment, the e-commerce server 83 may have a plurality of independent stations, each of which is used for providing independent station services corresponding to different online stores, and for each page of an independent station, one or more of the electronic mailboxes may be used as recipients for reflecting customer service problems, and when a corresponding customer user triggers a mail sending event for submitting a customer service problem, the corresponding independent station sends a corresponding mail to the corresponding recipient, and enters the corresponding mailbox server 81, so that the mail can be read by the application server 82.

Referring to fig. 2, a problem analysis model exemplarily employed by the present application is an intention analysis model, which is constructed based on a deep learning model, and is configured to extract deep semantic information of a given text information according to the deep learning model, perform classification mapping according to the deep semantic information, map the deep semantic information to each intention type in a preset classification space, obtain a classification probability corresponding to each intention type, and determine an intention type with the largest classification probability as a target intention corresponding to the text information.

In one embodiment, the intent analysis model employs a basic neural network model adapted to process serialized information as a backbone network, followed by a classifier based on the backbone network. The backbone network may include, but is not limited to, any of the following: a Recurrent Neural Network (RNN), a Long Short-Term Memory (LSTM), a transform encoder, a Bert, and the like; the classifier may be a multi-classifier constructed using a Softmax function.

The intention analysis model adopts training samples in a training data set and corresponding supervision labels in advance to carry out iterative training to a convergence state and then is put into an on-line reasoning for use. The training samples are text information, and usually include partial or all text contents for describing the customer service problem, and the supervision labels may be target intents corresponding to the customer service problem. When the intention analysis model needs to be subjected to one-time iterative training, one training sample is adopted for coding and then is input into the intention analysis model, corresponding deep semantic information is extracted, then, a corresponding result intention is predicted, then, the loss value corresponding to the result intention is calculated by adopting the supervision label, when the loss value reaches a target threshold value for verifying whether the model is converged, the intention analysis model is represented to be converged, and therefore, the iterative training process is terminated, otherwise, the intention analysis model is represented not to reach a convergence state, therefore, gradient updating is carried out on the intention analysis model according to the loss value, weight parameters of all links of the intention analysis model are corrected through back propagation, the model is promoted to be further converged, then, the next training sample is continuously called from the training data set to carry out the next iteration, and the rest is done until the intention analysis model reaches the convergence state.

Referring to fig. 3, for example, a named entity tagging model used in the present application is built based on a deep learning model, and is configured to extract deep semantic information according to given text information, and then extract each keyword and a corresponding type tag thereof according to the deep semantic information, so as to obtain mapping relationship data between the type tag and the keyword. In this application, the type tag may be an element tag corresponding to various types of element information required for describing information for defining a customer service problem, and the keyword is a type tag corresponding thereto, that is, an element tag.

In one embodiment, the named entity tagging model is implemented by taking a basic neural network model suitable for processing serialized information as a backbone network and then connecting a conditional random field network on the basis of the backbone network. The backbone network may include, but is not limited to, any of the following: a Recurrent Neural Network (RNN), a Long Short-Term Memory (LSTM), a transform encoder, a Bert, and the like; the Conditional Random Field network may be a CRF (Conditional Random Field) model.

Similarly, the named entity labeling model adopts training samples in a training data set and corresponding supervision labels in advance to carry out iterative training to a convergence state, and then is put into an online reasoning mode for use. The training sample is text information, and usually contains partial or all text contents corresponding to the customer service problem, and the supervision tag may be element information and an element tag thereof corresponding to the customer service problem. When the named entity tagging model needs to be subjected to one-time iterative training, one training sample is adopted for coding and then is input into the named entity tagging model, corresponding deep semantic information is extracted, corresponding element information and corresponding element labels thereof are predicted to serve as result data, then the supervision labels are adopted for calculating loss values corresponding to the result data, when the loss values reach a target threshold value used for verifying whether the model converges, the named entity tagging model is represented to be converged, and therefore the iterative training process is terminated, otherwise, the named entity tagging model does not reach a convergence state, so that gradient updating is carried out on the named entity tagging model according to the loss values, weight parameters of all links of the named entity tagging model are corrected through back propagation, the model is promoted to be further converged, then, a next training sample is continuously called from the training data set for carrying out the next iteration, and the analogy is carried out until the named entity tagging model reaches the convergence state.

Further embodiments referring to fig. 4, an embodiment of another exemplary network architecture of the problem analysis model is disclosed, in which the problem analysis model integrates the named entity tagging model on the basis of the intention analysis model, the intention analysis model and the named entity tagging model share the same backbone network for extracting deep semantic information of text information input therein, and then are processed separately through two branch networks, namely, a target intention is determined by a classifier in a first branch network, and a named entity is determined by a conditional random field network in a second branch network. It is understood that for such a network architecture, the intention analysis model and the named entity tagging model in the whole problem analysis model are jointly trained, and in each iterative training, respective loss values of the intention analysis model and the named entity tagging model can be superimposed to be an overall loss value, and then whether to continue the iteration and update the weight is decided according to the overall loss value.

In an embodiment, any one of the intention analysis model, the named entity tagging model, the problem analysis model, and the like, which needs to be independently run, may be deployed in any running space that can be called by the application server, so as to provide a service for the relevant requirements of the application.

In another embodiment, the problem analysis model may be encapsulated by a standardized service, which provides an interface for calling to receive text information input, and then the problem analysis model is directly called within the standardized service to obtain corresponding result data and then returned correspondingly.

In another embodiment, the text information processed by the named entity tagging model may be first text information, after predicting each element information and its corresponding element tag according to the first text information by the named entity tagging model, constructing all predicted element information as session information indicating the same session, where the session information is still text information, and then inputting the session information as second text information in the intention analysis model in combination with the first text information for predicting a corresponding target intention. Therefore, the input standardization degree of the intention analysis model can be improved, and the intention analysis model can predict the corresponding target intention more accurately.

Based on the above principle, please refer to fig. 5, in an embodiment of the method for implementing a mail service system according to the present application, comprising the following steps:

step S1100, pulling a new mail to a service mailbox corresponding to each channel according to mailbox configuration information of a plurality of online shop channels;

a merchant of a cross-border e-commerce typically operates one or more online stores, each of which runs in a respective independent station, each of which may be affiliated with a different e-commerce platform.

