CN113282708B - Method and device for replying to robot dialog, computer equipment and storage medium - Google Patents
Method and device for replying to robot dialog, computer equipment and storage medium Download PDFInfo
- Publication number
- CN113282708B CN113282708B CN202110605316.3A CN202110605316A CN113282708B CN 113282708 B CN113282708 B CN 113282708B CN 202110605316 A CN202110605316 A CN 202110605316A CN 113282708 B CN113282708 B CN 113282708B
- Authority
- CN
- China
- Prior art keywords
- dialogue
- data
- skill
- conversation
- dialog
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3343—Query execution using phonetics
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- Human Computer Interaction (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Acoustics & Sound (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Machine Translation (AREA)
Abstract
The application relates to the field of artificial intelligence, and discloses a method, a device, computer equipment and a storage medium for robot dialogue reply, wherein the method comprises the following steps: analyzing the first dialogue data to obtain a first dialogue intention and a first analysis parameter; determining a first dialogue skill according to the first analysis parameter, controlling the robot to enter a dialogue management process of the first dialogue skill, and generating first reply data; acquiring and analyzing second dialogue data in the dialogue management process of the first dialogue skill to obtain a second dialogue intention and a second analysis parameter of the second dialogue data, and judging whether the second dialogue data matches the second dialogue skill according to the second analysis parameter; if so, controlling the robot to enter a combined conversation management process and outputting second reply data; and if not, combining the first dialogue intention and the second dialogue intention to generate second reply data of the second dialogue data. The method and the device can improve the efficiency of the robot in replying the conversations in different fields.
Description
Technical Field
The present application relates to the field of artificial intelligence, and in particular, to a method and an apparatus for a robot dialog reply, a computer device, and a storage medium.
Background
At present, the questions and answers of a plurality of robots are all table questions and answers which are used as a method for intelligently asking and answering according to structured data and are widely applied to the fields of e-commerce, insurance and the like. In a common table question-answering method, a structured data file uploaded by a user is used as a knowledge base, the intention and key information of the user are analyzed in a mode of converting natural language into database statements (NL 2 SQL), SQL query statements are generated, and corresponding answers are returned to the knowledge base. The realization method has a single dialogue form, the intention analysis is required to be carried out again in each dialogue, the jumping and the connection between different scenes cannot be carried out, and the question-answering efficiency is low.
Disclosure of Invention
The application mainly aims to provide a method, a device, computer equipment and a storage medium for replying a robot dialogue, and aims to solve the problem that the current robot dialogue cannot jump to a problem scene quickly.
In order to achieve the above object, the present application provides a method for replying to a robot dialog, comprising:
acquiring first dialogue data, and analyzing the first dialogue data to obtain a first dialogue intention and a first analysis parameter of the first dialogue data;
determining a first dialogue skill matched with the first dialogue data according to the first analysis parameter, controlling the robot to enter a dialogue management process of the first dialogue skill, generating first reply data of the first dialogue data according to the first dialogue intention, and outputting the first reply data;
acquiring second dialogue data in the dialogue management process of the first dialogue skill, analyzing the second dialogue data to obtain a second dialogue intention and a second analysis parameter of the second dialogue data, and judging whether the second dialogue data matches the second dialogue skill according to the second analysis parameter; wherein the second dialogue data is dialogue data acquired after the first dialogue data, and the second dialogue skill is a different dialogue skill from the first dialogue skill;
if yes, entering a combined conversation management process of the first conversation skill and the second conversation skill, generating second reply data of the second conversation data according to the second conversation intention, and outputting the second reply data;
if not, in the dialogue management process of the first dialogue skill, combining the first dialogue intention and the second dialogue intention to generate second reply data of the second dialogue data, and then outputting the second reply data.
Further, the first reply data and/or the second reply data comprise call service type data; the outputting the first reply data comprises:
calling an application service corresponding to the first recovery data, and jumping to a function page of the application service;
the outputting the second reply data includes:
and calling the application service corresponding to the second reply data, and jumping to a function page of the application service.
Further, after determining the first dialog skill matched with the first dialog data according to the first parsing parameter and controlling the robot to enter a dialog management process of the first dialog skill, the method further includes:
acquiring associated conversation data of the first conversation data based on the acquired user conversation associated big data;
determining associated dialog skills from the associated dialog data;
acquiring configuration parameters of the associated conversation skills;
and storing the configuration parameters of the related conversation skills into a conversation management flow of the first conversation skill.
Further, the determining associated dialog skills from the associated dialog data includes:
acquiring keywords in the associated dialogue data;
and matching in a database according to the keywords, and determining the associated conversation skills matched with the keywords.
Further, before the obtaining the second session data in the session management process of the first session technology, the method further includes:
judging whether second dialogue data are received within preset time of entering a dialogue management process of the first dialogue technology;
if not, taking the received conversation data after the preset time is exceeded as new first conversation data;
and if so, acquiring second dialogue data in the dialogue management process of the first dialogue technology.
Further, after entering the combined dialogue management process of the first dialogue skill and the second dialogue skill, generating second reply data of the second dialogue data according to the second dialogue intention, and outputting the second reply data, the method further includes:
outputting guide information of the first dialogue skill; the guidance information includes questions associated with the conversation skills;
acquiring third dialogue data aiming at the guide information, and judging whether the third dialogue data is matched with second dialogue skills;
and if not, controlling the robot to return to the dialogue management process of the first dialogue technique, and storing the parameters of the second dialogue technique.
