CN112528002B

CN112528002B - Dialogue identification method, device, electronic equipment and storage medium

Info

Publication number: CN112528002B
Application number: CN202011545960.8A
Authority: CN
Inventors: 王猛; 赵筱军; 黄庆伟
Original assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Current assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Priority date: 2020-12-23
Filing date: 2020-12-23
Publication date: 2023-07-18
Anticipated expiration: 2040-12-23
Also published as: CN112528002A

Abstract

The application discloses a dialogue identification method, a dialogue identification device, electronic equipment and a storage medium, and relates to the field of Natural Language Processing (NLP). The implementation scheme is as follows: acquiring a target dialogue in at least one round of dialogue, and determining a target problem set to be identified from the at least one round of dialogue; performing entity word recognition on the target dialogue to obtain and store a plurality of entity words in the target dialogue; matching each question in the target question set with a plurality of stored entity words respectively to determine the entity word corresponding to each question; and determining answers of the questions according to the entity words corresponding to the questions. Therefore, when the user answers a plurality of questions in one round of dialogue, the system can automatically recognize the answers of the questions, unnecessary interaction times can be reduced, and the use experience of the user is improved.

Description

Dialogue identification method, device, electronic equipment and storage medium

Technical Field

The application relates to the technical field of AI (Artificial Intelligence ) such as NLP (Natural Language Processing, natural language processing), and in particular provides a dialog recognition method, a dialog recognition device, an electronic device and a storage medium.

Background

With the continuous evolution of human society informatization and the continuous rise of human service cost, people increasingly want to communicate with computers through natural language, and man-machine conversation systems are a product of birth in such a historical background. By means of a man-machine dialogue system, a human can use natural language to conduct dialogue with a machine, and instruct or consult a computer through the dialogue to complete specific operations, such as instructing intelligent hardware to complete short message reading and replying, inquiring weather, renting cars, booking tickets, scheduling routes and the like.

Disclosure of Invention

The application provides a method, a device, electronic equipment and a storage medium for dialog identification.

According to an aspect of the present application, there is provided a dialog recognition method, including:

acquiring a target dialogue in at least one round of dialogue, and determining a target problem set to be identified from the at least one round of dialogue;

performing entity word recognition on the target dialogue to obtain and store a plurality of entity words in the target dialogue;

matching each question in the target question set with a plurality of stored entity words respectively to determine the entity word corresponding to each question;

And determining answers of the questions according to the entity words corresponding to the questions.

According to another aspect of the present application, there is provided a dialogue identifying device, including:

the dialogue acquisition module is used for acquiring a target dialogue in at least one round of dialogue;

the set determining module is used for determining a set of target problems to be identified from the at least one round of dialogue;

the recognition module is used for recognizing entity words of the target dialogue, obtaining and storing a plurality of entity words in the target dialogue;

the matching module is used for respectively matching each question in the target question set with a plurality of stored entity words so as to determine the entity word corresponding to each question;

and the answer determining module is used for determining the answer of each question according to the entity word corresponding to each question.

According to still another aspect of the present application, there is provided an electronic apparatus including:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein,,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the dialog recognition method set forth in the above embodiments of the present application.

According to yet another aspect of the present application, there is provided a non-transitory computer-readable storage medium of computer instructions for causing the computer to perform the dialog recognition method proposed by the above-described embodiments of the present application.

According to a further aspect of the present application, a computer program product is provided, which, when instructions in the computer program product are executed by a processor, performs a dialog recognition method as set forth in the above embodiments of the present application.

It should be understood that the description of this section is not intended to identify key or critical features of the embodiments of the application or to delineate the scope of the application. Other features of the present application will become apparent from the description that follows.

Drawings

The drawings are for better understanding of the present solution and do not constitute a limitation of the present application. Wherein:

FIG. 1 is a schematic diagram of a partially continuous entity collection unit configuration;

FIG. 2 is a flow chart of a dialogue recognizing method according to an embodiment of the present application;

fig. 3 is a flow chart of a dialogue identification method according to a second embodiment of the present application;

fig. 4 is a flow chart of a dialogue identification method provided in the third embodiment of the present application;

Fig. 5 is a flow chart of a dialogue identification method according to a fourth embodiment of the present application;

fig. 6 is a schematic structural diagram of a dialogue identifying device provided in a fifth embodiment of the present application;

FIG. 7 shows a schematic block diagram of an example electronic device that may be used to implement embodiments of the present application.

Detailed Description

Exemplary embodiments of the present application are described below in conjunction with the accompanying drawings, which include various details of the embodiments of the present application to facilitate understanding, and should be considered as merely exemplary. Accordingly, one of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the present application. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.

With the continuous development of natural language processing technology, the multi-round dialogue technology is increasingly applied to intelligent customer service, intelligent outbound and other scenes, and the cost of enterprises is greatly reduced while the service efficiency of clients is improved.

At present, the multi-round dialogue system is based on a slot filling method (called slot filling method for short) or an improved version of the slot filling method, realizes multi-round dialogue according to a series of process node jump control of a dialogue tree, and completes interaction of dialogue according to a business process of a fixed order for collecting entity words. However, in the actual conversation process, the user replies to the expressions collected by the entity words, so that some other entity word information is easily carried as the following information of the conversation according to the conversation habit, and the following information is often randomly disordered. For example, in a car renting scene, the entity word collection flow preset by the man-machine dialogue system is three entity attributes of query location, car type and time in sequence, when the first entity word is collected, the user is likely to make a pre-judgment on the entity word to be collected by the subsequent collection node according to the context of the current dialogue and carry the pre-judged entity word, or some entity words to be collected subsequently may be recorded in subconsciousness after several complete dialogue interactions. For example, the human-computer interaction system issues a question: "ask where to rent? "user reply: "Peking upper me rents large cars" or "open sky Peking upper me rents large cars".