In one embodiment, when a customer user of any online store needs to submit a customer service question, the customer service question can be submitted after editing a relevant question description through a corresponding page of the corresponding online store, and the question description submitted by the user is sent to a default customer service mailbox address as a mail text to enter a corresponding mailbox server. In another embodiment, the user can edit the corresponding problem description through the email box of the user and send the corresponding problem description to the designated customer service email address by himself to enter the corresponding email server.

Therefore, in the customer service system of the merchant, the data communication connection to each customer service mailbox address is established in advance, and the e-mails are incrementally pulled to each corresponding e-mail mailbox in a timing or non-timing mode, so that the newly added mails in each customer service mailbox are continuously acquired. In order to establish the data communication connection, mailbox configuration information corresponding to the customer service mailboxes of all online store channels can be preset. Specifically, the same merchant may have a plurality of online stores, the online stores may be distributed in a plurality of e-commerce platforms, the online stores may share the same email box as a customer service mailbox, or may use different email boxes, each online store forms an independent channel, and the mailbox configuration information is provided in one-to-one correspondence with the corresponding customer service mailbox.

The mailbox configuration information is determined according to the mail communication protocol adopted by the corresponding customer service mailbox, such as POP, IMAP, SMTP or other protocols predefined by a mailbox server, and the like, wherein the POP and the IMAP are receiving protocols, and the SMTP is a sending protocol. The protocol is suitable for receiving the mail and provides various corresponding parameters according to the protocol, so that the corresponding mailbox configuration information can be formed. The various parameters may be, for example, an account number, an authorization code, a port number, and the like of the customer service mailbox, which are specifically determined according to an interface specification of a corresponding protocol.

After the mailbox configuration information of the customer service mailbox is provided, the mail can be pulled to the corresponding customer service mailbox at a certain reasonable frequency, for example, every time of 10 to 20 minutes. Of course, the corresponding customer service mailbox or other mailboxes can be used for sending the mails at any time. The frequency of pulling the mails is reasonably set, so that the balance between mail updating and system computing power can be realized. When a mail is pulled to a target mailbox, i.e. a target customer service mailbox, according to a certain frequency, the mail is usually acquired in an increment mode, and the mail pulled each time is treated as a newly added mail.

After a new mail of a customer service mailbox is obtained, the new mail can be analyzed, so that various information in the new mail can be read, and the new mail can be used for determining mail content, determining a sender mailbox address and the like. The mail content may be specified in advance according to actual needs, and specific components thereof may be specified, for example, the mail content may include only a mail body, may include a mail subject and a mail body, may include a mail body and a signature file, or further integrate the mail subject, the mail body, the mail signature file, and the like.

The extraction of each part of content from the newly added mail can be flexibly applied, for example, the mail content can be formed by combining the mail subject and the mail body, or the mail content can be formed by combining the mail body and the mail signature file, or the mail content can be formed by combining the mail subject, the mail body and the mail signature file, and the like. Currently, it is still feasible to use the mail body alone as the mail content.

And reserving a sender mailbox address extracted from the newly-added mail, configuring the sender mailbox address as a target mailbox address for replying the newly-added mail after determining a corresponding mail template aiming at the newly-added mail, namely a receiver mailbox address, and simultaneously using the mail template as a mail body and referring to an original topic of the newly-added mail as a reply topic so as to construct a reply mail, and sending the reply mail to the newly-added mail through a preset sending interface.

In some embodiments, the newly added mails generated by a plurality of online store channels can be converted into mail messages, and centralized and ordered scheduling is performed by means of a task list, which can be processed according to the following procedures:

firstly, respectively pulling new mails to corresponding target customer service mailboxes according to mailbox configuration information of each customer service mailbox preset corresponding to each online shop channel;

then, for each newly added mail, constructing a corresponding mail message: the customer service system may be responsible for processing newly added mails of a plurality of customer mailboxes simultaneously, and the newly added mails of the plurality of customer service mailboxes are easy to block when the number of the newly added mails is too large, therefore, a large number of the newly added mails can be processed based on a message mechanism, and accordingly, a mail message is correspondingly constructed for each newly added mail, and the newly added mail is encapsulated into a message body;

and finally, marking the mail message as a to-be-executed state, adding the mail message into a task list, and sequentially performing the following consumption: and adding the mail message into a task list, wherein the task list is used for storing the processing state of each mail message and assisting in realizing a message mechanism of mail processing, and accordingly, when the mail message of a newly added mail is added into the task list, the state initialization of the mail message is marked as a to-be-executed state.

Corresponding to the above process, when it is required to sequentially read each mail message from the task list, the following process may be performed:

firstly, responding to the consumption event of the mail message in the state to be executed in the task list, extracting a sender mailbox address, a mail subject, a mail body and a mail signature file in the newly-added mail, and on the basis of establishing a message mechanism by means of the task list, the mail message in the state to be executed in the task list is sequentially listed according to a queue mechanism, and one consumption event is correspondingly triggered when one mail message is listed. In response to the consumption event, the newly added mails carried in the consumption event can be analyzed, and the mailbox address of the sender, the mail subject, the mail body and the mail signature file in the newly added mails are respectively extracted.

And then packaging any items in the mail subject, the mail text and the mail signature file into mail content, and simultaneously determining the sender mailbox address as a target mailbox address for replying the newly added mail.

According to the embodiment, in order to prevent the response from being out of order caused by a large number of added mails, the ordered consumption of each mail message is realized by establishing a message mechanism by virtue of the task list and recording the state of the mail message of each added mail, so that the higher scheduling efficiency of each added mail can be ensured. In addition, in the process of consuming each mail, standardized analysis can be carried out on the corresponding newly-added mail, mail content is packaged, and a sender mailbox address is determined, so that early-stage standardized work is done for a subsequent consumer-answering user, and the information processing efficiency is improved.

Step S1200, predicting a target intention corresponding to the newly added mail by adopting a preset problem analysis model, and matching a mail template corresponding to the target intention;

in the present application, a standardized service is implemented in advance for predicting the target intention of the corresponding customer service problem according to the mail content, and in some embodiments, the standardized service is also responsible for extracting the session information provided corresponding to the customer service problem.