Further, the generating first reply data for the first dialog data according to the first dialog intention includes:
acquiring a reply database corresponding to the first dialogue skill;
generating, in a reply database of the first dialog skill, first reply data of the first dialog data according to the first dialog intention.
The application also provides a device for robot dialogue reply, which comprises:
the first obtaining module is used for obtaining first dialogue data and analyzing the first dialogue data to obtain a first dialogue intention and a first analysis parameter of the first dialogue data;
the skill identification module is used for determining a first dialogue skill matched with the first dialogue data according to the first analysis parameter, controlling the robot to enter a dialogue management process of the first dialogue skill, generating first answer data of the first dialogue data according to the first dialogue intention, and outputting the first answer data;
the intention judging module is used for acquiring second dialogue data in the dialogue management process of the first dialogue skill, analyzing the second dialogue data to obtain a second dialogue intention and a second analysis parameter of the second dialogue data, and judging whether the second dialogue data is matched with a second dialogue skill according to the second analysis parameter, wherein the second dialogue skill is a dialogue skill different from the first dialogue skill;
the skill switching module is configured to enter a combined conversation management process of the first conversation skill and the second conversation skill if it is determined that the second conversation data matches the second conversation skill, generate second reply data of the second conversation data according to the second conversation intention, and output the second reply data;
and if the second dialogue data is judged not to match the second dialogue skill, combining the first dialogue intention and the second dialogue intention to generate second reply data of the second dialogue data in the dialogue management process of the first dialogue skill, and then outputting the second reply data.
The present application further provides a computer readable storage medium having stored thereon a computer program which, when being executed by a processor, carries out the steps of the method of robotic dialog reply according to any of the above.
The application provides a conversation reply method for flexibly switching conversation skills of a robot, wherein in a scene of a question-answer sentence, the robot acquires first conversation data, analyzes the first conversation data to obtain a first conversation intention and a first analysis parameter of the first conversation data, the intention represents a theme and a main body of the conversation data, the analysis parameter represents a type range of the conversation data, a first conversation skill matched with the first conversation data is determined according to the first analysis parameter, namely, which conversation scene the first conversation data matches is determined, then the robot is controlled to enter a conversation management process of the first conversation skill, namely, the robot is controlled to enter the conversation scene and then converse with a user, under a conversation management process of the first conversation skill, a first reply data of the first conversation data is generated and then the first reply data is output, a second conversation data is acquired, and if the second conversation data matches the second conversation skill, the robot is controlled to enter a combination management process of the first conversation and the second conversation skill, and then the second conversation data is output according to the second conversation management process of the first conversation skill; and if the second dialogue skill is not matched, under the scene of the first dialogue skill, combining the first dialogue intention and the second dialogue intention to generate second reply data of the second dialogue data and then outputting the second reply data, wherein the context information of the whole dialogue is considered, the context connection of dialogue reply is improved, and the accuracy of the dialogue reply is improved.
Drawings
FIG. 1 is a flowchart illustrating an embodiment of a method for a robot dialog reply according to the present application;
FIG. 2 is a flowchart illustrating a method for a robotic dialog reply according to an embodiment of the present invention;
FIG. 3 is a schematic structural diagram of an embodiment of a device for robot dialogue reply according to the present application;
FIG. 4 is a block diagram illustrating a computer device according to an embodiment of the present invention.
The implementation, functional features and advantages of the objectives of the present application will be further explained with reference to the accompanying drawings.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of and not restrictive on the broad application.
Referring to fig. 1, an embodiment of the present application provides a method for replying to a robot dialog, which includes steps S10 to S50, and the following details are provided for the steps of the method for replying to a robot dialog, which may be implemented by an application program with a robot built in, where the robot is not specifically a physical machine device, and may refer to a "robot" function built in the application program, and the "robot" function is used to implement a dialog question and answer.
S10, first dialogue data are obtained and analyzed, and a first dialogue intention and a first analysis parameter of the first dialogue data are obtained.
In the dialog recognition scenario between the robot and the user, after the user inputs dialog data, defining currently received data as first dialog data, and then parsing the first dialog data, where the dialog skills of the robot include question-answer skills, chat skills, and task skills, the parsing includes but is not limited to parsing with NLU (natural language understanding) or parsing with NL2SQL (natural language to SQL statement), parsing the first dialog data to obtain intent of the first dialog data and parsing parameters, the intent represents a topic of the dialog data, i.e., semantic meaning expressed by the dialog data, and content of consultation with the dialog data can be determined according to intent of the dialog data, the parsing parameters represent a type range of the dialog data, i.e., a scene range to which the dialog data applies, dialog manners used by dialog data in different types of ranges are different, including consultation types, task types, chat types, and the like, and the embodiment defines the first dialog data as the first dialog data, and defines the intent of the first dialog data as the first dialog data parsing parameters. In one embodiment, a user converses with the robot through a dialog box, and at the moment, the information input by the user in the dialog box is acquired, so that the conversation data is acquired; in another embodiment, the user has a conversation with the robot by voice, and the voice data of the user is picked up by a microphone at this time, thereby acquiring the conversation data. The manner of acquiring the session data may also include other manners, and this embodiment is not limited in particular.
