CN113515705A

CN113515705A - Response information generation method, device, equipment and computer readable storage medium

Info

Publication number: CN113515705A
Application number: CN202110853383.7A
Authority: CN
Inventors: 陈奕坤
Original assignee: Ping An Puhui Enterprise Management Co Ltd
Current assignee: Ping An Puhui Enterprise Management Co Ltd
Priority date: 2021-07-27
Filing date: 2021-07-27
Publication date: 2021-10-19

Abstract

The application relates to the field of intelligent decision making, and provides a response information generation method, a response information generation device, response information generation equipment and a storage medium, wherein the method comprises the following steps: acquiring task exception information sent from a client; determining task identification of the task according to the task abnormal information; based on a process management system, acquiring process node information of the task according to the task identifier; extracting task keywords from the task abnormal information based on a keyword extraction model; judging whether the task keywords are matched with the process node information; if the process node information is judged to be matched with the task keywords, determining a target reply text according to the task keywords; generating response information according to the task identification and the target reply text based on a response information generation model; and sending the response information to the client. The accuracy of automatic response can be effectively improved. The application also relates to a blockchain technology, and the task exception information and the response information can be stored in the blockchain.

Description

Response information generation method, device, equipment and computer readable storage medium

Technical Field

The present application relates to the field of intelligent decision making technologies, and in particular, to a method, an apparatus, a device, and a computer-readable storage medium for generating response information.

Background

With the development of computer technology, more and more technologies are applied in the financial field, and the traditional financial industry is gradually changing to financial technology. Due to the complexity of the financial environment, users often present a number of very difficult or very specialized financial problems. Since financial matters are related to the personal interests of customers, it is necessary to answer these questions accurately in time.

At present, the answer to the financial question is mostly solved by a manual answer mode. When some complicated problems are faced or many problems are processed simultaneously through manual answering, the processing speed is slow, and meanwhile, the problem that the customer cannot be responded timely can also cause that the customer cannot be responded timely to complain, so that negative effects are caused. Meanwhile, there is also an automatic answer mode, but most of the current automatic answers are only automatic answers according to the types of questions reported by users, such as user reports: why has the task done and has not yet completed? The automatic response can feed back to a user for various reasons of incomplete tasks, the user needs to analyze the task by himself, and meanwhile, the problem that the task exception information reported by the user cannot be really solved due to the fact that the information of the client is not updated timely enough, the process seen by the user is different from the real task process, the automatic response still gives an error response according to the information reported by the user, and therefore the user needs to contact with a manual answer, and the task exception processing efficiency is low.

Disclosure of Invention

The present application mainly aims to provide a method, an apparatus, a device and a computer readable storage medium for generating response information, and aims to improve the efficiency and accuracy of automatically responding to abnormal information reported by a client.

In a first aspect, the present application provides a response information generating method, including:

acquiring task exception information sent from a client;

determining task identification of the task according to the task abnormal information;

based on a process management system, acquiring process node information of the task according to the task identifier;

extracting task keywords from the task abnormal information based on a keyword extraction model;

judging whether the task keywords are matched with the process node information;

if the process node information is judged to be matched with the task keywords, determining a target reply text according to the task keywords;

generating response information according to the task identification and the target reply text based on a response information generation model;

and sending the response information to the client.

In a second aspect, the present application further provides a response information generating apparatus, including:

the abnormal information acquisition module is used for acquiring task abnormal information sent from the client;

the task identifier determining module is used for determining the task identifier of the task according to the task abnormal information;

the flow node information determining module is used for acquiring the flow node information of the task according to the task identifier based on a flow management system;

the keyword extraction module is used for extracting task keywords from the task abnormal information based on a keyword extraction model;

the information matching judgment module is used for judging whether the task keywords are matched with the process node information;

the target reply text determining module is used for determining a target reply text according to the task keyword if the flow node information is judged to be matched with the task keyword;

the response information generation module is used for generating response information according to the task identifier and the target reply text based on a response information generation model;

and the response information sending module is used for sending the response information to the client.

In a third aspect, the present application further provides a computer device comprising a processor, a memory, and a computer program stored on the memory and executable by the processor, wherein the computer program, when executed by the processor, implements the steps of the response information generating method as described above.

In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, where the computer program, when executed by a processor, implements the steps of the response information generating method as described above.