The objective intention means that after customer service problems are classified in advance, a plurality of types of problems are formed, a corresponding intention type is calibrated corresponding to each type of problem, when one mail content is predicted to be one intention type, the intention type is an objective intention corresponding to the mail content, and therefore a customer service problem gate class to which the mail content belongs is determined.

The session information is a set of each element information for describing a certain customer service problem, and this set of element information constitutes description information of the customer service problem. Generally, complete description of a customer service problem requires providing several different element information according to a predetermined specification, and when a session information completely contains all the element information in the predetermined specification, a complete description of the corresponding customer service problem is formed. Each element information may be referred to using a corresponding element tag for calling, and thus, may be defined using a corresponding set of element tags when a predetermined specification of a customer service problem is given. It is understood that the element label sets of different customer service problems may be the same or different, and may be predetermined according to actual needs.

For example, describing a customer service problem corresponding to a fault, according to a preset specification, element information corresponding to element tags such as an order number, a contact number, a commodity identification and the like needs to be provided, when all the element information is included in session information, a complete description of the customer service problem is formed, otherwise, the session information is incomplete in describing the customer service problem.

Mail content extracted from the newly added mail is transferred to the standardized service by calling an interface provided by the standardized service, and the standardized service further determines corresponding target intentions and session information according to the mail content, wherein the target intentions are usually determined and unique, and the session information usually includes element information corresponding to one or more element labels, but it should be noted that the element information included in the session information may still not completely correspond to all element information required by the customer service problem to which the target intentions are directed.

In one embodiment of the standardized service, an intention analysis model can be independently adopted as a problem analysis model to independently determine a target intention, and then the mail template is directly called for replying a newly-added mail according to a preset mapping relation between the target intention and the mail template; in another embodiment, the email template can be determined in a refined manner by using the session information, and a problem analysis model integrating an intention analysis model and a named entity tagging model can be respectively adopted to correspondingly determine corresponding target intention and session information according to the email content; in another embodiment, the named entity tagging model may be used to predict the mail content to obtain corresponding session information, the intention analysis model may be used to predict a corresponding target intention according to the session information and the text set of the mail content, and the mail template may be determined by refining the target intention and the session information.

In some embodiments, encapsulation of the standardized service can be eliminated, and corresponding calls can be directly made to the required models, so as to predict corresponding target intentions and session information according to the mail content.

The mail template can independently contain a preset format file for providing instruction information for guiding a user to solve the corresponding customer service problem. In some embodiments, the mail templates may further include prompt information for guiding the customer user to supplement materials to complete the conversation information required for solving the problem, in which case, each mail template may correspond to a corresponding specification, in which a plurality of given element information may be described, and on the basis of determining the mail template set corresponding to the target intention, when the element information in the conversation information is identical to the element information in a certain mail template, the corresponding mail template is confirmed to be a usable target mail template.

Step S1300, creating a work order based on the newly added mails, wherein the work order comprises channel identifiers of source channels of the newly added mails, commodity identifiers of commodity items pointed by mail contents of the newly added mails and the target intentions, and the work order is correspondingly associated with the mail template;

through the above process, for a new mail, various corresponding information can be obtained, including but not limited to: the target intention corresponding to the customer service question, the commodity identification of the commodity item determined according to the corresponding element information provided in the mail content of the newly-added mail, such as the order number, the online shop channel identification determined according to the platform identification determined by the suffix part of the sender mailbox address of the newly-added mail, and the like can be followed up for clue, and various specific information related to the customer service question can be obtained. Therefore, a corresponding work order can be created for the newly added mail, so that the work order at least comprises a channel identification corresponding to a source channel of the newly added mail, a commodity identification of a commodity item pointed by mail content of the newly added mail and mapping relation data between target intentions, and various information required by the work order is perfected according to preset specifications.

And step S1400, matching corresponding agent users according to the channel identifications and the commodity identifications, and adding the work order to a work list of the agent users.

For a customer service system, usually a plurality of agent users are included, each agent user handles a customer service problem corresponding to a certain kind of features by dividing work, for example, one agent user may handle a customer service problem of an online store, one agent user may handle a customer service problem corresponding to a product line, one agent user may be responsible for centrally handling a customer service problem related to a specific product line of a specific online store, and the like, and the configuration information may be pre-configured to determine the work order processing configuration information corresponding to each agent user.

In an exemplary application, the work order processing configuration information of the agent user may include a channel identifier and/or a commodity identifier, for example, when one work order processing configuration information includes a channel identifier of the online store a and a commodity identifier of the product line B, the corresponding agent user is only responsible for the work order corresponding to the commodity item from the product line B of the online store a.

Therefore, for the work orders created for each newly added mail, because the work orders carry the corresponding channel identifiers and commodity identifiers, the corresponding channel identifiers and commodity identifiers are matched with the work order processing configuration information of each agent user, the work orders are distributed to the agent users who realize matching, and the work orders are added into the work lists of the agent users, so that the automatic distribution of the work orders is completed.

When the agent user starts to process the work order, customer service problems can be quickly grasped through channel identification, commodity identification, target intention and the like in the work order, editing is carried out as required on the basis of the mail template associated with the work order, and then the mail template is quickly replied to the sender of the newly added mail.

According to the embodiment, the newly added mails submitted when customer service problems are fed back by customer users in a plurality of online shop channels are pulled to the customer service mailbox, the target intentions of the newly added mails are analyzed by using the preset problem analysis model, the corresponding mail templates are automatically matched according to the target intentions, the newly added mails are converted into work orders and are associated with the corresponding mail templates, then the work orders are automatically distributed to the seat users, the centralized management of the customer service problems based on the E-mail feedback is realized, the reply effectiveness and the reply efficiency are realized between the machine automatic processing and the manual seat answering, the centralized automatic preprocessing and the automatic distribution of the customer service problems of all online shops scattered on different E-commerce platforms are conveniently realized by merchants, the centralized data management is realized, the customer service efficiency is embodied, and the user experience is improved.

On the basis of any one of the embodiments of the present application, referring to fig. 6, predicting a target intention corresponding to the newly added email by using a preset problem analysis model, and matching a email template corresponding to the target intention, includes:

step S1210, coding the mail content of the newly added mail by referring to a preset mapping word list to obtain coding information;

before the mail content is used for determining the target intention and the session information, in one embodiment, the mail content is subjected to formatting preprocessing, including but not limited to: and removing punctuation marks, stop words, invalid characters and the like to obtain the purified text of the mail content.