However, the existing man-machine dialogue system can only recognize a problem in one round of dialogue, for example, can only recognize "Beijing" in the above example, but can not recognize two entity words of time ("tomorrow") and vehicle type ("large vehicle"), in this way, even if the user carries other entity words (such as tomorrow and large vehicle) in one problem, the man-machine dialogue system can not recognize, so that multiple rounds of dialogue interaction can only be completed according to a business process of a fixed order to recognize all entity words.

That is, in the prior art, only one question can be identified by one dialog, and the method is not suitable for the situation that the user answers a plurality of questions simultaneously in one dialog, so that the interaction times of the user are increased, and the use experience of the user is reduced.

Therefore, the method mainly aims at the technical problems that in the prior art, only one question can be identified by one round of dialogue, and the method cannot be suitable for the situation that a user answers a plurality of questions simultaneously in one round of dialogue, so that the interaction times of the user are increased, and the use experience of the user is reduced.

According to the dialogue identification method, a target dialogue in at least one round of dialogue is obtained, and a target problem set to be identified in the at least one round of dialogue is determined; performing entity word recognition on the target dialogue to obtain and store a plurality of entity words in the target dialogue; matching each question in the target question set with a plurality of stored entity words respectively to determine the entity word corresponding to each question; and determining answers of the questions according to the entity words corresponding to the questions. Therefore, when the user answers a plurality of questions in one round of dialogue, the system can automatically recognize the answers of the questions, unnecessary interaction times can be reduced, and the use experience of the user is improved.

The following describes a dialogue recognition method, a dialogue recognition device, an electronic device and a storage medium according to embodiments of the present application with reference to the accompanying drawings. Before describing the embodiments of the present invention in detail, for ease of understanding, the general technical words are first introduced:

intent refers to a business operation that a user is to perform, such as renting cars, querying weather, booking tickets, etc., with intent similar to a function in the code. The intention can be divided into a top-level intention and a sub-intention, wherein the top-level intention is the intention which can be triggered at any time in the conversation process, and the sub-intention can be triggered only when in a corresponding scene.

A scene consists of one intention and all dialog interactions under that intention (e.g., collect entity words, clarify, confirm, sub-intents, etc.), and only when the user expresses a new intention or ends a dialog, the scene is switched or ended.

The entity words refer to parameters required for completing business actions, such as time, place, vehicle type and the like, the entity words are similar to parameters in functions, and one intention plus a plurality of entity words can complete business transaction.

The conversation, a series of conversations between the same user and the man-machine conversation system in a certain time period is a conversation, such as a telephone in telephone customer service.

The dialogue, the user and the man-machine dialogue system can be a round of dialogue, namely, the round of dialogue comprises a question and the answer or answer corresponding to the question.

The entity collecting unit refers to node configuration required by completing an entity word collecting flow in the conversation process. The entity collecting unit generally comprises a father node and two child nodes, wherein the father node and the child nodes are nodes in a dialogue tree, the father node is used for judging whether entity words to be collected are collected by dialogue contexts as node selection entering conditions, if the entity words are not collected by dialogue contexts, the first child node judges whether contents expressed by users contain entity words, if the entity words are contained, the collection is completed by assigning values of variables in the contexts, and if the entity words are not contained, the second child node is executed and jumps to the father node again to initiate entity word collection, and the whole collecting flow consists of three nodes which are closed loops until the entity word collection is completed.

And the entity collection unit set consists of entity collection units. The entity collection unit set includes a plurality of continuous entity collection units, i.e., the number of continuous entity collection units included in the entity collection unit set is greater than 1. For example, referring to fig. 1, fig. 1 is a schematic diagram of a partially continuous entity collection unit set flow configuration. Assuming that the fourth entity collecting unit from the left is an illegal entity collecting unit, the first, second and third entity collecting units from the left form an entity collecting unit set, and only in the same entity collecting unit set, entity word information temporarily stored in the memory in the conversation process is effectively available for entity words to be collected in the current entity collecting unit set, and memories among different entity collecting unit sets are independent.

Wherein the dialogue tree is layered according to parent-child relationship, and the top node in fig. 1 refers to the first parent node of the dialogue tree.

For example, if the entity collection unit set 1 is applied to a car rental scene, the entity collection unit set 1 may identify entity words in each dialogue in the car rental scene, and if the entity collection unit set 2 is applied to a query weather scene, the entity collection unit set 2 may identify entity words in each dialogue in the query weather scene. The memory between the entity collection unit set 1 and the entity collection unit set 2 is independent, and entity words temporarily stored in the memory are not influenced.

And the entity collecting unit detector is used for searching whether a legal entity collecting unit set exists in the layer where the current node is located, stopping detection when the legal entity collecting unit set does not exist, and storing the detected entity collecting unit set and the detected node track information after detection is finished each time.

The validity of the entity collection unit set may be determined according to the number of continuous entity collection units included in the entity collection unit set, where when the number of continuous entity collection units included in the entity collection unit set is greater than 1, the entity collection unit set is determined to be legal, and when the number of continuous entity collection units included in the entity collection unit set is less than or equal to 1, the entity collection unit set is determined to be illegal.

As shown in fig. 1, a complete detection process is that the first entity collection unit of the left number starts to detect from the right, and since the fourth entity collection unit of the left number is an illegal entity collection unit structure, the first detection result is an entity collection unit set formed by the first entity collection unit of the left number, the second entity collection unit of the left number and the third entity collection unit of the left number, the entity collection unit set information and the detection track node information are recorded, the second detection process starts from the fifth entity collection unit of the left number, and since only one legal entity collection unit is not satisfied, the second detection process does not detect the entity collection unit set, and only the detection track node information is recorded.