S20, determining a first dialogue skill matched with the first dialogue data according to the first analysis parameter, controlling the robot to enter a dialogue management process of the first dialogue skill, generating first reply data of the first dialogue data according to the first dialogue intention, and outputting the first reply data;
in this embodiment, after obtaining a first dialogue intention and a first analysis parameter of the first dialogue data, determining a first dialogue skill matched with the first dialogue data according to the first analysis parameter, that is, determining which dialogue scene type of the robot the first dialogue data is matched with; and then controlling the robot to enter a dialogue management process of the first dialogue skill, namely controlling the robot to enter a corresponding dialogue scene and then have a dialogue with the user, completing collection of key information of the first dialogue data according to the first dialogue intention under the dialogue management process of the first dialogue skill, generating first reply data of the first dialogue data according to the key information, and then outputting the first reply data to reply the first dialogue data.
And S30, acquiring second dialogue data in the dialogue management process of the first dialogue skill, analyzing the second dialogue data to obtain a second dialogue intention and a second analysis parameter of the second dialogue data, and judging whether the second dialogue data is matched with a second dialogue skill according to the second analysis parameter, wherein the second dialogue skill is a dialogue skill different from the first dialogue skill.
In this embodiment, after the robot outputs the first answer data, the user may check the first answer data, and after the robot outputs the first answer data, the user has performed a dialog with the robot again, at this time, the dialog data sent again by the user is obtained in the dialog management process of the first dialog technology, and is defined as second dialog data, so as to obtain second dialog data, and then the second dialog data is analyzed, where the analysis includes but is not limited to analysis by using NLU (natural language understanding) or analysis by using NL2SQL (natural language to SQL statement), a second dialog intention and a second analysis parameter of the second dialog data are obtained after the second dialog data is analyzed, and then it is determined whether the second dialog data matches a second dialog skill, that is, the dialog skill of the second dialog data is determined according to the second analysis parameter, and if the dialog skill of the second dialog data is the same as the dialog skill of the first dialog data, the second dialog data is a new skill, and the matched dialog data is still the first dialog skill; and if the conversation skill of the second conversation data is not the same as that of the first conversation data, matching a new conversation skill with the second conversation data, and defining the second conversation skill as a second conversation skill which is different from the first conversation skill.
S40, if yes, controlling the robot to enter a combined conversation management process of the first conversation skill and the second conversation skill, generating second reply data of the second conversation data according to the second conversation intention, and outputting the second reply data;
in this embodiment, after the second dialog data is judged to match the second dialog skill, the robot is controlled to enter a combined dialog management process of the first dialog skill and the second dialog skill, the combined dialog management process is realized by merging the two different dialog skills, and a question and a response of the two different dialog data can be quickly completed in the combined dialog management process, that is, a first analysis parameter and a second analysis parameter are reserved in the combined dialog management process, then second reply data of the second dialog data is generated according to the second dialog intention, and the second reply data is output, so that a dialog reply is in a new dialog management process, and the two different analysis parameters are reserved as contextual connections of a dialog, and the accuracy of the dialog reply can be improved through the contextual connections, and after the second reply data is output, third dialog data is subsequently received, the first dialog skill or the second dialog skill can be quickly linked, so that the contextual connections of the dialog reply are improved.
And S50, if not, in the dialogue management process of the first dialogue skill, combining the first dialogue intention and the second dialogue intention to generate second reply data of the second dialogue data, and then outputting the second reply data.
In this embodiment, after it is determined that the second dialog data does not match the second dialog skill, that is, the second dialog data and the first dialog data are both dialog data under the first dialog skill, at this time, the robot is still in the dialog management process of the first dialog skill, and then generates second reply data of the second dialog data by combining the first dialog intention and the second dialog intention and outputs the second reply data, that is, generates the second reply data of the second dialog data by combining the first dialog data and the second dialog data of the context, thereby improving contextual connection of dialog reply and improving accuracy of dialog reply.
The embodiment provides a dialog reply method for flexibly switching dialog skills of a robot, in a scene of a question-answer sentence, the robot acquires first dialog data, analyzes the first dialog data to obtain a first dialog intention and a first analysis parameter of the first dialog data, the intention represents a theme and a main body of the dialog data, the analysis parameter represents a type range of the dialog data, a first dialog skill matched with the first dialog data is determined according to the first analysis parameter, namely, which dialog scene the first dialog data matches is determined, then the robot is controlled to enter a dialog management process of the first dialog skill, namely, the robot is controlled to enter the dialog scene and then has a dialog with a user, in the dialog management process of the first dialog skill, after generating first reply data of the first dialog data, the first reply data is output, a second dialog data is acquired, and if the second dialog data matches with the second dialog skill, the robot is controlled to enter a combination management of the first dialog technology and the second dialog skill management process, and then the second dialog data is output according to the second dialog intention data; and if the second dialogue skill is not matched, under the scene of the first dialogue skill, combining the first dialogue intention and the second dialogue intention to generate second reply data of the second dialogue data and then outputting the second reply data, wherein the context information of the whole dialogue is considered, the context connection of dialogue reply is improved, and the accuracy of the dialogue reply is improved.