The application provides a response information generation method, a response information generation device, response information generation equipment and a computer readable storage medium, wherein task abnormal information sent from a client is obtained; determining task identification of the task according to the task abnormal information; based on a process management system, acquiring process node information of the task according to the task identifier; extracting task keywords from the task abnormal information based on a keyword extraction model; judging whether the task keywords are matched with the process node information; if the process node information is judged to be matched with the task keywords, determining a target reply text according to the task keywords; generating response information according to the task identification and the target reply text based on a response information generation model; and sending the response information to the client. The task keywords are extracted through the keyword extraction model, and the flow node information of the task is acquired based on the flow management system, so that the abnormal information of the task and the task flow node information reported by a user can be more accurately determined; by comparing the extracted task keywords with the flow node information, error response caused by untimely updating of the task flow of the client can be avoided; and the response information can be determined more accurately by determining the response text through the task keywords, and meanwhile, the response efficiency is improved.

Drawings

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

Fig. 1 is a schematic flowchart of a method for generating response information according to an embodiment of the present application;

fig. 2 is a schematic view of a scene for implementing the method for generating response information according to the present embodiment;

fig. 3 is a schematic block diagram of a response information generating apparatus according to an embodiment of the present application;

fig. 4 is a block diagram illustrating a structure of a computer device according to an embodiment of the present application.

Detailed Description

The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some, but not all, embodiments of the present application. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.

The flow diagrams depicted in the figures are merely illustrative and do not necessarily include all of the elements and operations/steps, nor do they necessarily have to be performed in the order depicted. For example, some operations/steps may be decomposed, combined or partially combined, so that the actual execution sequence may be changed according to the actual situation.

The embodiment of the application provides a response information generation method and device, computer equipment and a computer readable storage medium. The response information generation method can be applied to terminal equipment and/or a server, and the terminal equipment can be electronic equipment such as a mobile phone, a tablet computer, a notebook computer, a desktop computer, a personal digital assistant and wearable equipment. The server may be, for example, a single server or a cluster of servers.

Some embodiments of the present application will be described in detail below with reference to the accompanying drawings. The embodiments described below and the features of the embodiments can be combined with each other without conflict.

Referring to fig. 1, fig. 1 is a schematic flowchart illustrating a method for generating response information according to an embodiment of the present application.

For example, as shown in fig. 2, fig. 2 is a scene diagram provided in an embodiment of the present application, where a server obtains task exception information of a client, processes the task exception information, and sends obtained response information to the client, so as to solve an exception problem provided by a user.

As shown in fig. 1, the response information generating method includes steps S101 to S108.

And step S101, acquiring task abnormal information sent from the client.

Exemplarily, task exception information reported from the client is obtained, wherein the task may be a loan task established by the user in an application program of the client or an upgrade task of the application program, the task exception information may be a problem that a loan cannot be made, or a loan is not yet made by a due date, or the like, or a problem that the loan cannot be upgraded, and a problem occurs in a using process of the client, and the task exception information may be reported to the server through the client.

Illustratively, the task exception information includes a plurality of fields, the plurality of fields are input by the client, and through the plurality of fields in the task exception information, the server can identify which piece of the task the client reports, and identify what problem occurs. It can be understood that the task may be characterized by a unique task identifier, so that the server can identify the currently reported task by the task identifier and obtain information about an abnormal condition by using other fields.

For example, the task exception information may be stored in the blockchain, so that the server can obtain the task exception information from the blockchain.

And S102, determining task identification of the task according to the task abnormal information.

Illustratively, the reported task exception information includes both a task identifier for indicating the task and a description of an exception condition occurring in the task, and the task identifier needs to be extracted from the reported task exception information.

For example, the task identifier includes a field generated based on a preset identifier generation rule, and in general, the task identifier has a preset generation rule, the generation rule may be composed of a plurality of characters according to a certain arrangement order, for example, an arrangement combination of characters and numbers, and the task identifier may be identified from the plurality of fields of the task exception information based on the generation rule.

In some embodiments, the determining a task identifier of a task according to the task exception information includes: determining the position relation among a plurality of fields in the task exception information; and determining the task identifier of the task according to the position relation among the fields.

For example, each field in the task exception information may be located, that is, fields appearing in the task exception information are numbered by reading the sequence of each field in the task exception information, where the number of the field is used to indicate the position of the field in the task exception information, and the positional relationship between the fields can be obtained by the numbering, such as what is the last field or the next field of the current field.