Then, a bag-of-words model or other word segmentation algorithms such as N-Gram and the like are adopted to segment the purified text to obtain a plurality of lemmas, a preset mapping dictionary is adopted to inquire out the characteristic value corresponding to each lemma, and the characteristic values of all the lemmas are constructed into the coded information corresponding to the purified text.

In one embodiment, in the encoded information, position information of each lemma in the mail content may be further superimposed, and the position information may be absolute position information and/or relative position information.

Step S1220, inputting the coded information into a preset problem analysis model, and obtaining target intention and session information predicted by the problem analysis model according to deep semantic information of the coded information, wherein the session information comprises at least one element information for describing customer service problems;

after the coded information corresponding to the mail content is determined, the coded information is input into a problem analysis model, and the problem analysis model adopted in this embodiment is a comprehensive model obtained by integrating an intention submodel model and a named entity tagging model, so that the problem analysis model extracts corresponding deep semantic information according to the coded information through a backbone network of the problem analysis model, and then on one hand, each named entity in the problem analysis model is extracted according to the deep semantic information through a first branch network, and element tags corresponding to the named entities are determined and serve as corresponding element information. The determined mapping relation data between the element label and the element information can form the conversation information corresponding to the mail content; on the other hand, the deep semantic information is mapped to a preset classification space by the second branch network, and the intention type indicated by the category corresponding to the maximum classification probability is determined as the target intention.

And step S1230, matching the mail template from the template library according to the target intention and the session information.

In one embodiment, a single standardized mail template can be provided corresponding to each type of customer service problem, so that the matching efficiency of the mail template can be improved, specifically, the corresponding mail template can be directly determined in a one-to-one correspondence mode according to the target intention, and the session information can be reserved for later use.

In another embodiment, for the same type of customer service problem, a plurality of email templates may be preset, where a preset specification corresponding to each email template describes an element tag set corresponding to element information included in the matched session information, and when a corresponding email template needs to be matched, the element tag of the element information of the session information is compared with the element tag set of the email template, so that the email template matching the session information may be determined, and the element tag provided in the preset specification of the email template is exactly the element tag of a part of element information included in the session information corresponding to the customer service problem.

It is easy to understand that due to the fact that the deep learning model is applied when the target intention and the session information are predicted, the target intention and the session information can be predicted more intelligently, efficiency and accuracy are high, manual intervention is avoided, and batch processing of a large amount of mail contents is facilitated.

On the basis of any embodiment of the present application, please refer to fig. 7, matching an email template from a template library according to the target intention and the session information includes:

step S1241, determining a corresponding problem tree according to the target intention, wherein the problem tree comprises a plurality of problem units which are pre-associated according to a tree structure, and each problem unit correspondingly provides the mail template;

in this embodiment, a large number of preset email templates are organized by a problem tree, so as to effectively organize multiple-customer-service problems and multiple element information missing situations, and the determined email template plays a more delicate and precise role in a guiding process of completing session information for a consumer user, and has steps. To facilitate the search, a question tree may be built up for each target intent, whereby when a target intent is determined, its corresponding question tree may be determined.

The problem tree, in one embodiment, may be constructed based on relational data, and includes a plurality of problem units, each problem unit is associated with and stores a corresponding email template and hierarchical dependency information thereof, the email template provides a corresponding preset specification in addition to a preset format text, the preset specification includes an element tag set, and the element tag set includes element tags corresponding to known element information when the element information provided by the session information corresponding to the customer service problem is incomplete, so that matching may be achieved by subsequently comparing the session information with the email template to perform element tag comparison. It can be seen that the preset specification of the mail template actually describes the element configuration that the conversation information should provide. Different problem units, namely different email templates have different element information missing situations corresponding to each other, so that different element tag sets are included.

In another embodiment, the problem tree may also be constructed based on a knowledge graph, in which mail templates corresponding to the problem units are stored in a node manner, and edge connection information is constructed between any two problem units according to actual hierarchical dependency relationship information.

It can be seen that the hierarchical relationship between the mass email templates is constructed in a problem tree manner, which is helpful for improving the matching efficiency between the session information and the email templates, and thus, in some embodiments, after a consumer user replies to an email template provided by a first newly added email, the steps S1100 to S1400 may not be iterated again for a subsequent newly added email replied by the user, but only the first determined target intent is used, in the corresponding problem tree, the element information supplemented by the user is merged with the first determined session intent, and then the merged element information is compared with the corresponding preset specification of the email template in the lower-level problem unit of the first determined email template to match the subsequent email template, so that the consumer user is guided to provide corresponding element information gradually, and a multi-branch and multi-detail guiding flow is realized.

Step S1242, determining a problem unit pointed by the session information according to matching of the element information included in the session information with the element configuration of each problem unit in the problem tree, where the element configuration includes a plurality of element tags, and each element tag is used to indicate one type of element information;

when the email template is matched for the session information in a problem tree, because the session information includes mapping relationship data between the element tags and the element information, it is only necessary to compare the element tag sets of each element information existing in the session information with the element configuration of each email template under the target intention, that is, the preset specification, and when the element tag sets of the session information and the element tag sets configured for the element configuration of the email template are all equal, the email template is the matched target email template, and accordingly, the problem unit pointed by the session information is determined.

And step S1243, taking the email template corresponding to the question unit pointed by the session information as a matched email template.

In some embodiments, for the same problem unit, mail templates corresponding to multiple styles and languages may also be provided, for this reason, preset specifications of the mail templates corresponding to the respective styles and languages may require that the session information includes a corresponding channel identifier as a platform identifier for identifying a specific e-commerce platform, when the session information is matched with the mail templates, matching of the platform identifiers is performed at the same time, and in multiple mail templates of the same problem unit, the mail template carrying the corresponding platform identifier is determined to be used as a target mail template.