In one embodiment, the first reply data and/or the second reply data comprise call service type data; the outputting the first reply data comprises:
calling an application service corresponding to the first recovery data, and jumping to a function page of the application service;
the outputting the second reply data includes:
and calling the application service corresponding to the second reply data, and jumping to a function page of the application service.
In this embodiment, the first reply data and/or the second reply data include call service type data, that is, the reply data may not only be text data, but also be an application service, when the first reply data is output, the application service corresponding to the first reply data is called, and a functional interface of the application service is output, for example, a robot is applied to a consultation of an application, and when the application service corresponding to the first reply data is called, the application service is output, that is, a function page of the application service is skipped; similarly, the second reply data may also be call service type data, and when the second reply data is output, the application service corresponding to the second reply data is called, and a function interface of the application service is output. By the calling type service data, the application service matched with the dialogue data of the user can be quickly called, and the dialogue question-answer solving efficiency is improved.
In an embodiment, as shown in fig. 2, after the determining, according to the first parsing parameter, a first session skill matched with the first session data and controlling the robot to enter a session management process of the first session skill, the method further includes:
s21, acquiring associated conversation data of the first conversation data based on the acquired user conversation associated big data;
s22, determining the associated conversation skill according to the associated conversation data;
s23, acquiring configuration parameters of the associated dialogue skills;
and S24, storing the configuration parameters of the associated dialogue skills into the dialogue management flow of the first dialogue skill.
In this embodiment, after determining a first session skill matched with the first session data according to the first parsing parameter, and controlling the robot to enter a session management process of the first session skill, a current session skill scene is determined, meanwhile, multiple times of session data of multiple users are collected through big data, the big data establishes any two pieces of session data which are associated with each other, for example, from the collected big data of the session association of the users, it is found that there is a probability X1 that there is a consultation session data 2 after consulting the session data 1, and there is a probability X2 that there is a consultation session data 4 after consulting the conversation data 3, and when the probability X1 or X2 exceeds a certain preset value, it is determined that the two pieces of session data are associated with each other. In this embodiment, after a session management process of a first session skill is determined, a correlation intention of first session data is obtained based on collected user session correlation big data, then a correlation session skill is determined according to the correlation session data, configuration parameters of the correlation session skill are obtained, and then the configuration parameters of the correlation session skill are stored in the session management process of the first session skill, when the correlation session data is subsequently consulted, reply data matching of the correlation session data can be quickly performed, and efficiency of session reply is improved; or, the associated intentions correspond to different conversation skills, and when the associated intentions are consulted subsequently, the conversation skill can be switched to another conversation skill quickly, so that the conversation reply efficiency is improved.
In one embodiment, the determining associated dialog skills from the associated dialog data comprises:
acquiring keywords in the associated dialogue data;
and matching in a database according to the keywords, and determining the associated conversation skills matched with the keywords.
In this embodiment, in an implementation manner of determining associated conversation skills according to the associated conversation data, keywords in the associated conversation data are obtained, and then matching is performed in a database according to the keywords, where a large number of keywords included in the conversation skills are stored in the database, and matching is performed in the database according to the keywords, so as to determine the associated conversation skills matched with the keywords. And if the keywords belong to the keywords contained in the first dialogue skill, determining that the associated dialogue data matches the first dialogue skill, so that the robot can be rapidly controlled to switch under different dialogue skills, the dialogue question and answer under the dialogue skill scene can be completed, and the accuracy of the dialogue question and answer can be improved.
In one embodiment, after entering the combined session management procedure of the first session skill and the second session skill, generating second reply data of the second session data according to the second session intention, and outputting the second reply data, the method further includes:
outputting guide information of the first dialogue skill; the guidance information includes questions associated with the conversation skills;
acquiring third dialogue data aiming at the guide information, and judging whether the third dialogue data is matched with second dialogue skills;
and if not, controlling the robot to return to the dialogue management process of the first dialogue technique, and storing the parameters of the second dialogue technique.
In this embodiment, after the robot is controlled to enter the combined session management flow of the first session skill and the second session skill, the robot outputs the second reply data, and then outputs the guidance information of the first session skill, where the guidance information includes a problem associated with the session skill, and guides the user to jump back to the first session skill, and after the guidance information is output, obtains third session data, and then determines whether the third session data matches the second session skill, if not, it indicates that the second session skill may be caused by a jump of user thinking, and at this time, the robot is controlled to return to the session management flow of the first session skill, and a preset duration of a parameter of the second session skill is saved, so that the robot is conveniently controlled to flexibly switch to the second session skill, and further, after the preset duration of the parameter of the second session skill is saved, if no received session data matches the second session skill, the parameter of the second session skill is deleted, so as to narrow a data range of session reply, and improve efficiency of session reply.
In an embodiment, before acquiring the second session data in the session management process of the first session technology, the method further includes:
judging whether second dialogue data are received within the preset time of entering the dialogue management process of the first dialogue technology;
if not, taking the received conversation data after the preset time is exceeded as new first conversation data;
and if so, acquiring second dialogue data in the dialogue management process of the first dialogue technology.