Illustratively, the fields may further include a plurality of characters, the characters may be at least one of chinese characters, english letters, numbers, and punctuation marks, or may include other special symbols, such as greek letters, and the task identifier may be determined by the characters included in the fields after the acquiring sequence.

Exemplarily, determining the task identifier of the task according to the position relationship among the fields includes judging whether the characters located in the preset field length meet a preset sorting condition; and if the characters in the preset field length are judged to accord with the preset sorting condition, determining the task identifier according to the characters in the preset length.

For example, the preset sorting condition may be a permutation and combination of a plurality of words and a plurality of numbers, and all the words are located before the number, for example, the preset length is 13 fields, where the sorting condition is that 5 english letters are located before 8 numbers (HPDAC123456789), when determining the task identifier, it may be determined whether the english field is located before the number field by obtaining the sequence of the fields, if so, then the character in the field is determined to be the preset sorting condition, for example, whether 5 english characters appear continuously in the english field, whether 8 numeric characters appear continuously in the number field, and if both are true, it is determined that a field of 13 fields length composed of the 5 english characters and the 8 numeric characters is the task identifier.

For example, the task identifier may be determined by determining a position relationship between characters according to numbers corresponding to the characters in the fields.

The task identifier is determined in the task abnormal information through the preset sequencing condition, so that the precision of determining the task identifier from a plurality of fields can be effectively improved, and the error probability is reduced.

And step S103, acquiring the process node information of the task according to the task identifier based on the process management system.

For example, the task identifier determined from the task exception information may be input into the process management system to obtain the task corresponding to the task identifier in the process management system, so as to obtain the process node information of the task, and it may be understood that the process node information may be a step to which the task is performed, for example, a creditor uploads a credential.

For example, the process node information may be stored in a blockchain, that is, the process management system may store the process node information of the task in a blockchain manner, after obtaining the task identifier, the server broadcasts the task identifier to a blockchain network, where the task identifier may include a channel type, a client type, and/or a flow number, and in the blockchain network, a storage address of the task may be determined by the channel type, the client type, and/or the flow number.

The task identifier may include a channel type, a client type and/or a serial number, and when the task identifier is broadcast to the blockchain network, the blockchain network may obtain the storage address according to a mapping relationship between the channel type, the client type and/or the serial number and the task storage address. The storage address is used to indicate the storage location of the task in the block chain, such as on a block of the block chain; the block chain network knows the storage position of the task which needs to be called by the current server in the block chain according to the storage address, and can find the corresponding block of the block chain to extract the task which is needed by the server.

The block chain is a novel application mode of computer technologies such as distributed data storage, point-to-point transmission, a consensus mechanism and an encryption algorithm. A block chain (Blockchain), which is essentially a decentralized database, is a series of data blocks associated by using a cryptographic method, and each data block contains information of a batch of network transactions, so as to verify the validity (anti-counterfeiting) of the information and generate a next block. The blockchain may include a blockchain underlying platform, a platform product service layer, an application service layer, and the like.

For example, the process management system may further store the task identifier and the process node information based on a mapping relationship, that is, in the process management system, the task identifier corresponds to the process node information of the task one by one, and the process node information of the corresponding task can be queried through the task identifier.

Illustratively, based on the process management system, each time the server provides a service to the client, the server sends a process update instruction to the process management system, where the process update instruction is used to instruct the process management system to update a process node of a task corresponding to the currently provided service, and it is possible to analyze whether an abnormal problem of the task at the client is the same as the process node information in the process management system through the task abnormal information reported by the client and the process node information of the task acquired by the server, so that it is possible to determine whether an abnormal situation is caused by an update of the client being not in time, and reduce the situation of false response.

And step S104, extracting task keywords from the task abnormal information based on a keyword extraction model.

Illustratively, the task keywords are extracted from the task exception information, and it can be understood that the extracted task keywords may be descriptions of exception conditions of the task written when the user reports the task exception information.

For example, the task keyword may include "question", "unable", and the like, and when the above-mentioned character is detected, the characters near the above-mentioned character, such as the characters at the front and back 5 positions, may be simultaneously obtained, or the first punctuation mark is detected forward and the second punctuation mark is detected backward, and all the characters located in the first punctuation mark and the second punctuation mark are extracted to obtain the task keyword.

Illustratively, keywords may be extracted for the task exception information based on a keyword extraction model. The keyword extraction model can be obtained by training the neural network model according to the labeled keyword data, and the parameters of the neural network model can be obtained by learning and adjusting from the labeled keyword data based on an algorithm framework of online machine learning.