The platform mark is essentially a channel mark, and the core function of the platform mark is to distinguish different online shops, namely, to distinguish different independent stations, so as to distinguish different e-commerce platforms. Specifically, in order to distinguish different independent stations of a merchant, a dedicated customer service mailbox, namely a recipient mailbox of a customer service problem, can be configured for an online store of each independent station, so that each independent station can share one customer service mailbox, and different independent stations can be distinguished through the customer service mailbox. Considering that different independent stations are usually supported by different e-commerce platforms, an enterprise mailbox provided by a mailbox server of a corresponding e-commerce platform is generally provided, and such enterprise mailbox also usually takes the own website of the e-commerce platform or the independent station as a suffix and can be distinguished by the suffix, so that after a new added mail is received, a suffix part of a recipient mailbox, which specifically refers to the address of the recipient mailbox, can be extracted and used as a channel identifier corresponding to the new added mail, and then used as a platform identifier for representing a source independent station of the new added mail, namely on a source line.

It is understood that when the mail template is used to respond to the new mail and the consumer user responds to the mail template to generate a new mail, the above steps S1100 to S1400 may still be performed to perform iterative processing on such new mail, so as to further determine whether the session information is complete, and if the session information is still incomplete, the user may be further guided to complete the session information until the description of the customer service problem by the session information meets the requirement of the preset specification corresponding to the customer service problem.

According to the embodiment, the mail template for guiding the consumer user to perfect the customer service problem description can be determined according to the target intention and the session information, so that the mail template has a flow guiding function, and before the consumer user incompletely describes the customer service problem, an intermediate process before the customer service problem is solved is intelligently guided, so that the solving efficiency of the customer service problem can be improved, and manpower and material resources are saved.

On the basis of any embodiment of the present application, the mail message is marked as a to-be-executed state, and after being added to the task list for in-order dequeue consumption, state switching of the mail message in the task list can be maintained according to the following procedures, including:

firstly, orderly reading mail messages in a task list in a to-be-executed state for consumption;

when the newly added mails are processed, orderly scheduling is carried out according to the mail messages in the task list, specifically, only the mail messages marked as the to-be-executed state are consumed, and the mail content of the corresponding newly added mails is extracted from the mail messages to be used for determining the mail template. To indicate the status of the mail message in a timely manner, when the mail message begins to be consumed, its status may be modified to be in a processing state to indicate a distinction.

In one embodiment, multiple concurrent threads can be started, and multiple mail messages marked as a to-be-executed state are consumed at a time, so that the consumption efficiency of the mail messages is improved.

Secondly, marking the mail message replied by the mail template as a finished state;

after each mail message is consumed, that is, the mail template of the corresponding newly-added mail is determined and replied, in order to avoid repeated processing or processing omission of the mail message, the state of the mail message may be further set as the finished state.

In addition, the mail message which is consumed but not replied by the mail template is transmitted to a manual customer service interface for processing, and the mail message is marked as a manual processing state.

When the consumption process of one mail message is finished but a corresponding mail template is not obtained, aiming at the situation, the problem which cannot be solved by the automation process usually exists, and manual agent intervention is needed, so that the corresponding mail message is further transmitted to a manual agent interface and sent to a manual agent user, and the manual agent user can perform further processing.

According to the embodiment, the state of the mail messages is maintained in time, so that massive mail messages passing through the task list can be guaranteed to be orderly and stably consumed, and the efficient operation of the customer service system is guaranteed.

On the basis of any one of the embodiments of the present application, please refer to fig. 8, where matching out a corresponding agent user according to the channel identifier and the product identifier, and adding the work order to the work list of the agent user includes:

step S2100, responding to the work order query request and correspondingly pushing a work order statistical page;

the agent user can initiate a work order query request through a work page provided by a customer service system realized by the application, the customer service system receives the work order query request and then responds the work order query request, correspondingly, a work order statistical page is pushed to the agent user, the work order statistical page comprises codes corresponding to various data in the acquisition page, and the codes are executed after the codes are analyzed and operated by a browser of terminal equipment of the agent user to realize bidirectional interaction with an application server.

Step S2200, responding to a list loading request automatically triggered by the work order counting page, counting the total work order amount corresponding to each commodity item in the work list of the agent user, and sequencing the work orders in the work list according to the total work order amount;

and in the process of analyzing and executing the work order statistical page, automatically triggering a list loading request, and indicating a customer service system to push a corresponding work list for the customer service system. And the customer service system responds to the list loading request and carries out data statistics on the working list of the seat user.

Considering that each work order is generally used as an essential item of the corresponding session information of the work order in the construction process, each item of data of the corresponding commodity item can be acquired from the corresponding independent station through the order number, and therefore, actually, each work order in the work list of the seat user can also be converted into corresponding index according to the commodity item. Therefore, the total work order quantity obtained by each commodity item can be counted according to each commodity item in the work list of the agent user, then, the commodity items are sequenced according to the total work order quantity, and generally, the commodity items are sequenced according to the total work order quantity from high to low, so that the content in the work list is updated. It will be appreciated that the greater the total amount of work orders for a commodity item, the more customer service problems it encounters.

And S2300, pushing the sorted work list to the work order counting page for display, and associating a problem counting control for each commodity item in the work order counting page.

Further, the sorted work list can be pushed to the terminal device of the agent user, and the work list is loaded to the work order statistical page after being correspondingly analyzed by the browser of the terminal device. In the analysis process, according to the preset business logic, a problem counting control can be set and displayed in a correlated mode corresponding to each commodity item, so that an agent user can conveniently and quickly master all customer service problem general views of the related commodity items through the problem counting control.

According to the embodiment, the work order data can be further converted into commodity item data, statistics and sequencing of customer service problems based on the E-mails are achieved, overall information of various customer service problems comprehensively appearing in a plurality of online shop channels can be conveniently mastered, ordered arrangement of the information is achieved, and customer service problem processing efficiency is improved.

On the basis of any one embodiment of the present application, after pushing the sorted work list to the work order statistics page for display, the method includes:

step S2400, responding to a problem detail browsing request triggered by a problem counting control acting on a target commodity item in the work list, and counting the total work order quantity corresponding to each target intention of the target commodity item;

on the work order statistics page, an agent user can touch a problem statistics control corresponding to any one target commodity item to trigger a problem detail browsing request, and after receiving the problem detail browsing request, a customer service system can perform work order statistics on various customer service problems of the target commodity item.