In this embodiment, before the second session data is obtained in the session management flow of the first session technology, because the second session data may not be received for a long time after the first session data is received, it is determined whether the second session data is received within a preset time after the session management flow of the first session technology is entered, if not, the session data received after exceeding the preset time is used as new first session data, and the new first session data is re-determined, and if so, the second session data is obtained in the session management flow of the first session technology, so that the old session data is prevented from being misused as data for identifying the first session technology due to long waiting time, and timeliness of a session reply is provided.
In one embodiment, the generating first reply data for the first dialog data according to the first dialog intention comprises:
acquiring a reply database corresponding to the first dialogue skill;
generating, in the reply database of the first dialogue skill, first reply data of the first dialogue data according to the first dialogue intention.
In this embodiment, when the first answer data of the first dialogue data is generated according to the first dialogue intention, the answer database corresponding to the first dialogue skill is first obtained, and then matching is performed on the answer database of the first dialogue skill according to the first dialogue intention, so as to generate the first answer data of the first dialogue data, which does not need global matching, thereby improving the retrieval efficiency of the answer data.
Referring to fig. 3, the present application further provides a device for robot dialog reply, including:
a first obtaining module 10, configured to obtain first dialogue data and analyze the first dialogue data to obtain a first dialogue intention and a first analysis parameter of the first dialogue data;
the skill identification module 20 is configured to determine a first dialogue skill matched with the first dialogue data according to the first analysis parameter, control the robot to enter a dialogue management process of the first dialogue skill, generate first answer data of the first dialogue data according to the first dialogue intention, and output the first answer data;
the intention judging module 30 is configured to obtain second dialogue data in the dialogue management process of the first dialogue skill, analyze the second dialogue data to obtain a second dialogue intention and a second analysis parameter of the second dialogue data, and judge whether the second dialogue data matches the second dialogue skill according to the second analysis parameter; wherein the second dialogue data is dialogue data acquired after the first dialogue data, and the second dialogue skill is a different dialogue skill from the first dialogue skill;
a skill switching module 40, configured to enter a combined conversation management process of the first conversation skill and the second conversation skill if it is determined that the second conversation data matches the second conversation skill, generate second reply data of the second conversation data according to the second conversation intention, and output the second reply data;
and if it is determined that the second dialogue data does not match the second dialogue skill, in the dialogue management process of the first dialogue skill, combining the first dialogue intention and the second dialogue intention to generate second reply data of the second dialogue data, and then outputting the second reply data.
As described above, it is understood that the components of the device for robotic dialog reply presented in the present application may implement the functions of any of the methods for robotic dialog reply described above.
In one embodiment, the first reply data and/or the second reply data comprise call service type data; the skill identification module 20 further performs:
calling an application service corresponding to the first recovery data, and jumping to a function page of the application service;
the skill switching module 40 or the skill preservation module 50 further performs:
and calling the application service corresponding to the second reply data, and jumping to a function page of the application service.
In one embodiment, the skill identification module 20 further performs:
acquiring associated conversation data of the first conversation data based on the acquired user conversation associated big data;
determining associated dialog skills from the associated dialog data;
acquiring configuration parameters of the associated conversation skills;
and storing the configuration parameters of the associated conversation skills into a conversation management flow of the first conversation skill.
In one embodiment, the skill identification module 20 further performs:
acquiring keywords in the associated dialogue data;
and matching in a database according to the keywords, and determining the associated conversation skills matched with the keywords.
In one embodiment, the intent determination module 30 further performs:
judging whether second dialogue data are received within the preset time of entering the dialogue management process of the first dialogue technology;
if not, taking the conversation data received after the preset time is exceeded as new first conversation data;
and if so, acquiring second dialogue data in the dialogue management process of the first dialogue skill.
In one embodiment, the skill switching module 50 further comprises performing:
outputting guide information of the first dialogue skill; the guidance information includes questions associated with the conversation skills;
acquiring third dialogue data aiming at the guide information, and judging whether the third dialogue data is matched with a second dialogue skill or not;
and if not, controlling the robot to return to the dialogue management process of the first dialogue technique, and storing the parameters of the second dialogue technique.
In one embodiment, the skill identification module 20 further performs:
acquiring a reply database corresponding to the first dialogue skill;
generating, in the reply database of the first dialogue skill, first reply data of the first dialogue data according to the first dialogue intention.
Referring to fig. 4, an embodiment of the present application further provides a computer device, where the computer device may be a mobile terminal, and an internal structure of the computer device may be as shown in fig. 4. The computer equipment comprises a processor, a memory, a network interface, a display device and an input device which are connected through a system bus. Wherein, the network interface of the computer equipment is used for connecting and communicating with an external terminal through a network. The input means of the computer device is for receiving input from a user. The computer designed processor is used to provide computational and control capabilities. The memory of the computer device includes a storage medium. The storage medium stores an operating system, a computer program, and a database. The database of the computer device is used for storing data. The computer program is executed by a processor to implement a method for robotic dialog reply.