For example, the labeled keyword data may include keyword data of a common corpus and/or a business corpus, wherein the common corpus is, for example, open-source corpus participle data, and the business corpus data may be business corpus participle data stored on the process management system.

Illustratively, the extraction of the keywords may be performed on the task exception information based on a keyword extraction model and sequence labeling of the words. For the word sequence of the input task abnormal information, the keyword extraction model can mark a mark for identifying a word boundary for each word in the task abnormal information, and the task keywords in the task abnormal information can be determined according to the mark for identifying the word boundary.

Illustratively, the extraction of the keywords may also be performed on the task exception information based on the keyword extraction model and the labeled keyword data. For the acquired task abnormal information, the keyword extraction model can compare the task abnormal information with the labeled keyword data, and according to the comparison result, the same or similar phrases are determined as the task keywords in the task abnormal information.

In some embodiments, the extracting task keywords in the task exception information based on a keyword extraction model includes: traversing all fields in the task abnormal information based on a traversing network of the keyword extraction model to obtain key value pairs corresponding to the fields; and acquiring key value pairs of preset key words, inputting the key value pairs of the preset key words and the key value pairs corresponding to the fields in the task abnormal information into a comparison network of the key word extraction model, and determining the task key words.

Illustratively, the keyword extraction model may include a traversal network and a comparison network, and as can be understood, the traversal network is used to traverse the task exception information to obtain key-value pairs of all fields in the task exception information; and the comparison network is used for comparing the key value pairs of all the fields in the task abnormal information with the key value pairs of the preset keywords to determine the task keywords.

For example, in the task exception information, the storage form of the field may be stored in a form of a Key-Value pair (Key-Value), where Key is fixed as a field and is represented by using a field object, and Value may be any one of a field, a list, a hash, a set, and an ordered set object, and traversing all fields in the task exception information may obtain the Key-Value pair of each field.

And comparing the preset key value pairs of the keywords with the key value pairs of each field in the task abnormal information one by one through a comparison network, determining the fields with the key value pairs same as the preset key value pairs, and determining the fields as the task key words of the task abnormal information.

For example, the preset key value pair includes A, B, C, the traversed key value pairs of the fields include a, B, C, d, and e, through comparing a with a, B, C, d, and e one by one, if the key value pairs corresponding to a and any one of a, B, C, d, and e are not the same, a is not the task key of the task exception information, it can be understood that B and C are operated the same as a, and if B and C are both the same as the corresponding key value pair of one of a, B, C, d, and e, B and C are determined as the task key of the task exception information.

The method comprises the steps of traversing the abnormal information of the task through a traversing network to obtain key value pairs corresponding to all fields in the abnormal information, comparing the key value pairs corresponding to the fields with key value pairs of preset keywords through a comparison network to determine the task keywords, effectively improving the precision of determining the task keywords, and determining the target reply text capable of answering the user appeal according to the user appeal when determining the reply text.

And step S105, judging whether the task keywords are matched with the process node information.

Illustratively, whether a task keyword determined from the task exception information matches with the process node information is judged, and it can be understood that if the task keyword does not match with the process node information, it may be that network delay is not timely fed back to the current process node information of the client at the client, and if the task keyword matches with the process node information, it may be determined that an exception condition occurs in the current process node.

In some embodiments, the determining whether the task keyword matches the process node information includes: determining matching degree according to the process node information and the task keywords; and if the matching degree is greater than a preset matching threshold value, judging that the process node information is matched with the task keywords, wherein the matching degree is positively correlated with the number of the task keywords appearing in the process node information.

For example, since the description of the user on the problem may be different from the description of the task by the flow node information updated in the server, a matching threshold may be set to determine whether the task keyword of the task matches the flow node information. It can be understood that the matching degree may be the same number of the text characters in the process node information as the number of the text characters in the task keywords, that is, the matching degree is positively correlated with the number of the task keywords appearing in the process node information.

For example, the matching threshold may be set to 10, 5 keywords in the task keywords are the same as 5 words in the process node information, it is determined that the matching degree of the task keywords and the process node information is 5, and if the matching degree is less than the matching degree threshold, the task keywords are not matched with the process node information; if 15 keywords of the task keywords are the same as 15 words in the process node information, determining that the matching degree of the task keywords and the process node information is 15, and if the matching degree is greater than a matching degree threshold value, matching the task keywords with the process node information.