The entity collection unit set formed by the first entity collection unit, the second entity collection unit and the third entity collection unit supports intelligent collection of entity words when the entity words are collected, and the fifth entity collection unit (without forming the entity collection unit set which is continuous and has the number larger than 1) does not support intelligent collection of the entity words. And the fifth entity collecting unit of left number can not use the entity word information temporarily stored in the dialogue domain memory of the entity collecting unit set formed by the first entity collecting unit of left number, the second entity collecting unit of left number and the third entity collecting unit of left number when collecting entity words.

Fig. 2 is a flowchart of a dialog recognition method according to an embodiment of the present application.

The embodiment of the application is exemplified by the dialogue identification method being configured in the dialogue identification device, and the dialogue identification device can be applied to any electronic equipment so that the electronic equipment can execute dialogue identification functions.

The electronic device may be any device with computing capability, for example, a PC (Personal Computer ), a mobile terminal, a server, and the like, and the mobile terminal may be, for example, a mobile phone, a tablet computer, a personal digital assistant, a wearable device, an in-vehicle device, a robot, and other hardware devices with various operating systems, touch screens, and/or display screens.

As shown in fig. 2, the dialog recognition method may include the steps of:

step 201, obtaining a target dialogue in at least one round of dialogue, and determining a target problem set to be identified from the at least one round of dialogue.

In the embodiment of the application, the target problem set is related to the business handled by the user. For example, when a user rents a car, the target set of questions contains questions related to the car, e.g., the target set of questions may be { "please ask where car is to be rented? "please ask what model of car to rent? "," ask when to rent? "}. Alternatively, when a user queries for weather, the target set of questions contains various questions related to the queried weather, such as the target set of questions may be { "please ask where weather is to be queried? "please ask when to check weather? "}. Alternatively, when the user orders a ticket, such as a flight reservation, the target set of questions may be { "what day of flights you want to reserve? "do your departure time? "do you go to city? "do you arrive at city? "}.

It should be appreciated that in order to identify the user's intent, and thereby perform an operation that matches the user's intent, in embodiments of the present application, the target dialog may comprise individual dialogs that the user has answered in the current session. For example, when the human-computer interaction system presents a question of "please ask where to rent? ", the answer of the user is: "Beijing help me rent large car in tomorrow", only one round of dialogue is acquired, and the dialogue is "please ask which car to rent? "dialog is hereinafter" tomorrow rents me large car in Beijing group ".

When the target dialogue is each dialogue which contains the answer of the user in the current dialogue, the target question set is the question set which contains the questions in the target dialogue.

In one possible implementation of the embodiments of the present application, a user intent may be identified, and a set of target questions is determined based on the user intent. And asking the user in sequence according to each question in the target question set, and taking each dialogue containing the answer of the user as a target dialogue.

An example, controls for handling business, such as car renting, weather inquiring, ticket purchasing and the like, can be displayed on the interface of the man-machine interaction system, so that a user can enter a corresponding business handling page by triggering the corresponding controls, the man-machine interaction system can determine the intention of the user based on the corresponding business, determine a target question set according to the intention of the user, and ask questions to the user in sequence based on the questions in the target question set, and accordingly the user can answer, and each dialogue containing the answer of the user is used as a target dialogue.

As another example, a user may directly input a service to be transacted (e.g., renting a car, querying weather, purchasing a ticket, etc.), where the input manner includes, but is not limited to, touch input (e.g., sliding, clicking, etc.), keyboard input, voice input, etc., so that the user intent may be identified from the service input by the user, a set of target questions may be determined from the user intent, and questions may be asked to the user in sequence based on the questions in the set of target questions, so that the user may make an answer, and each dialog including the answer of the user may be used as the target dialog.

The user intention may be identified based on a semantic analysis technology in an NLP technology in the AI field, or may be identified based on a template matching method, a text classification method, or the like in the NLP technology, which is not limited in this application.

Among them, AI is a discipline of researching and making a computer simulate some thinking process and intelligent behavior (such as learning, reasoning, thinking, planning, etc.) of a person, and has a technology at both hardware and software level. Artificial intelligence hardware technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing, and the like; the artificial intelligence software technology mainly comprises a computer vision technology, a voice recognition technology, a natural language processing technology, a machine learning/deep learning technology, a big data processing technology, a knowledge graph technology and the like.

In another possible implementation manner of the embodiment of the present application, a question set may be extracted from a plurality of question sets according to a set rule, and used as a target question set, and questions are sequentially asked to the user according to questions in the target question set, and each dialog including the answer of the user is used as a target dialog.

For example, taking a set rule as an example in sequence, N question sets in the man-machine interaction system are all selected, for example, a first question set may be selected first, or a question set may be selected randomly, and questions may be asked to the user in sequence according to each question in the selected question set. After the user answers m questions, assuming that the currently selected question set is a t-th question set, the t-th question set can be used as a target question set, questions can be sequentially asked to the user according to questions in the target question set, and each dialogue containing the answer of the user is used as a target dialogue.

For another example, taking a set rule as a correlation between scenes as an example, determining that a last scene is a booking scene according to a user intention, possibly involving local tourism, car renting, accommodation, etc. after booking, the current scene may be a car renting, accommodation, etc., after determining the current scene, a target problem set corresponding to the current scene may be directly determined, for example, the target problem set may be determined according to the intention corresponding to the current scene, so that each question in the target problem set may be sequentially asked to the user, and each dialog including the answer of the user may be regarded as a target dialog.

Step 202, entity word recognition is performed on the target dialogue, and a plurality of entity words in the target dialogue are obtained and stored.