Customer service problems encountered by a commodity item are various, for example, quality problems, logistics problems, after-sales maintenance problems and the like may be generated, each customer service problem is converted into a work order through an electronic mail, and therefore, from the dimension of the commodity item, the distribution of the number of work orders of each commodity item on different customer service problems is dynamically changed, so that in order to present the detailed information, for the target commodity item, the total number of work orders can be counted according to the target intention corresponding to the work order, and the total number of work orders corresponding to each target intention can be counted. Since an objective is intended to correspond to a class of customer service problems, it is actually the work order quantity distribution that counts the customer service problems.

And S2500, generating a data chart according to each target intention and the corresponding work order total amount, and pushing the data chart to the terminal equipment of the agent user for displaying.

After the number of work orders of the target commodity items under each target intention is determined, a corresponding data chart can be generated by means of a chart generation tool, then the data chart is pushed to terminal equipment of the seat user, and the data chart is analyzed by a browser and displayed.

According to the embodiments, the customer service system of the application can perform statistics of work order quantity distribution of various customer service problems encountered by the commodity item based on the work orders obtained from the customer service problems fed back by the e-mails, so that various outstanding problems of the commodity item can be conveniently and intensively grasped, and an efficiency tool and useful information are provided for customer service processing and production management.

On the basis of any one embodiment of the present application, please refer to fig. 9, where matching out a corresponding agent user according to the channel identifier and the commodity identifier, and adding the work order to the work list of the agent user includes:

step S3100, acquiring invitation configuration information, wherein the invitation configuration information comprises object specifying information and invitation description information, and the object specifying information is used for confirming a specific shop identifier and/or a specific commodity identifier;

the method and the system allow for establishing an invitation mechanism for inviting the consumer user to evaluate the satisfaction degree of the customer service problem solution, the satisfaction degree of the commodity item, the satisfaction degree of the merchant overall service and the like, and also realize an advertisement mechanism. For this reason, invitation configuration information, which is configurable by an agent user, a customer service system administrator, or the like, and generally includes object specifying information and invitation description information, may be set in advance.

The object specifying information may be generalized, and may be expressed as a channel identifier for specifying a specific online store, which is used as a specific store identifier to cover the entire number of merchandise items of the store, or may be expressed as a specific merchandise identifier to cover a specific merchandise item or each merchandise item below a specific product line. And the like, can be flexibly set as required.

The invitation description information can be a self-defined text used for introducing the invitation and evaluation event and can be flexibly defined.

Step S3200, determining a historical work order according to the object designation information in the invitation configuration information, and extracting a sender mailbox address of a newly added mail corresponding to the historical work order;

when the invitation is required to be executed, searching the total or the historical work orders in the specified range in the customer service system according to the object specifying information in the invitation configuration information, such as the specific shop identifier or the specific commodity identifier, so as to search out part of the target historical work orders.

In some embodiments, the target user group may be further filtered by using a preset rule, for example, the target user group is filtered according to age, gender, and region, and the target user group meeting the expectation is finally determined.

And S3300, sending an invited comment mail to the sender mailbox address, wherein the mail text of the invited comment mail contains the invited comment description information.

After the target user group is determined, the invitation mail can be sent to the mailbox address list group of each consumer user in the target user group, and the purpose of invitation is achieved. The body text of the invitation mail may include the description information of the invitation, and of course, may also include information edited by the agent user or the administrator.

According to the embodiment, on the basis of the customer service system, the invitation mechanism is further realized by using the convenience of e-mail communication, the merchant can conveniently realize the maintenance of the customer relationship, the further feedback information can be acquired through the subsequent communication with the user, the popularization aim is realized, and the functions of the customer service system are enriched.

On the basis of any one of the embodiments of the present application, please refer to fig. 10, where a corresponding agent user is matched according to the channel identifier and the product identifier, and after the work order is added to the work list of the agent user, the method includes:

step S4100, responding to the customer service bulletin board browsing request, and correspondingly pushing a customer service bulletin board page;

the method comprises the steps that an agent user can initiate a customer service bulletin board browsing request through a working page provided by a customer service system, the customer service system receives the customer service bulletin board browsing request and then responds, correspondingly, a customer service bulletin board page is pushed to the agent user, codes corresponding to various data in the page are obtained in the customer service bulletin board page, and the codes are executed after the codes are analyzed and operated by a browser of terminal equipment of the agent user to realize bidirectional interaction with an application server.

Step S4200, responding to a customer service effect data loading request automatically triggered by the customer service billboard page, and counting customer service effect data based on work order processing result data of an agent user, wherein the customer service effect data comprises a plurality of statistical data dimensions;

after each agent user processes the work orders in the work list, the relevant users can evaluate each work order to obtain the corresponding evaluation data of each work order, in addition, the relevant data such as the processing time, the processing progress and the like of each work order can be included to form the work order processing result data of each work order, and the customer service effect data of the corresponding agent user can be generated according to the work order processing result data.

And in the process of analyzing and executing the customer service bulletin board page, automatically triggering a data loading request, and indicating a customer service system to push corresponding customer service effect data for the customer service system. And the customer service system responds to the data loading request and carries out data statistics on the customer service effect data of the seat user in the customer service system.

When the customer service effect data of the agent user in the customer service system is unified, based on the processing result data of each worksheet of the agent user, statistics of multiple statistical data dimensions can be included, for example, single-dimension statistics is performed according to the number of good comments, the number of praise and the amount of complaints of the agent user, classification statistics is performed according to the amount of complaints and channel identification, and the like, so that various customer service effect data can be obtained.

And S4300, generating data charts of the customer service effect data of each statistical data dimension, and pushing each data chart to the customer service billboard page for display.

After determining the customer service effect data of the target commodity item under each target intention, generating a corresponding data chart by means of a chart generation tool, and then pushing the data chart to the customer service billboard page for display.

According to the embodiment, the customer service system is convenient to count the solution of the customer service problems and the working performance of the seat users, is convenient to intensively master various outstanding problems in customer service management, and provides an efficiency tool and useful information for customer service processing and production management.