It will be appreciated that the match threshold may be determined based on the total word count of the task key and/or the flow node information.

For example, the extraction of the keyword may be performed on the flow node information based on the keyword extraction model and the sequence label of the word. For the input word sequence of the process node information, the keyword extraction model can label each word in the process node information with a mark for identifying a word boundary, and can determine one or more keywords in the process node information according to the mark for identifying the word boundary. Or extracting keywords from the process node information based on the keyword extraction model and the labeled keyword data. For the process node information, the keyword extraction model can compare the process node information with the labeled keyword data, and determine the same or similar phrases as one or more node keywords in the process node information according to the comparison result. And determining whether the process node information is matched with the task keywords or not by comparing the node keywords with the task keywords.

The matching degree is determined through the task keywords and the node keywords in the process node information, whether the task keywords are matched with the process node information or not can be judged more accurately, and therefore the target reply text can be determined more accurately, the target reply text can be responded to the user, and the use experience of the user is improved.

And S106, if the process node information is judged to be matched with the task keywords, determining a target reply text according to the task keywords.

For example, if it is determined that the process node information does not match the task keyword, an alarm message may be sent to a terminal of the network provider, so that the network provider updates the corresponding task process on the client. And alarm information can be sent to a terminal of a maintainer so that the maintainer can analyze the reason for the unmatched flow nodes.

For example, if it is determined that the process node information matches the task keyword according to the matching degree determined by the process node information and the task keyword, the target reply text may be determined according to the task keyword.

Illustratively, the reply text includes a field for answering an abnormal condition, and if the abnormal condition is the same or similar, the task keyword and the field in the plurality of reply texts have certain similarity. Through the similarity, target reply texts for answering task keywords corresponding to task abnormal information can be determined.

In some embodiments, the method further comprises: and determining a target database from a plurality of service class databases according to the task identifier, wherein each service class database comprises a plurality of preset reply texts.

Illustratively, the target traffic class database may be determined from a plurality of traffic class databases through task identification, wherein each traffic class database includes a plurality of preset reply texts.

It is understood that the task identifier includes a channel category and/or a customer type, and a task corresponding to the task identifier can be determined by the channel category and/or the customer type, so as to determine the target business category database.

The determining a target reply text according to the task keyword comprises: traversing all preset reply texts in the target database according to the task keywords to obtain the text matching degree of the task keywords and each preset reply text; and determining a target reply text according to the text matching degree.

Illustratively, after the target service category database is determined, traversing all preset reply texts in the target service category database through the task keywords to obtain the text matching degrees of the task keywords and the preset reply texts, wherein it can be understood that the definition of the text matching degrees can be the same as the definition of the task keywords and the flow node information, that is, the number of fields (not including punctuations) in the task keywords and the reply texts are the same.

Illustratively, the reply text corresponding to the highest text matching degree is determined as the target reply text, and the reply information is generated based on the target reply text, so as to reply the task exception information reported by the user.

And S107, generating response information according to the task identification and the target response text based on a response information generation model.

Illustratively, the response information generation model is used for splicing the task identifier and the response text to generate the response information, and as can be understood, the response information generation model may embed the task identifier into the response text and then send the response text to the client, so that the user can obtain the response information through the client, thereby solving the appeal of the user.

In some embodiments, the generating response information from the task identifier and the target response text based on a response information generation model includes: vectorizing the task identifier based on the information vectorization network of the response generation model to obtain a first vector, and vectorizing the target reply text to obtain a second vector; splicing the first vector and the second vector based on a vector splicing network of the response generation model to obtain a third vector; converting the third vector into the response information based on a conversion network of the response generation model.

Illustratively, the response generation model includes an information vectorization network, a vector splicing network and a conversion network, wherein the vectorization network may perform vectorization processing on information, the vector splicing network is used to splice vectors, and the conversion network may perform inverse vectorization processing on the vectors to obtain information indicated by the vectors.

Illustratively, the task identifier and the target response text can be vectorized through an information vectorization network to obtain a first vector and a second vector, and then the first vector and the second vector are input into a vector splicing network, and the vector splicing network splices the first vector and the second vector to obtain a third vector; and inputting the third vector into a conversion network, and converting to obtain response information, wherein the response information comprises a task identifier and a target reply text.

It can be understood that the conversion network may also be disposed in the client, the server may send the spliced third vector to the client, and the client converts the third vector through the conversion network to obtain the response information.