In the embodiment of the application, the entity word recognition can be performed on the target dialogue, so that a plurality of entity words in the target dialogue are obtained, and the entity words are stored.

For example, the target dialogue may be subjected to entity word recognition based on keyword matching, template matching, statistical model and other methods in the NLP technology, so as to obtain a plurality of entity words in the target dialogue. It should be noted that, the above-mentioned entity word recognition method is only an exemplary embodiment, but not limited thereto, and includes other entity word recognition methods known in the art, as long as the entity word in the target dialogue can be obtained, that is, the method for recognizing the entity word is not limited in the present application.

Step 203, matching each question in the target question set with a plurality of stored entity words respectively, so as to determine the entity word corresponding to each question.

In this embodiment of the present application, after obtaining a plurality of entity words in a target dialogue, each question in a target question set may be respectively matched with a plurality of entity words in the target dialogue, so as to determine an entity word corresponding to each question.

As an application scenario, when the dialogue identification method is applied to a car rental scenario, a target problem set to be identified is as follows: { "please ask where car rental is to be? "please ask what model of car to rent? "," ask when to rent? "}, in the first dialog, the question sent by the man-machine dialog system is: "ask where to rent? ", assuming that the answer of the user is: "Beijing upper me rents large cars in tomorrow", therefore, it can be determined that the entity word corresponding to question 1 ("please ask what car should be rented in.

Step 204, determining answers to the questions according to the entity words corresponding to the questions.

In the embodiment of the present application, after determining the entity word corresponding to each question, the answer of each question may be determined according to the entity word corresponding to each question. For example, based on the slot filling method, the entity words can be filled into the corresponding slots to obtain answers to the corresponding questions.

Still as an example of the above, the answer to question 1 may be obtained as: "rent in Beijing", answer corresponding to question 2 is "rent large car", answer corresponding to question 3 is "rent time is tomorrow". Therefore, the man-machine dialogue system can execute corresponding business operation according to answers corresponding to all the questions, and the user is helped to rent large-sized vehicles in Beijing, and the time of renting the vehicles is tomorrow. Therefore, when the user answers a plurality of questions in one round of dialogue, the man-machine interaction system can automatically recognize the answers of the questions, unnecessary interaction times can be reduced, the business handling efficiency of the user is improved, and the use experience of the user is improved.

According to the dialogue identification system, a target dialogue in at least one round of dialogue is acquired, and a target problem set to be identified in the at least one round of dialogue is determined; performing entity word recognition on the target dialogue to obtain and store a plurality of entity words in the target dialogue; matching each question in the target question set with a plurality of stored entity words respectively to determine the entity word corresponding to each question; and determining answers of the questions according to the entity words corresponding to the questions. Therefore, when the user answers a plurality of questions in one round of dialogue, the system can automatically recognize the answers of the questions, unnecessary interaction times can be reduced, and the use experience of the user is improved.

In one possible implementation manner of the embodiment of the present application, the entity words in the target dialogue may be identified through the entity collecting unit, so as to determine the entity words corresponding to each problem. The following describes the above process in detail with reference to the second embodiment.

Fig. 3 is a flow chart of a dialog recognition method according to a second embodiment of the present application.

As shown in fig. 3, the dialog recognition method may include the steps of:

step 301, a target session in at least one round of sessions is acquired.

The execution of step 301 may refer to the execution of step 201 in the above embodiment, which is not described herein.

Step 302, determining a target entity collection unit set from a plurality of entity collection unit sets corresponding to at least one round of dialogue according to a set rule; the target entity collection unit set comprises one or more entity collection units corresponding to the questions in the target question set, and the entity collection units are used for identifying entity words matched with the corresponding questions.

In this embodiment of the present application, the setting rule is preset, for example, the setting rule may be sequential, or the setting rule may be other, which is not limited in this application.

In this embodiment of the present invention, entity words in a conversation process may be collected or identified through a plurality of entity collection unit sets, so after at least one conversation is acquired, a plurality of entity collection unit sets corresponding to the at least one conversation may be determined first, and based on a set rule, a target entity collection unit set is determined from a plurality of entity collection unit sets corresponding to the at least one conversation, where the target entity collection unit set may include one or more entity collection units corresponding to a problem in a target problem set, and each entity collection unit is configured to identify entity words matched with the corresponding problem.

For example, taking the rule set as an example in sequence, assume that the plurality of entity collection units corresponding to at least one round of dialogue are entity collection unit set 1, entity collection unit set 2, and entity collection unit set 3, respectively. Assuming that the current scene is a ticket booking, the subsequent scenes can discuss local renting, accommodation and travel, so that the entity word recognition can be performed on the dialogs in the current scene according to the entity collecting units in the entity collecting unit set 1, the entity word recognition can be performed on the dialogs in the renting scenes by using the entity collecting units in the entity collecting unit set 2, and the entity word recognition can be performed on the dialogs in the accommodation scene by using the entity collecting units in the entity collecting unit set 3.

The above description is merely given by way of example in order of setting rules, and the present application is not limited thereto, and the setting rules may be related to user intention, for example, entity word recognition may be performed on dialogs of a booking scene by each entity collecting unit in the entity collecting unit set 1, entity word recognition may be performed on dialogs of a renting scene by each entity collecting unit in the entity collecting unit set 2, entity word recognition may be performed on dialogs of an accommodation scene by each entity collecting unit in the entity collecting unit set 3, and entity word recognition may be performed on dialogs of a query weather scene by each entity collecting unit in the entity collecting unit set 4. Assuming that at least one round of dialogue is subjected to intention recognition, and the intention of the dialogue is determined to be a car renting, the target entity collection unit set may be the entity collection unit set 2, so that entity word recognition can be performed on the dialogue in a car renting scene through each entity collection unit in the entity collection unit set 2.