On the basis of any one of the embodiments of the present application, please refer to fig. 11, where a corresponding agent user is matched according to the channel identifier and the product identifier, and after the work order is added to the work list of the agent user, the method includes:

step S5100, responding to a mail sending request of the agent user for confirming sending the mail template, and sending the mail template to a receiver mailbox address of the newly-added mail;

when the agent user reads the work order and processes one work order, the mail template can be sent to the sender mailbox of the newly added mail after the mail template associated with the work order is confirmed, so that the processing of the work order is completed.

Step S5200, monitoring a reply email replied by the sender of the newly added email based on the email template, and determining an update target intention corresponding to the email content of the reply email;

after a mail template is used for replying a newly added mail, the mail template is sent to a corresponding consumer user, the consumer user can reply according to the prompt message given by the mail template, element information corresponding to the element label in the mail template is supplemented, and then the mail is re-sent, so that the customer service system can pull the corresponding reply mail.

In an embodiment, for such a reply email supplemented based on the email template, since the information therein is relatively straightforward, it is possible to determine various element information provided in the email content of the reply email based on the email content in the reply email directly by adopting a rule matching manner or by adopting a named entity tagging model of the present application without iterative processing of the processes of the foregoing embodiments of the present application, and supplement the element information into the session information that has been previously determined for the newly added email, so that the session information can completely correspond to the description information of one customer problem.

In another embodiment, the problem analysis model of the present application may be invoked again to determine its target intent according to the mail content in the reply mail, which is expected to obtain more accurate prediction results because the mail content after the customer user supplements the material provides more complete key semantics.

In another embodiment, the mail content in the newly added mail and the complete conversation information obtained according to the reply mail may be merged and then input into the problem analysis model, the mail content provides comprehensive semantics, the complete conversation information provides key semantics, and the corresponding target intention is predicted more accurately by virtue of the reasoning ability of the problem analysis model.

The target intention determined in the above manner may be changed with a very small probability with respect to the target intention predicted solely for the mail content of the newly added mail, at which time the process of matching the mail template may be regressed. However, in most cases, the target intention determined here is merely a confirmation of the target intention predicted from the mail content of the new mail, and serves to accurately understand the type of the customer service problem. The target intention re-predicted from the reply mail may be referred to as an update target intention to show a difference.

Step S5300, determining a reply text mapped with the update target intention according to the update target intention, and responding to the reply email with the reply text.

After confirming the target intent of updating, the reply text corresponding to the target intent can be called, and the reply text is sent to the mailbox address of the receiver of the newly added mail and is sent to the consumer user. The reply text is usually a description text for instructing the user to solve the corresponding customer care question, and may be preset by a person skilled in the art, and in principle, each customer care question may provide a reply text correspondingly.

According to the embodiment, after the user of the consumer completes the session information, the type of the customer service problem proposed by the user is confirmed by predicting the target intention again, and the preset reply text is called after the confirmation to automatically provide the solution of the customer service problem, so that the automatic response is realized, the digestion efficiency of the customer service problem is improved, and the accuracy and the reliability of automatically solving the customer service problem are improved.

Referring to fig. 12, an apparatus for implementing a mail service system according to an aspect of the present application includes a mail obtaining module 1100, a problem analyzing module 1200, a work order creating module 1300, and a work order splitting module 1400, where: the mail acquisition module 1100 is configured to pull a new mail to a service mailbox corresponding to each of the channels according to mailbox configuration information of a plurality of online store channels; the problem analysis module 1200 is configured to predict a target intention corresponding to the newly added email by using a preset problem analysis model, and match an email template corresponding to the target intention; the work order creating module 1300 is configured to create a work order based on the newly added email, where the work order includes a channel identifier of a source channel of the newly added email, a commodity identifier of a commodity item pointed by email content of the newly added email, and the target intention, and the work order is correspondingly associated with the email template; the work order distribution module 1400 is configured to match corresponding agent users according to the channel identifiers and the commodity identifiers, and add the work orders to the work list of the agent users.

On the basis of any embodiment of the present application, the problem analysis module 1200 includes: the content coding unit is set to code the mail content of the newly added mail by referring to a preset mapping word list to obtain coding information; the problem analysis unit is used for inputting the coded information into a preset problem analysis model and obtaining target intention and session information predicted by the problem analysis model according to deep semantic information of the coded information, wherein the session information comprises at least one element information for describing customer service problems; and the template matching unit is used for matching the mail template from the template library according to the target intention and the conversation information.

On the basis of any one embodiment of the present application, the template matching unit includes: the problem tree determining subunit is set to determine a corresponding problem tree according to the target intention, the problem tree comprises a plurality of problem units pre-associated according to a tree structure, and each problem unit correspondingly provides the mail template; a question determining subunit configured to determine a question unit that is hit by the session information, based on matching of element information included in the session information with an element configuration arrangement of each question unit in the question tree, the element configuration arrangement including a plurality of element tags each indicating one type of element information; and the template determining subunit is configured to take the mail template corresponding to the question unit pointed by the session information as the matched mail template.

On the basis of any one of the embodiments of the present application, the work order splitting module 1400 includes: the list request module is set to respond to the work order query request and correspondingly pushes a work order counting page; the statistical processing module is set to respond to the list loading request automatically triggered by the work order statistical page, count the total work order quantity corresponding to each commodity item in the work list of the agent user, and sort the work orders in the work list according to the total work order quantity; and the list pushing module is used for pushing the sorted work list to the work order counting page for display, and associating the problem counting control for each commodity item in the work order counting page.

On the basis of any one of the embodiments of the present application, the list push module includes: the browsing response module is set to respond to a question detail browsing request triggered by a question counting control acting on the target commodity item in the working list and count the total work order amount corresponding to each target intention of the target commodity item; and the chart pushing module is set to generate a data chart according to each target intention and the corresponding work order total amount, and pushes the data chart to the terminal equipment of the agent user for display.

On the basis of any one of the embodiments of the present application, the list push module includes: the system comprises a configuration acquisition module, a configuration acquisition module and a configuration processing module, wherein the configuration acquisition module is used for acquiring invitation configuration information which comprises object designation information and invitation description information, and the object designation information is used for confirming a specific shop identifier and/or a specific commodity identifier; the work order calling module is set to determine a historical work order according to the object designation information in the invitation configuration information and extract a sender mailbox address of a newly added mail corresponding to the historical work order; and the invitation evaluation notification module is set to send an invitation evaluation mail to the sender mailbox address, and the mail body of the invitation evaluation mail contains the invitation evaluation description information.