Illustratively, the third vector may also be encrypted, and the encrypted third vector may be sent to the client. The encryption may be performed, for example, by using a hash value or a public key, and if there is a corresponding hash value or a corresponding secret key in the client, the third vector may be decrypted and converted to obtain the response information, so as to avoid a situation that the response information is sent incorrectly, which may result in information leakage.

And step S108, sending the response information to the client.

For example, when the task exception information is obtained, the identifier of the client may be obtained, or the connection port between the client and the server is recorded, and after the response information is generated, the server sends the response information to the client through the identifier of the client or the corresponding connection port, so that the user can know the response information.

The response information is sent to the client to complete automatic response, and the response information is highly matched with the task abnormal information provided by the user, so that the probability of wrong automatic response is reduced, and the experience of the user is improved.

For example, the server may also send the response information to the blockchain, so that the client can obtain the response information from the blockchain.

In further embodiments, the response information generation model includes an interface generation network, the method further comprising: and generating a network based on the interface, marking a first identifier for the task identifier and a second identifier for the target reply text to obtain marked response information, wherein the first identifier is used for indicating the client to display the task identifier on a first interface of a display interface of the client, and the second identifier is used for indicating the client to display the target reply text on a second interface of the display interface of the client. The sending the response information to the client includes: and sending the marked response information to the client.

For example, after receiving the marked response information, the client may display the task identifier and the target response text on different display portions in a display interface corresponding to the client according to the first identifier and the second identifier in the marked response information, for example, the task identifier is displayed on the first interface of the display interface, and the target response text is displayed on the second interface of the display interface, so that the user can conveniently view the task identifier and the target response text corresponding to the task, when the user reports a plurality of task exception information, the user can quickly know which task exception information the obtained response information corresponds to, and the use experience of the user is improved.

In some embodiments, the method further comprises: generating a question text according to the task keywords; and generating response information according to the task identification, the reply text and the question text based on a response information generation model.

Illustratively, the task keywords are spliced to generate a question text, and the obtained numbers or key value pairs corresponding to the task keywords are spliced to generate the question text.

Illustratively, the response message is generated through the task identifier, the response text and the question text, and it is understood that the step of generating the response message may refer to step S107, which is not described herein.

The response information comprising the question text enables a user to check whether the reported task abnormal information is the same as the generated question text when checking the response information, so that a more accurate solution is obtained, and the problem can be seen more visually, so that the solution in the target reply text is obtained or the question text is fed back.

Illustratively, response information is generated through the task identifier, the response text and the question text, so that a user can know the task abnormal information provided by the user through the task identifier or the question text, and the abnormal condition of the task can be solved by using the obtained target response text more conveniently or the target response text has higher recognition.

According to the response information generation method provided by the embodiment, the task and the task abnormal information reported by the user can be more accurately acquired by determining the task identifier, meanwhile, through whether the task keyword is matched with the process node information or not, the error response caused by untimely updating of the client is avoided, the response information for responding to the task abnormal information can be more accurately determined by determining the reply text through the task keyword, and the efficiency of responding to the abnormal information reported by the user can be improved. The accuracy and the efficiency of automatic response can be effectively improved, and the use experience of a user is improved.

Referring to fig. 3, fig. 3 is a schematic diagram of a response information generating device according to an embodiment of the present application, where the response information generating device may be configured in a server or a terminal, and is used to execute the response information generating method.

As shown in fig. 3, the response information generating apparatus includes: the system comprises an abnormal information acquisition module 110, a task identification determination module 120, a flow node information determination module 130, a keyword extraction module 140, an information matching judgment module 150, a target reply text determination module 160, a response information generation module 170 and a response information sending module 180.

An exception information obtaining module 110, configured to obtain task exception information sent from the client.

And a task identifier determining module 120, configured to determine a task identifier of the task according to the task exception information.

And a flow node information determining module 130, configured to obtain, based on the flow management system, flow node information of the task according to the task identifier.

And the keyword extraction module 140 is configured to extract task keywords from the task abnormal information based on a keyword extraction model.

And an information matching judgment module 150, configured to judge whether the task keyword matches the process node information.

And a target reply text determining module 160, configured to determine a target reply text according to the task keyword if it is determined that the process node information matches the task keyword.

And the response information generating module 170 is configured to generate response information according to the task identifier and the target reply text based on a response information generating model.

A response message sending module 180, configured to send the response message to the client.