The method for the processor to execute the robot dialogue reply comprises the following steps: acquiring first dialogue data, and analyzing the first dialogue data to obtain a first dialogue intention and a first analysis parameter of the first dialogue data; determining a first dialogue skill matched with the first dialogue data according to the first analysis parameter, controlling a robot to enter a dialogue management process of the first dialogue skill, generating first reply data of the first dialogue data according to the first dialogue intention, and outputting the first reply data; acquiring second dialogue data in the dialogue management process of the first dialogue skill, analyzing the second dialogue data to obtain a second dialogue intention and a second analysis parameter of the second dialogue data, and judging whether the second dialogue data matches the second dialogue skill according to the second analysis parameter; wherein the second dialogue data is dialogue data acquired after the first dialogue data, and the second dialogue skill is a different dialogue skill from the first dialogue skill; if yes, entering a combined conversation management process of the first conversation skill and the second conversation skill, generating second reply data of the second conversation data according to the second conversation intention, and outputting the second reply data; if not, in the dialogue management process of the first dialogue skill, combining the first dialogue intention and the second dialogue intention to generate second reply data of the second dialogue data, and then outputting the second reply data.
The computer equipment provides a conversation reply method for flexibly switching conversation skills of a robot, in a scene of a question-answer sentence, the robot acquires first conversation data, analyzes the first conversation data to obtain a first conversation intention and a first analysis parameter of the first conversation data, the intention represents a theme and a main body of the conversation data, the analysis parameter represents a type range of the conversation data, a first conversation skill matched with the first conversation data is determined according to the first analysis parameter, namely, which conversation scene of the robot is matched with the first conversation data is determined, then the robot is controlled to enter a conversation management process of the first conversation skill, namely, the robot is controlled to enter the conversation scene and then converse with a user, in the conversation management process of the first conversation skill, the first conversation data is output after the first conversation data is generated, a second conversation data is acquired, and if the second conversation data is matched with the second conversation skill, the robot is controlled to enter a combination management process of the first conversation and the second conversation skill, and then the second conversation data is output after the second conversation management process is switched, and the second conversation data is output according to the second conversation scene; and if the second dialogue skill is not matched, under the scene of the first dialogue skill, combining the first dialogue intention and the second dialogue intention to generate second reply data of the second dialogue data and then outputting the second reply data, wherein the context information of the whole dialogue is considered, the context connection of dialogue reply is improved, and the accuracy of the dialogue reply is improved.
An embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored, and the computer program, when executed by the processor, implements a method for robotic dialog reply, including the steps of: acquiring first dialogue data, and analyzing the first dialogue data to obtain a first dialogue intention and a first analysis parameter of the first dialogue data; determining a first dialogue skill matched with the first dialogue data according to the first analysis parameter, controlling the robot to enter a dialogue management process of the first dialogue skill, generating first reply data of the first dialogue data according to the first dialogue intention, and outputting the first reply data; acquiring second dialogue data in the dialogue management process of the first dialogue skill, analyzing the second dialogue data to obtain a second dialogue intention and a second analysis parameter of the second dialogue data, and judging whether the second dialogue data matches the second dialogue skill according to the second analysis parameter; wherein the second dialog data is dialog data acquired after the first dialog data, and the second dialog skill is a different dialog skill from the first dialog skill; if yes, entering a combined conversation management process of the first conversation skill and the second conversation skill, generating second reply data of the second conversation data according to the second conversation intention, and outputting the second reply data; if not, in the dialogue management process of the first dialogue skill, combining the first dialogue intention and the second dialogue intention to generate second reply data of the second dialogue data, and then outputting the second reply data.
The computer readable storage medium provides a conversation reply method for flexibly switching conversation skills of a robot, in a scene of a question-answer sentence, the robot acquires first conversation data, analyzes the first conversation data to obtain a first conversation intention and a first analysis parameter of the first conversation data, the intention represents a theme and a main body of the conversation data, the analysis parameter represents a type range of the conversation data, a first conversation skill matched with the first conversation data is determined according to the first analysis parameter, namely, which conversation scene of the robot is matched with the first conversation data is determined, then the robot is controlled to enter a conversation management flow of the first conversation skill, namely, the robot is controlled to enter the conversation scene and then converse with a user, in the conversation management flow of the first conversation skill, the first conversation data is output after the first conversation data is generated, a second conversation data is acquired, and if the second conversation data is matched with the second conversation skill, the robot is controlled to enter a combination management flow of the first conversation and the second conversation skill is switched to the second conversation management flow, and then the second conversation data is output according to the second conversation management flow of the first conversation skill; and if the second dialogue skill is not matched, under the scene of the first dialogue skill, combining the first dialogue intention and the second dialogue intention to generate second reply data of the second dialogue data and then outputting the second reply data, wherein the context information of the whole dialogue is considered, the context connection of dialogue reply is improved, and the accuracy of the dialogue reply is improved.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium, and can include the processes of the embodiments of the methods described above when the computer program is executed.
Any reference to memory, storage, database, or other medium provided herein and used in the embodiments may include non-volatile and/or volatile memory.
Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (SSRDRAM), enhanced SDRAM (ESDRAM), synchronous Link (SLDRAM), rambus (Rambus) direct RAM (RDRAM), direct Rambus Dynamic RAM (DRDRAM), and Rambus Dynamic RAM (RDRAM), among others.
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, apparatus, article, or method 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, apparatus, article, or method. Without further limitation, an element defined by the phrase "comprising one of 8230, and" comprising 8230does not exclude the presence of additional like elements in a process, apparatus, article, or method comprising the element.
The above description is only a preferred embodiment of the present application and is not intended to limit the scope of the present application.