In the embodiment of the application, one or more entity collecting units corresponding to the problems in the target problem set, which are contained in the target entity collecting unit set, are adopted to identify entity words matched with the corresponding problems. Because the entity collecting units are configured with the nodes required by one entity word collecting flow in the conversation process, the accuracy of the entity word identifying result can be improved based on the fact that each entity collecting unit identifies the entity word matched with the corresponding problem.

Step 303, entity word recognition is performed on the target dialogue, and a plurality of entity words in the target dialogue are obtained and stored.

The execution of step 303 may refer to the execution of step 202 in the above embodiment, which is not described herein.

And 304, sequentially identifying a plurality of entity words in the target dialogue by adopting each entity collecting unit in the target entity collecting unit set to obtain entity words matched with the problems corresponding to each entity collecting unit.

In this embodiment, when the target entity collection unit set includes a plurality of entity collection units, each entity collection unit in the target entity collection unit set may be used to sequentially identify a plurality of entity words in the target session, so as to obtain entity words matched with problems corresponding to each entity collection unit.

Step 305, determining answers to the questions according to the entity words corresponding to the questions.

The execution of step 305 may refer to the execution of step 204 in the above embodiment, and will not be described herein.

In the embodiment of the application, the entity collecting units are configured with the nodes required by one entity word collecting flow in the conversation process, so that the accuracy of the entity word recognition result can be improved based on the fact that each entity collecting unit recognizes the entity word matched with the corresponding problem.

In a possible implementation manner of the embodiment of the present application, when all entity collecting units in the target entity collecting unit set are identified, the use of the target entity collecting unit set for identification may be stopped, or when there is an entity collecting unit that does not identify a matching entity word, the use of the target entity collecting unit set for identification may be stopped. Therefore, after the entity words contained in the target dialogue are identified, the adoption of the target entity collection unit set is stopped to continue the identification, and the processing pressure of the system can be reduced.

It may be understood that when there is an entity collection unit that does not recognize a matching entity word, it indicates that the user does not answer all the questions in the target question set, and at this time, the user cannot be helped to complete the corresponding service, so, as a possible implementation manner of the embodiment of the present application, in the case that there is an entity collection unit that does not recognize a matching entity word, the following text of the target dialogue may be obtained from at least one round of dialogue, the following entity word may be recognized and stored, and the target entity collection unit set may be adopted to continuously recognize each stored entity word until all entity collection units in the target entity collection unit set recognize the entity word that is matched with the corresponding question.

The above process will be described in detail with reference to the third embodiment.

Fig. 4 is a flow chart of a dialogue identification method provided in the third embodiment of the present application.

As shown in fig. 4, the dialog recognition method may include the steps of:

step 401, obtaining a target dialogue in at least one round of dialogue.

Step 402, determining a target entity collection unit set from a plurality of entity collection unit sets corresponding to at least one round of dialogue according to a set rule; the target entity collection unit set comprises one or more entity collection units corresponding to the questions in the target question set, and the entity collection units are used for identifying entity words matched with the corresponding questions.

Step 403, performing entity word recognition on the target dialogue, obtaining and storing a plurality of entity words in the target dialogue.

And step 404, sequentially identifying a plurality of entity words in the target dialogue by adopting each entity collecting unit in the target entity collecting unit set to obtain entity words matched with the problems corresponding to each entity collecting unit.

The execution of steps 401 to 404 may be referred to the execution of the above embodiment, and will not be described herein.

In step 405, if there is an entity collection unit that does not recognize a matching entity word, the recognition with the target entity collection unit set is stopped.

For example, the current scenario is a rental car scenario, the target set of questions is { "please ask where to rent? "please ask what model of car to rent? "," ask when to rent? "when the human-computer interaction system presents a question of" please ask where to rent? ", the answer of the user is: "in Beijing help me rent large car", it can be determined that the entity word corresponding to question 1 ("please ask where car is to be rented. Thus, the recognition of the target session by the target entity collecting unit may be stopped.

Step 406, obtaining the context of the target session from at least one round of sessions.

It may be understood that when there is an entity collecting unit that does not recognize a matching entity word, it indicates that the user does not answer all questions in the target question set, and at this time, the user cannot be helped to complete the corresponding service, so, as a possible implementation manner of the embodiment of the present application, in the case that there is an entity collecting unit that does not recognize a matching entity word, the following text of the target dialogue may also be obtained from at least one round of dialogue.

Step 407, storing the entity words in the following.

In the embodiment of the application, the entity word can be identified below, and the identified entity word can be stored.

And 408, continuing to identify each stored entity word by adopting the target entity collection unit set.

In this embodiment of the present application, a target entity collection unit set may be used to identify each stored entity word continuously, that is, each entity collection unit in the target entity collection unit set is used to identify each stored entity word in sequence, so as to obtain an entity word matched with a problem corresponding to each entity collection unit. Therefore, the entity words matched with the questions in the target question set can be identified, and the answers of the questions can be determined later, so that the user can be helped to accurately handle the corresponding business.

Still with the above example, when the human-machine interaction system presents a question in the context of a target dialogue that "please ask when to rent a car? ", the answer of the user is: and if the entity word is "tomorrow", the target entity collection unit set is adopted to continuously identify each stored entity word, and the entity word corresponding to the problem 3 can be determined as "tomorrow".

Step 409, determining answers to the questions according to the entity words corresponding to the questions.

The execution of step 409 may be referred to the execution of the above embodiment, and will not be described herein.

In one possible implementation manner of the embodiment of the present application, when each entity collecting unit in the target entity collecting unit set identifies a matching entity word, the stored plurality of entity words may be deleted. Therefore, when the entity words matched with the problems are identified, the stored entity words are deleted, so that on one hand, the occupation of storage resources can be reduced, and on the other hand, other dialogs can be identified by adopting the target entity collecting unit, the stored entity words are deleted, and the influence on the identification results of the other dialogs can be avoided.

For example, the booking scene may relate to time identification, while the accommodation and taxi scene may also relate to time identification, and assuming that the target dialogue scene is the booking scene, the stored entity words may include time, after identifying the entity words matched with each question in the target question set, if the stored entity words are not deleted, the stored entity words will occur in the accommodation scene or taxi scene, and when there is no entity word in the dialogue, the stored time is used as the entity words identified by the target entity collection unit set, which will cause misjudgment, and seriously reduce the user experience of the user.

For example, the entity words stored in the booking scene are: booking time, departure city and arrival city, and the entity words obtained by recognition in the renting scene are as follows: if the entity words stored in the ticket booking scene are not deleted, the departure time in the ticket booking scene is identified as the ticket booking time in the ticket booking scene, and obviously, the identification result is incorrect, so that the stored entity words can be deleted under the condition that each entity collecting unit in the target entity collecting unit set identifies the matched entity words, on one hand, the accuracy of the identification result can be improved, and on the other hand, the occupation of storage resources can be reduced.

As an example, referring to fig. 5, fig. 5 is a flow chart of a dialog recognition method provided in the fourth embodiment of the present application.

1. The user may input a query sentence (query), wherein the input means includes, but is not limited to, touch input, keyboard input, voice input, etc.;

2. the man-machine interaction system can conduct intention recognition and entity word recognition on the query;

3. entering scenes matched with the intention of the user, such as a ticket booking scene, a renting scene and the like;

4. Judging whether the entity word recognition result is empty, and under the condition that the entity word recognition result is empty, indicating that the round of dialogue does not recognize the entity word, thus directly ending and executing the processing flow corresponding to the current dialogue node; and executing the step 5 under the condition that the entity word recognition result is not null;

5. judging whether the current dialogue node is detected by an entity collecting detector, if not, starting the entity collecting detector, searching whether a legal entity collecting unit set exists in the layer where the current dialogue node is located by using the entity collecting detector, temporarily storing a detection result, and continuously executing the step 6; under the condition of being detected by the entity collecting detector, no processing is carried out, and the step 6 is continuously executed;

6. judging whether the current dialogue node belongs to a node in a certain entity collecting unit in the entity collecting unit set detected by the entity collecting detector, and under the condition that the current dialogue node does not accord with the entity collecting unit configuration rule, the intelligent collection of the entity cannot be carried out, the intelligent collection of the dialogue entity words is finished in advance, and executing the corresponding processing flow of the current dialogue node; under the condition of belonging to the category, the whole recognition result of the round of entity words is selectively temporarily stored based on the current scene, for example, when the entity words are wound, for example, the mobile phone number and the digital entity words are wound, at the moment, the entity words which are wound are required to be independently collected, the entity words which are wound are not temporarily stored, the temporary storage process is a selective memory temporary storage process, and the step 7 is continuously executed;

7. Judging whether an entity collecting unit corresponding to the current dialogue node exists or not, and judging whether entity words temporarily stored in a memory in the step 6 contain entity words which need to be collected by the entity collecting unit existing currently or not;

under the condition that the two conditions are met, performing entity word collection of the current entity collection unit, pointing the current entity collection unit to the next entity collection unit in the entity collection unit set detected by the entity collection detector in the step 5, and continuing to perform the step 7 to perform the entity word collection process of the next entity collection unit until the condition of the step 7 is not met. And if the two conditions are not satisfied at the same time, continuing to execute the step 8;

8. judging whether the entity collecting unit set detected by the entity collecting detector in the step 5 completely completes entity word collection in the step 7; under the condition that the collection of the entity words is completed completely, all temporary entity words in the step 6 are emptied, at the moment, the collection of the entity collection unit set detected in the step 5 is completed completely, and the temporarily stored entity word information has no meaning, so that the entity words can be emptied, the intelligent collection of the entity words is completed completely, and answers of corresponding dialogue nodes are replied; and the entity collecting unit is used for collecting incomplete entity words in the entity collecting unit set, so that the intelligent collection of the dialogue entity words is finished, and answers of corresponding dialogue nodes are replied.

Therefore, in a certain scene of a multi-round dialogue, automatic collection of a plurality of continuous entity words can be supported, the intelligent collection process of the entity words has flexible short-term memory capacity for the dialogue, so that more effective dialogue information can be captured in the dialogue, the intelligent collection of the entity words has the memory capacity in the dialogue, the unnecessary interaction times are effectively reduced, the intelligence and humanization of the dialogue capacity are improved, and meanwhile, the business handling efficiency and experience effect of users are improved.

It should be noted that in the prior art, intelligent collection of entity words can be achieved through complex configuration, for example, at each collection node, the object is achieved through a pre-collection means by implanting the entity to be collected later in advance in a manner of "hidden buried points". For example, while collecting the place, it is determined whether there are collectable vehicle types and time entity words, and if there are any, it is collected in advance, the next collection node and so on.

However, the above-mentioned entity word collection manner introduces additional configuration cost, and too much and complicated configuration is not beneficial to expansion maintenance, even a configuration error may occur, and in addition, no cost migration is possible, because the dialogue flow configuration and the information collection requirement of each scene are different.

In the method, an additional complex configuration flow is not required to be introduced, the use threshold is reduced, and the potential uncontrollable risk existing in the traditional configuration mode is effectively and automatically avoided. In addition, the dialogue processing method is not a scheme customized and developed based on a certain service scene, is an abstract unified solution for the type of requirements in the dialogue system, can migrate to other scenes at zero cost, and reduces the construction cost of the multi-round dialogue system for meeting the type of requirements.

Corresponding to the session identification method provided by the embodiments of fig. 1 to 4, the present application further provides a session identification device, and since the session identification device provided by the embodiments of the present application corresponds to the session identification method provided by the embodiments of fig. 1 to 4, the implementation of the session identification method is also applicable to the session identification device provided by the embodiments of the present application, which will not be described in detail in the embodiments of the present application.

Fig. 6 is a schematic structural diagram of a dialogue identifying device provided in a fifth embodiment of the present application.

As shown in fig. 6, the dialogue recognizing device 600 includes: a dialog acquisition module 610, a set determination module 620, an identification module 630, a matching module 640, and an answer determination module 650.

The session obtaining module 610 is configured to obtain a target session in at least one round of session.

The set determination module 620 is configured to determine a set of target questions to be identified from at least one dialog.

The recognition module 630 is configured to perform entity word recognition on the target dialogue, obtain a plurality of entity words in the target dialogue, and store the plurality of entity words.

And the matching module 640 is configured to match each question in the target question set with a plurality of stored entity words, so as to determine an entity word corresponding to each question.

The answer determining module 650 is configured to determine an answer to each question according to the entity word corresponding to each question.

In one possible implementation manner of the embodiment of the present application, the set determining module 620 is specifically configured to: determining a target entity collection unit set from a plurality of entity collection unit sets corresponding to at least one round of dialogue according to a set rule; the target entity collection unit set comprises one or more entity collection units corresponding to the questions in the target question set, and the entity collection units are used for identifying entity words matched with the corresponding questions.

In one possible implementation manner of the embodiment of the present application, the target entity collection unit set includes a plurality of entity collection units, and the matching module 640 is specifically configured to: and sequentially identifying a plurality of entity words in the target dialogue by adopting each entity collecting unit in the target entity collecting unit set so as to obtain entity words matched with the problems corresponding to each entity collecting unit.

In one possible implementation manner of the embodiment of the present application, the session identifying apparatus 600 may further include:

and the stopping module is used for stopping the recognition by adopting the target entity collection unit set if all the entity collection units in the target entity collection unit set are recognized or if the entity collection units which do not recognize the matched entity words exist.

and the lower-text acquisition module is used for acquiring the lower text of the target dialogue from at least one round of dialogue.

And the storage module is used for storing the entity words below.

The recognition module 630 is further configured to use the target entity collection unit set to continuously recognize each stored entity word.

and the deleting module is used for deleting the stored plurality of entity words under the condition that each entity collecting unit in the target entity collecting unit set identifies the matched entity word.

According to the dialogue identification device, a target dialogue in at least one round of dialogue is obtained, and a target problem set to be identified in at least one round of dialogue is determined; performing entity word recognition on the target dialogue to obtain and store a plurality of entity words in the target dialogue; matching each question in the target question set with a plurality of stored entity words respectively to determine the entity word corresponding to each question; and determining answers of the questions according to the entity words corresponding to the questions. Therefore, when the user answers a plurality of questions in one round of dialogue, the system can automatically recognize the answers of the questions, unnecessary interaction times can be reduced, and the use experience of the user is improved.

According to embodiments of the present application, there is also provided an electronic device, a readable storage medium and a computer program product.

Fig. 7 shows a schematic block diagram of an example electronic device 700 that may be used to implement embodiments of the present application. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the disclosure described and/or claimed herein.

As shown in fig. 7, the apparatus 700 includes a computing unit 701 that can perform various appropriate actions and processes according to a computer program stored in a ROM (Read-Only Memory) 702 or a computer program loaded from a storage unit 708 into a RAM (Random Access Memory ) 703. In the RAM 703, various programs and data required for the operation of the device 700 may also be stored. The computing unit 701, the ROM 702, and the RAM 703 are connected to each other through a bus 704. An I/O (Input/Output) interface 705 is also connected to bus 704.

Various components in device 700 are connected to I/O interface 705, including: an input unit 706 such as a keyboard, a mouse, etc.; an output unit 707 such as various types of displays, speakers, and the like; a storage unit 708 such as a magnetic disk, an optical disk, or the like; and a communication unit 709 such as a network card, modem, wireless communication transceiver, etc. The communication unit 709 allows the device 700 to exchange information/data with other devices via a computer network, such as the internet, and/or various telecommunication networks.

The computing unit 701 may be a variety of general and/or special purpose processing components having processing and computing capabilities. Some examples of computing unit 701 include, but are not limited to, a CPU (Central Processing Unit ), a GPU (Graphic Processing Units, graphics processing unit), various dedicated AI (Artificial Intelligence ) computing chips, various computing units running machine learning model algorithms, a DSP (Digital Signal Processor ), and any suitable processor, controller, microcontroller, etc. The computing unit 701 performs the respective methods and processes described above, such as a dialogue recognition method. For example, in some embodiments, the dialog recognition method may be implemented as a computer software program tangibly embodied on a machine-readable medium, such as the storage unit 708. In some embodiments, part or all of the computer program may be loaded and/or installed onto device 700 via ROM 702 and/or communication unit 709. When a computer program is loaded into RAM 703 and executed by computing unit 701, one or more steps of the dialog recognition method described above may be performed. Alternatively, in other embodiments, the computing unit 701 may be configured to perform the dialog recognition method by any other suitable means (e.g., by means of firmware).

Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuit System, FPGA (Field Programmable Gate Array ), ASIC (Application-Specific Integrated Circuit, application-specific integrated circuit), ASSP (Application Specific Standard Product, special-purpose standard product), SOC (System On Chip ), CPLD (Complex Programmable Logic Device, complex programmable logic device), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.

Program code for carrying out methods of the present disclosure may be written in any combination of one or more programming languages. These program code may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus such that the program code, when executed by the processor or controller, causes the functions/operations specified in the flowchart and/or block diagram to be implemented. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.

In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. The machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, RAM, ROM, EPROM (Electrically Programmable Read-Only-Memory, erasable programmable read-Only Memory) or flash Memory, an optical fiber, a CD-ROM (Compact Disc Read-Only Memory), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: a display device (e.g., CRT (Cathode-Ray Tube) or LCD (Liquid Crystal Display ) monitor) for displaying information to a user; and a keyboard and pointing device (e.g., a mouse or trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic input, speech input, or tactile input.

The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: LAN (Local Area Network ), WAN (Wide Area Network, wide area network) and the internet.

The computer system may include a client and a server. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also called a cloud computing server or a cloud host, and is a host product in a cloud computing service system, so that the defects of high management difficulty and weak service expansibility in the traditional physical hosts and VPS service (Virtual Private Server, virtual special servers) are overcome. The server may also be a server of a distributed system or a server that incorporates a blockchain.

According to an embodiment of the present application, there is also provided a computer program product, which when executed by a processor, performs the dialog recognition method set forth in the above embodiment of the present application.

According to the technical scheme of the embodiment of the application, the target dialogue in at least one round of dialogue is obtained, and the target problem set to be identified in at least one round of dialogue is determined; performing entity word recognition on the target dialogue to obtain and store a plurality of entity words in the target dialogue; matching each question in the target question set with a plurality of stored entity words respectively to determine the entity word corresponding to each question; and determining answers of the questions according to the entity words corresponding to the questions. Therefore, when the user answers a plurality of questions in one round of dialogue, the system can automatically recognize the answers of the questions, unnecessary interaction times can be reduced, and the use experience of the user is improved.

It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps described in the present application may be performed in parallel, sequentially, or in a different order, provided that the desired results of the technical solutions disclosed in the present application can be achieved, and are not limited herein.

The above embodiments do not limit the scope of the application. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application are intended to be included within the scope of the present application.

Claims

1. A dialog recognition method, the method comprising:

acquiring a target dialogue in at least one round of dialogue, and determining a target entity collection unit set from a plurality of entity collection unit sets corresponding to the at least one round of dialogue according to a set rule; the target entity collection unit set comprises a plurality of entity collection units corresponding to the problems in the target problem set, and the entity collection units are used for identifying entity words matched with the corresponding problems; the entity collecting unit consists of a father node and two child nodes, wherein the father node and the child nodes are nodes in a dialogue tree, the father node is used for judging whether entity words to be collected are collected by dialogue context as node selection entering conditions, if the entity words are not collected by dialogue context, the first child node judges whether contents expressed by users contain entity words, if the entity words are contained, the collection of assignment of the variables in the context is completed, and if the entity words are not contained, the second child node is executed and jumps to the father node again to initiate entity word collection;

2. The dialog recognition method of claim 1, wherein the matching each question in the set of target questions with a plurality of stored entity words to determine an entity word corresponding to each question includes:

and sequentially identifying a plurality of entity words in the target dialogue by adopting each entity collecting unit in the target entity collecting unit set so as to obtain entity words matched with the problems corresponding to each entity collecting unit.

3. The dialog recognition method of claim 2, wherein the employing each entity-collection unit to sequentially recognize the plurality of entity words in the target dialog further comprises:

and stopping identifying by using the target entity collection unit set if all the entity collection units in the target entity collection unit set are identified, or if the entity collection units which do not identify the matched entity words exist.

4. The dialog recognition method of claim 3, wherein the entity-collection unit for which no matching entity word is recognized exists, and further comprising, after stopping recognition with the entity-collection unit set:

obtaining the following of the target dialogue from the at least one round of dialogue;

storing the entity words in the text;

and adopting the target entity collection unit set to continuously identify each stored entity word.

5. A dialog recognition method as claimed in claim 3, wherein the method further comprises:

and deleting the stored plurality of entity words under the condition that each entity collecting unit in the target entity collecting unit set identifies the matched entity word.

6. A dialog recognition device, comprising:

the set determining module is used for determining a target entity collecting unit set from a plurality of entity collecting unit sets corresponding to the at least one dialogue according to a set rule; the target entity collection unit set comprises a plurality of entity collection units corresponding to the problems in the target problem set, and the entity collection units are used for identifying entity words matched with the corresponding problems; the entity collecting unit consists of a father node and two child nodes, wherein the father node and the child nodes are nodes in a dialogue tree, the father node is used for judging whether entity words to be collected are collected by dialogue context as node selection entering conditions, if the entity words are not collected by dialogue context, the first child node judges whether contents expressed by users contain entity words, if the entity words are contained, the collection of assignment of the variables in the context is completed, and if the entity words are not contained, the second child node is executed and jumps to the father node again to initiate entity word collection;

7. The dialog recognition device of claim 6, wherein the set of target entity collection units includes a plurality of entity collection units, and the matching module is specifically configured to:

8. The dialog identification device of claim 7, wherein the device further comprises:

and the stopping module is used for stopping the recognition by the target entity collection unit set if all the entity collection units in the target entity collection unit set are recognized or the entity collection units which are not recognized to match entity words exist.

9. The dialog identification device of claim 8, wherein the device further comprises:

a context acquisition module, configured to acquire a context of the target session from the at least one round of session;

the storage module is used for storing the entity words in the text;

the identification module is further configured to continuously identify each stored entity word by using the target entity collection unit set.

10. The dialog identification device of claim 8, wherein the device further comprises:

11. An electronic device, comprising:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein,,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the dialog recognition method of any of claims 1-5.

12. A non-transitory computer readable storage medium storing computer instructions for causing the computer to perform the dialog recognition method of any of claims 1-5.