Illustratively, the keyword extraction module 140 further includes a traverse network sub-module and a comparison network sub-module.

And the traversal network sub-module is used for traversing all the fields in the task abnormal information based on the traversal network of the keyword extraction model to obtain the key value pairs corresponding to the fields.

And the comparison network sub-module is used for acquiring the key value pair of the preset key word, inputting the key value pair of the preset key word and the key value pair corresponding to the field in the task abnormal information into the comparison network of the key word extraction model, and determining the task key word.

Illustratively, the information matching judgment module 150 further includes a matching degree determination sub-module and a matching degree ratio sub-module.

And the matching degree determining submodule is used for determining the matching degree according to the process node information and the task keywords.

The matching degree comparison pair module is used for judging that the process node information is matched with the task keyword if the matching degree is greater than a preset matching threshold; wherein the matching degree is positively correlated with the number of the task keywords appearing in the flow node information.

Illustratively, the response information generating device further comprises a target database determining sub-module, a text matching degree determining sub-module and a target reply text determining sub-module.

And the target database determining submodule is used for determining a target database from a plurality of service class databases according to the task identifier, wherein each service class database comprises a plurality of preset reply texts.

And the text matching degree determining submodule is used for traversing all the preset reply texts in the target database according to the task keywords to obtain the text matching degree of the task keywords and each preset reply text.

And the target reply text determining submodule is used for determining the target reply text according to the text matching degree.

Illustratively, the response information generating module 170 further includes a vectorization network sub-module, a vector splicing network sub-module, and a conversion network sub-module.

And the vectorization network sub-module is used for vectorizing the task identification to obtain a first vector and vectorizing the target reply text to obtain a second vector based on the information vectorization network of the response generation model.

And the vector splicing network submodule is used for splicing the first vector and the second vector based on the vector splicing network of the response generation model to obtain a third vector.

And the conversion network sub-module is used for converting the third vector into the response information based on the conversion network of the response generation model.

Illustratively, the response information generating device further comprises an interface generating module.

And the interface generation module is used for generating a network based on the interface, marking a first identifier for the task identifier and marking a second identifier for the target reply text to obtain marked response information, wherein the first identifier is used for indicating the client to display the task identifier on a first interface of a display interface of the client, and the second identifier is used for indicating the client to display the target reply text on a second interface of the display interface of the client.

The response information sending module 180 is further configured to send the marked response information to the client.

Illustratively, the task identity determination module 120 further includes a field location determination submodule and an identity determination submodule.

And the field position determining submodule is used for determining the position relation among a plurality of fields in the task exception information.

And the identification determining submodule is used for determining the task identification of the task according to the position relation among the fields.

It should be noted that, as will be clear to those skilled in the art, for convenience and brevity of description, the specific working processes of the apparatus, the modules and the units described above may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.

The methods, apparatus, and devices of the present application are operational with numerous general purpose or special purpose computing system environments or configurations. For example: personal computers, server computers, hand-held or portable devices, tablet-type devices, multiprocessor systems, microprocessor-based systems, set top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments that include any of the above systems or devices, and the like.

The above-described methods and apparatuses may be implemented, for example, in the form of a computer program that can be run on a computer device as shown in fig. 4.

Referring to fig. 4, fig. 4 is a schematic block diagram of a computer device according to an embodiment of the present disclosure. The computer device may be a server or a terminal.

As shown in fig. 4, the computer device includes a processor, a memory, and a network interface connected by a system bus, wherein the memory may include a storage medium and an internal memory.

The storage medium may store an operating system and a computer program. The computer program includes program instructions that, when executed, cause a processor to perform any one of the response information generating methods.

The processor is used for providing calculation and control capability and supporting the operation of the whole computer equipment.

The internal memory provides an environment for running a computer program in the storage medium, which, when executed by the processor, causes the processor to perform any one of the response information generating methods.

The network interface is used for network communication, such as sending assigned tasks and the like. Those skilled in the art will appreciate that the architecture shown in fig. 4 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.

It should be understood that the Processor may be a Central Processing Unit (CPU), and the Processor may be other general purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) or other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components, etc. Wherein a general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.

Wherein, in one embodiment, the processor is configured to execute a computer program stored in the memory to implement the steps of:

acquiring task exception information sent from a client;

and sending the response information to the client.

In one embodiment, the processor, when implementing a keyword extraction model based on extracting task keywords from the task exception information, is configured to implement:

traversing all fields in the task abnormal information based on a traversing network of the keyword extraction model to obtain key value pairs corresponding to the fields;

and acquiring key value pairs of preset key words, inputting the key value pairs of the preset key words and the key value pairs corresponding to the fields in the task abnormal information into a comparison network of the key word extraction model, and determining the task key words.

In one embodiment, the processor, when determining whether the task keyword matches the process node information, is configured to:

determining matching degree according to the process node information and the task keywords;

if the matching degree is larger than a preset matching threshold value, judging that the process node information is matched with the task keyword;

wherein the matching degree is positively correlated with the number of the task keywords appearing in the flow node information.

In one embodiment, when implementing the response information generating method, the processor is configured to implement:

determining a target database from a plurality of service class databases according to the task identifier, wherein each service class database comprises a plurality of preset reply texts;

the processor, in implementing determining a target reply text from the task keyword, is configured to implement:

traversing all preset reply texts in the target database according to the task keywords to obtain the text matching degree of the task keywords and each preset reply text;

and determining a target reply text according to the text matching degree.

In one embodiment, the processor, in implementing a response message generation model based on response message generation, is configured to implement:

vectorizing the task identification to obtain a first vector and vectorizing the target reply text to obtain a second vector based on the information vectorization network of the response generation model;

splicing the first vector and the second vector based on a vector splicing network of the response generation model to obtain a third vector;

converting the third vector into the response information based on a conversion network of the response generation model.

generating a network based on the interface, marking a first identifier for the task identifier and a second identifier for the target reply text to obtain marked response information, wherein the first identifier is used for indicating the client to display the task identifier on a first interface of a display interface of the client, and the second identifier is used for indicating the client to display the target reply text on a second interface of the display interface of the client;

when the processor sends the response information to the client, the processor is configured to:

and sending the marked response information to the client.

In one embodiment, when the processor determines the task identifier of the task according to the task exception information, the processor is configured to:

determining the position relation among a plurality of fields in the task exception information;

and determining the task identifier of the task according to the position relation among the fields.

It should be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working process of generating the response information may refer to the corresponding process in the foregoing embodiment of the response information generation control method, and is not described herein again.

Embodiments of the present application further provide a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, where the computer program includes program instructions, and a method implemented when the program instructions are executed may refer to various embodiments of the method for generating response information in the present application.

The computer-readable storage medium may be an internal storage unit of the computer device described in the foregoing embodiment, for example, a hard disk or a memory of the computer device. The computer readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like provided on the computer device.

It is to be understood that the terminology used in the description of the present application herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used in the specification of the present application and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.

It should also be understood that the term "and/or" as used in this specification and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items. It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or system. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or system that comprises the element.

The above-mentioned serial numbers of the embodiments of the present application are merely for description and do not represent the merits of the embodiments. While the invention has been described with reference to specific embodiments, the scope of the invention is not limited thereto, and those skilled in the art can easily conceive various equivalent modifications or substitutions within the technical scope of the invention. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. A response information generating method, comprising:

acquiring task exception information sent from a client;

and sending the response information to the client.

2. The response information generating method according to claim 1, wherein the extracting a task keyword in the task abnormality information based on a keyword extraction model includes:

3. The response information generating method according to claim 2, wherein the determining whether the task keyword matches the flow node information includes:

4. A response message generation method according to any one of claims 1 to 3, characterized in that the method further comprises:

the determining a target reply text according to the task keyword comprises:

and determining a target reply text according to the text matching degree.

5. The answer information generation method of any one of claims 1-3, wherein the generating answer information from the task identifier and the target answer text based on an answer information generation model comprises:

6. The response message generation method according to any one of claims 1 to 3, wherein the response message generation model includes an interface generation network, and the generating of the response message based on the response message generation model from the task identification and the target reply text includes:

the sending the response information to the client includes:

and sending the marked response information to the client.

7. The method for generating response information according to claim 1, wherein the determining a task identifier of a task according to the task exception information includes:

8. A response information generating apparatus, characterized in that the response information generating apparatus comprises:

9. A computer arrangement, characterized in that the computer arrangement comprises a processor, a memory, and a computer program stored on the memory and executable by the processor, wherein the computer program, when executed by the processor, carries out the steps of the reply information generation method according to any of claims 1 to 7.

10. A computer-readable storage medium, characterized in that a computer program is stored thereon, wherein the computer program, when being executed by a processor, carries out the steps of the response information generating method according to any one of claims 1 to 7.