On the basis of any one of the embodiments of the present application, the list pushing module includes: the billboard response module is set to respond to the customer service billboard browsing request and correspondingly push a customer service billboard page; the billboard counting module is set to respond to a customer service effect data loading request automatically triggered by the customer service billboard page, and count customer service effect data based on the work order processing result data of the seat user, wherein the customer service effect data comprises a plurality of statistical data dimensions; and the graph display module is used for generating a data graph of the customer service effect data of each statistical data dimension and pushing each data graph to the customer service billboard page for display.

On the basis of any one of the embodiments of the present application, the list push module includes: the agent reply module is set to respond to the mail sending request of the agent user for confirming the mail template sending and send the mail template to the receiver mailbox address of the newly added mail; the tracking monitoring module is used for monitoring a reply email replied by the sender of the newly added email based on the email template and determining an updating target intention corresponding to the email content of the reply email; and the complete reply module is set to determine a reply text mapped with the updating target intention according to the updating target intention and reply the reply email by the reply text.

Another embodiment of the present application further provides a mail service system implementation device. As shown in fig. 13, the mail service system implements the internal structure of the device. The mail service system implementing device includes a processor, a computer-readable storage medium, a memory, and a network interface connected by a system bus. The computer-readable non-volatile readable storage medium of the mail service system implementation device stores an operating system, a database and computer-readable instructions, the database can store information sequences, and the computer-readable instructions can enable a processor to implement a mail service system implementation method when being executed by the processor.

The processor of the mail service system realizing device is used for providing calculation and control capacity and supporting the operation of the whole mail service system realizing device. The memory of the mail service system implementation device can store computer readable instructions, and when the computer readable instructions are executed by the processor, the processor can execute the mail service system implementation method of the application. The network interface of the mail customer service system realization equipment is used for connecting and communicating with the terminal.

It will be understood by those skilled in the art that the configuration shown in fig. 13 is a block diagram of only a portion of the configuration relevant to the present application, and does not constitute a limitation on the mail service system implementation apparatus to which the present application is applied, and a particular mail service system implementation apparatus may include more or less components than those shown in the figure, or may combine certain components, or have a different arrangement of components.

In this embodiment, the processor is configured to execute specific functions of each module, sub-module, and unit in fig. 12, and the memory stores program codes and various data required for executing the modules or sub-modules. The network interface is used for realizing data transmission between user terminals or servers. In the present embodiment, a nonvolatile readable storage medium stores program codes and data required for executing all modules in the mail client system implementation apparatus of the present application, and a server can call the program codes and data of the server to execute functions of all modules.

The present application further provides a non-transitory readable storage medium storing computer readable instructions which, when executed by one or more processors, cause the one or more processors to perform the steps of the mail service system implementation method of any of the embodiments of the present application.

The present application also provides a computer program product comprising computer programs/instructions which, when executed by one or more processors, implement the steps of the method as described in any of the embodiments of the present application.

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments of the present application may be implemented by hardware related to instructions of a computer program, which may be stored in a non-volatile readable storage medium, and when executed, may include the processes of the embodiments of the methods as described above. The storage medium may be a computer-readable storage medium such as a magnetic disk, an optical disk, a Read-Only Memory (ROM), or a Random Access Memory (RAM).

To sum up, this application realizes customer service system based on email, can realize the centralized management to the email that the dispersion produced in a plurality of online shops of different electricity merchant platforms, carry out intelligent analysis and reposition of redundant personnel to the customer service problem wherein, make the customer service problem based on email feedback obtain high-efficient handling, between machine automated processing and the response of manual seat, reply validity and reply efficiency have been taken into account, make things convenient for the merchant to carry out centralized automation preliminary treatment and automatic dispatch with the customer service problem of each online shop of dispersion at different electricity merchant platforms, realize centralized data management, embody customer service efficiency, improve user experience.

Claims

1. A mail customer service system implementation method is characterized by comprising the following steps:

2. The method as claimed in claim 1, wherein the step of predicting a target intention corresponding to the new email by using a preset problem analysis model and matching a email template corresponding to the target intention comprises:

3. The method as claimed in claim 2, wherein matching the mail template from the template library according to the target intention and the session information comprises:

and taking the mail template corresponding to the problem unit pointed by the session information as a matched mail template.

4. The method as claimed in any one of claims 1 to 3, wherein matching out a corresponding agent user according to the channel identifier and the product identifier, and adding the work order to the work list of the agent user comprises:

5. The method as claimed in claim 4, wherein pushing the sorted work list to the post-order statistics page comprises:

6. The method as claimed in any one of claims 1 to 3, wherein matching out a corresponding agent user according to the channel identifier and the product identifier, and adding the work order to the work list of the agent user comprises:

7. The method as claimed in any one of claims 1 to 3, wherein matching out a corresponding agent user according to the channel identifier and the article identifier, and adding the work order to the work list of the agent user comprises:

8. The method as claimed in any one of claims 1 to 3, wherein matching out a corresponding agent user according to the channel identifier and the product identifier, and adding the work order to the work list of the agent user comprises:

and determining a reply text mapped with the updating target intention according to the updating target intention, and responding the reply mail with the reply text.

9. An apparatus for implementing a mail service system, comprising:

the mail acquisition module is set to pull a newly added mail to the customer service mailboxes corresponding to all the channels according to the mailbox configuration information of a plurality of online shop channels;

the work order creating module is used for creating a work order based on the newly added mails, the work order comprises channel identification of a source channel of the newly added mails, commodity identification of commodity items pointed by mail contents of the newly added mails and the target intention, and the work order is correspondingly associated with the mail template;

10. A mail service system implementing device comprising a central processor and a memory, characterized in that said central processor is adapted to invoke the execution of a computer program stored in said memory to perform the steps comprised by the method according to any one of claims 1 to 8.

11. A non-transitory readable storage medium storing a computer program in the form of computer readable instructions, the computer program when invoked by a computer performing the steps comprised by the method according to any one of claims 1 to 8.

12. A computer program product comprising computer programs/instructions for performing the steps comprised by the method according to any one of claims 1 to 8 when called upon by a processor.