All the equivalent structures or equivalent processes performed by using the contents of the specification and the drawings of the present application, or directly or indirectly applied to other related technical fields, are included in the scope of protection of the present application.
Claims (9)
1. A method of robotic dialog reply, comprising:
acquiring first dialogue data, and analyzing the first dialogue data to obtain a first dialogue intention and a first analysis parameter of the first dialogue data;
determining a first dialogue skill matched with the first dialogue data according to the first analysis parameter, controlling a robot to enter a dialogue management process of the first dialogue skill, generating first reply data of the first dialogue data according to the first dialogue intention, and outputting the first reply data;
acquiring second dialogue data in the dialogue management process of the first dialogue skill, analyzing the second dialogue data to obtain a second dialogue intention and a second analysis parameter of the second dialogue data, and judging whether the second dialogue data matches the second dialogue skill according to the second analysis parameter; wherein the second dialog data is dialog data acquired after the first dialog data, and the second dialog skill is a different dialog skill from the first dialog skill;
if yes, entering a combined conversation management process of the first conversation skill and the second conversation skill, generating second reply data of the second conversation data according to the second conversation intention, and outputting the second reply data;
if not, in the dialogue management process of the first dialogue skill, combining the first dialogue intention and the second dialogue intention to generate second reply data of the second dialogue data, and then outputting the second reply data;
determining a first dialogue skill matched with the first dialogue data according to the first analysis parameter, and controlling the robot to enter a dialogue management process of the first dialogue skill, wherein the method further comprises the following steps:
acquiring associated conversation data of the first conversation data based on the acquired user conversation associated big data;
determining associated dialog skills from the associated dialog data;
acquiring configuration parameters of the associated conversation skills;
and storing the configuration parameters of the associated conversation skills into a conversation management flow of the first conversation skill.
2. A method of robotic dialog replying as claimed in claim 1, wherein the first reply data and/or the second reply data comprises call service type data; the outputting the first recovery data comprises:
calling an application service corresponding to the first recovery data, and jumping to a function page of the application service;
the outputting the second reply data includes:
and calling the application service corresponding to the second reply data, and jumping to a function page of the application service.
3. The method of robotic dialog reply according to claim 1, wherein the determining of associated dialog skills from the associated dialog data comprises:
acquiring keywords in the associated dialogue data;
and matching in a database according to the keywords, and determining the associated conversation skills matched with the keywords.
4. The method for robotic dialog reply according to claim 1, further comprising, before the obtaining second dialog data in the dialog management procedure of the first dialog technique:
judging whether second dialogue data are received within the preset time of entering the dialogue management process of the first dialogue technology;
if not, taking the received conversation data after the preset time is exceeded as new first conversation data;
and if so, acquiring second dialogue data in the dialogue management process of the first dialogue technology.
5. The method for robotic dialog reply according to claim 1, wherein the entering of the combined dialog management procedure of the first dialog skill and the second dialog skill, the generating of the second reply data for the second dialog data according to the second dialog intention, and the outputting of the second reply data further comprises:
outputting guide information of the first dialogue skill; the guidance information includes questions associated with the conversation skills;
acquiring third dialogue data aiming at the guide information, and judging whether the third dialogue data is matched with second dialogue skills;
and if not, controlling the robot to return to the dialogue management process of the first dialogue technique, and storing the parameters of the second dialogue technique.
6. A method of robotic dialog replying as claimed in claim 1, wherein the generating of first reply data for the first dialog data in accordance with the first dialog intent comprises:
acquiring a reply database corresponding to the first dialogue skill;
generating, in the reply database of the first dialogue skill, first reply data of the first dialogue data according to the first dialogue intention.
7. An apparatus for robotic dialog reply, comprising:
the first obtaining module is used for obtaining first dialogue data and analyzing the first dialogue data to obtain a first dialogue intention and a first analysis parameter of the first dialogue data;
the skill identification module is used for determining a first dialogue skill matched with the first dialogue data according to the first analysis parameter, controlling the robot to enter a dialogue management process of the first dialogue skill, generating first answer data of the first dialogue data according to the first dialogue intention, and outputting the first answer data;
the intention judging module is used for acquiring second dialogue data in the dialogue management process of the first dialogue skill, analyzing the second dialogue data to obtain a second dialogue intention and a second analysis parameter of the second dialogue data, and judging whether the second dialogue data is matched with the second dialogue skill according to the second analysis parameter; wherein the second dialogue data is dialogue data acquired after the first dialogue data, and the second dialogue skill is a different dialogue skill from the first dialogue skill;
the skill switching module is configured to enter a combined conversation management process of the first conversation skill and the second conversation skill if it is determined that the second conversation data matches the second conversation skill, generate second reply data of the second conversation data according to the second conversation intention, and output the second reply data;
a skill saving module, configured to, if it is determined that the second session data does not match the second session skill, generate second reply data of the second session data in combination with the first session intention and the second session intention in a session management process of the first session skill, and output the second reply data;
the skill identification module further performs:
acquiring associated conversation data of the first conversation data based on the acquired user conversation associated big data;
determining associated dialog skills from the associated dialog data;
acquiring configuration parameters of the associated conversation skills;
and storing the configuration parameters of the related conversation skills into a conversation management flow of the first conversation skill.
8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, performs the steps of the method of robotic dialog reply according to any of claims 1 to 6.
9. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method for robotic dialog reply according to any of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110605316.3A CN113282708B (en) | 2021-05-31 | 2021-05-31 | Method and device for replying to robot dialog, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110605316.3A CN113282708B (en) | 2021-05-31 | 2021-05-31 | Method and device for replying to robot dialog, computer equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113282708A CN113282708A (en) | 2021-08-20 |
| CN113282708B true CN113282708B (en) | 2023-04-07 |
Family
ID=77282839
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110605316.3A Active CN113282708B (en) | 2021-05-31 | 2021-05-31 | Method and device for replying to robot dialog, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113282708B (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112800195A (en) * | 2021-01-18 | 2021-05-14 | 南京奥拓电子科技有限公司 | Configuration method and system of conversation robot |
Family Cites Families (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107369443B (en) * | 2017-06-29 | 2020-09-25 | 北京百度网讯科技有限公司 | Dialog management method and device based on artificial intelligence |
| CN108415932B (en) * | 2018-01-23 | 2023-12-22 | 思必驰科技股份有限公司 | Man-machine conversation method and electronic equipment |
| WO2019161207A1 (en) * | 2018-02-15 | 2019-08-22 | DMAI, Inc. | System and method for conversational agent via adaptive caching of dialogue tree |
| CN108874967B (en) * | 2018-06-07 | 2023-06-23 | 腾讯科技（深圳）有限公司 | Dialogue state determining method and device, dialogue system, terminal and storage medium |
| CN109033223B (en) * | 2018-06-29 | 2021-09-07 | 北京百度网讯科技有限公司 | Method, apparatus, device and computer-readable storage medium for cross-type conversation |
| US10740371B1 (en) * | 2018-12-14 | 2020-08-11 | Clinc, Inc. | Systems and methods for intelligently configuring and deploying a machine learning-based dialogue system |
| CN109977208B (en) * | 2019-03-22 | 2021-04-09 | 北京中科汇联科技股份有限公司 | Dialogue system integrating FAQ (failure-based query language) and task and active guidance |
| US11551665B2 (en) * | 2019-09-12 | 2023-01-10 | Oracle International Corporation | Dynamic contextual dialog session extension |
| CN111400467B (en) * | 2020-03-09 | 2023-05-16 | 上海国民集团健康科技有限公司 | Robot chatting method |
| CN111639168B (en) * | 2020-05-21 | 2023-06-09 | 北京百度网讯科技有限公司 | Multi-round dialogue processing method and device, electronic equipment and storage medium |
| CN112035633B (en) * | 2020-08-21 | 2023-07-25 | 腾讯科技（深圳）有限公司 | Data processing method, device, dialogue equipment and storage medium |
| CN112579757A (en) * | 2020-12-25 | 2021-03-30 | 泰康保险集团股份有限公司 | Intelligent question and answer method and device, computer readable storage medium and electronic equipment |
-
2021
- 2021-05-31 CN CN202110605316.3A patent/CN113282708B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112800195A (en) * | 2021-01-18 | 2021-05-14 | 南京奥拓电子科技有限公司 | Configuration method and system of conversation robot |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113282708A (en) | 2021-08-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112037799B (en) | Voice interrupt processing method and device, computer equipment and storage medium | |
| CN111105782B (en) | Session interaction processing method and device, computer equipment and storage medium | |
| CN112492111B (en) | Intelligent voice outbound method, device, computer equipment and storage medium | |
| CN111104495A (en) | Information interaction method, device, equipment and storage medium based on intention recognition | |
| CN105355200B (en) | System and method for directly training and modifying robot interactive content | |
| CN109614627B (en) | Text punctuation prediction method and device, computer equipment and storage medium | |
| CN111177338B (en) | Context-based multi-round dialogue method | |
| CN113626568A (en) | Man-machine conversation control method and device for robot, computer equipment and medium | |
| CN112214589B (en) | Method for multi-round session framework based on cold start | |
| CN113157371A (en) | Manual customer service switching method and device, storage medium and computer equipment | |
| CN111666388B (en) | Dialogue data processing method, device, computer equipment and storage medium | |
| CN113282708B (en) | Method and device for replying to robot dialog, computer equipment and storage medium | |
| CN110931002B (en) | Man-machine interaction method, device, computer equipment and storage medium | |
| KR20090076318A (en) | Realtime conversational service system and method thereof | |
| CN110502631B (en) | Input information response method and device, computer equipment and storage medium | |
| CN116432665A (en) | Dialogue model construction method, text generation method, device, system and equipment | |
| CN109726002B (en) | Processing flow adjusting method and device | |
| CN108933870B (en) | Dialing control method and device for sales call, computer equipment and storage medium | |
| CN113190660A (en) | Questionnaire survey method and device | |
| CN116402057A (en) | Man-machine interaction system based on natural language | |
| CN114446325A (en) | Information pushing method and device based on emotion recognition, computer equipment and medium | |
| CN115658875A (en) | Data processing method based on chat service and related product | |
| US20190179902A1 (en) | Systems and methods for task automation using natural language processing | |
| CN114639379A (en) | Interaction method and device of intelligent electric appliance, computer equipment and medium | |
| CN115442480A (en) | Outbound method, device, medium, server and intelligent robot |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |