CN111816173A - Dialogue data processing method, device, storage medium and computer equipment - Google Patents
Dialogue data processing method, device, storage medium and computer equipment Download PDFInfo
- Publication number
- CN111816173A CN111816173A CN202010486061.9A CN202010486061A CN111816173A CN 111816173 A CN111816173 A CN 111816173A CN 202010486061 A CN202010486061 A CN 202010486061A CN 111816173 A CN111816173 A CN 111816173A
- Authority
- CN
- China
- Prior art keywords
- data
- text
- user
- key information
- operation strategy
- 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.)
- Granted
Links
- 238000003672 processing method Methods 0.000 title claims abstract description 8
- 238000013486 operation strategy Methods 0.000 claims abstract description 72
- 238000000034 method Methods 0.000 claims abstract description 51
- 238000004590 computer program Methods 0.000 claims description 12
- 230000003993 interaction Effects 0.000 abstract description 21
- 230000008569 process Effects 0.000 abstract description 10
- 238000010586 diagram Methods 0.000 description 11
- 230000006870 function Effects 0.000 description 9
- 230000002452 interceptive effect Effects 0.000 description 9
- 230000004044 response Effects 0.000 description 8
- 238000013500 data storage Methods 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 238000003058 natural language processing Methods 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000010411 cooking Methods 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 238000013075 data extraction Methods 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 235000014347 soups Nutrition 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
-
- 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
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/26—Speech to text systems
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
- G10L2015/223—Execution procedure of a spoken command
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computational Linguistics (AREA)
- General Engineering & Computer Science (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Human Computer Interaction (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Mathematical Physics (AREA)
- Multimedia (AREA)
- Acoustics & Sound (AREA)
- General Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The application relates to a dialogue data processing method, a device, a storage medium and computer equipment, which are applied to a dialogue system, wherein the method comprises the following steps: receiving a current text through an interpreter, and extracting key information in the current text, wherein the current text is data generated according to received user data; marking key information through predefined dialogue logic in the tracker; performing operation strategy matching on the marked key information through a strategy manager to obtain a target operation strategy; and executing the target operation strategy in a preset knowledge map database through an operator. According to the method and the system, the user information is combined with the historical interaction information, the user information set and the domain knowledge set in the knowledge base, the intellectual relation is established, the association between the knowledge map database and the data is enhanced, the data required by the user can be inferred when the key information corresponding to the user data is ambiguous, the conversation interaction process is simplified, and more accurate target data can be accurately or speculatively acquired.
Description
Technical Field
The present application relates to the field of search technologies, and in particular, to a method and an apparatus for processing session data, a storage medium, and a computer device.
Background
The way in which people express demands is changed in internet technology by expressing demands in a voice manner. The voice intelligent equipment is more and more abundant in variety, and intelligent voice functions are more and more, for example, voice control or voice search functions such as a home robot, a voice guidance kitchen, and a voice shopping guide. However, in the prior art, voice control or voice search is mainly based on logic search and template matching, and a large amount of logic judgment needs to be configured, so that the process of obtaining a search result becomes complicated when a voice conversation is complicated, and the search result is not accurate enough and cannot necessarily meet the user requirements; when the requirement is expressed in a voice mode, sometimes the intention of the user for voice expression may not be specific enough or fuzzy and wide in range, and the search result obtained during fuzzy retrieval is not accurate enough, so that the requirement for accurate search cannot be met, the actual requirement of the user cannot be met, and the user experience is poor.
Disclosure of Invention
In order to solve the technical problems that the search is complex and the search result is not accurate enough due to logic search or template matching, embodiments of the present application provide a method and an apparatus for processing dialog data, a storage medium, and a computer device.
In a first aspect, an embodiment of the present application provides a dialog data processing method, which is applied to a dialog system, where the dialog system includes an interpreter, a tracker, a policy manager, and an operator, and the method includes:
receiving a current text through an interpreter, and extracting key information in the current text, wherein the current text is data generated according to received user data;
marking key information through predefined dialogue logic in the tracker;
performing operation strategy matching on the marked key information through a strategy manager to obtain a target operation strategy;
and executing the target operation strategy in a preset knowledge map database through an operator.
Optionally, the dialog system includes a plurality of interpreters, the dialog system further includes a mode selector, and the method further includes:
receiving an original text through a mode selector, wherein the original text carries an equipment identifier;
determining a corresponding target interpreter according to the equipment identifier;
receiving, by an interpreter, current text, comprising: the current text is received by the target interpreter, the current text being the original text or a portion of the original text.
Optionally, the dialog system further includes a data receiver and a voice recognition gateway, and the method further includes:
receiving user data by a data receiver;
and when the user data is voice data, recognizing the voice data of the user through the voice recognition gateway to obtain an original text.
Optionally, a data type judger is further included in the dialog system, and the method further includes:
judging the data type of the user data through a data type judger;
when the data type is a text type, taking the user data as an original text;
and when the data type is the voice type, identifying the voice data of the user through the voice identification gateway to obtain an original text.
Optionally, the key information is marked by predefined dialog logic in the tracker, including:
and marking key information by historical text of the current text stored in the tracker and dialog logic, wherein the historical text and the current text are the same session text.
Optionally, the preset knowledge profile database comprises a domain knowledge profile database and a personal knowledge profile database, the target operation policy is a first operation policy or a second operation policy,
executing a target operation strategy in a preset knowledge graph database through an operator, wherein the target operation strategy comprises the following steps:
executing the first operation strategy in the domain knowledge profile database through the operator when the target operation strategy is the first operation strategy,
and when the target operation strategy is a second operation strategy, executing the second operation strategy in the personal knowledge map database through the operator to obtain a third operation strategy, and executing the third operation strategy in the domain knowledge map database through the operator.
Optionally, the method further comprises:
obtaining target data according to the execution target operation strategy;
and displaying the target data.
In a second aspect, an embodiment of the present application provides a dialog data processing apparatus, which is applied to a dialog system, where the dialog system includes an interpreter, a tracker, a policy manager, and an operator, and the apparatus includes:
the receiving and extracting module is used for receiving the current text through the interpreter and extracting key information in the current text, wherein the current text is data generated according to the received user data;
the marking module is used for marking key information through conversation logic defined in advance in the tracker;
the matching module is used for matching the operation strategies to the marked key information through the strategy manager to obtain a target operation strategy;
and the execution module is used for executing the target operation strategy in the preset knowledge map database through the operator.
In a third aspect, embodiments of the present application provide a computer-readable storage medium, on which a computer program is stored, which, when executed by a processor, causes the processor to perform the steps of the method according to any one of the preceding claims.
In a fourth aspect, embodiments of the present application provide a computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor executing the program to perform the steps of the method according to any of the preceding claims.
Compared with the prior art, the technical scheme provided by the embodiment of the application has the following advantages:
the method comprises the steps that a current text is received through an interpreter, and key information in the current text is extracted, wherein the current text is data generated according to received user data; marking key information through predefined dialogue logic in the tracker; performing operation strategy matching on the marked key information through a strategy manager to obtain a target operation strategy; and executing the target operation strategy in a preset knowledge map database through an operator. The knowledge graph, the database and the data are combined to establish a knowledge connection, the association between the knowledge graph database and the data is enhanced, the data required by the user can be inferred when the key information corresponding to the user data is unclear, accurate and rational data extraction is realized, a large amount of logic judgment is not needed, and the conversation interaction process is simplified.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and together with the description, serve to explain the principles of the invention.
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without inventive exercise.
FIG. 1 is an architecture diagram of a conversational data processing system, in one embodiment;
FIG. 2 is a flow diagram illustrating a method for session data processing in one embodiment;
FIG. 3 is a block diagram showing a configuration of a dialogue data processing apparatus according to an embodiment;
FIG. 4 is a block diagram of a session data processing system in one embodiment;
FIG. 5 is a diagram illustrating an internal structure of a computer device according to an embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
FIG. 1 is an architecture diagram of a conversational data processing system, in one embodiment; referring to fig. 1, the dialogue data processing system includes a front-end interface layer 10, an application service layer 20, a dialogue system layer 30, and a data storage layer 40.
Where the front-end interface layer 10 includes a search box, which may be used to receive a single voice search or a single text search, and a dialog box, which may be used to receive a dialog voice search or a dialog text search for multiple back and forth dialogs, where the user may select whether to use the search box mode or the dialog box mode.
The application service layer 20 includes an access service module for receiving the user data received by the front-end interface layer, and a voice recognition service module for recognizing whether the user data is text data or voice data.
The dialogue system layer 30 includes a semantic understanding module, a dialogue management module, and an information acquisition module, where the semantic understanding module may be configured to perform semantic parsing on user data and extract key information through a natural language processing technique, and the dialogue management module and the information acquisition module may be configured to acquire an operation policy according to the key information and acquire target data from the data storage layer 40 according to the operation policy. The data storage layer 40 includes a domain knowledge map, a domain database, a historical interaction data knowledge map, a historical interaction data database, a user information knowledge map, and a user information database.
The dialogue data processing system can realize voice recognition and analysis, can also realize logical reasoning, automatic matching and query, and stores various user information, historical interactive information of all users and field knowledge information of various fields, wherein the user information comprises age, gender, occupation, hobbies and the like, the historical interactive information comprises various interactive data used by each user through a network, and the field knowledge information comprises various resource information such as picture resources, music resources, character resources and the like.
In one embodiment, the entire dialog data processing system may be a shared search system. And the data processing system is used for acquiring the target data according to the user data. The dialogue data processing system may communicate with the intelligent device. The device can be a terminal (such as a mobile phone, a tablet, a computer, a notebook, etc.) with an input device and a display device, the device can also be an intelligent electrical appliance (such as an intelligent television, an intelligent kitchen ware, etc.) with an input device and a display device, the input device is used for receiving text data input by a user, the display device is used for displaying target data returned by the dialogue data processing system, and the displayed target data can be visual contents such as pictures, characters, videos, recipes and the like. The device may be a terminal (e.g., a mobile phone, a tablet, a computer, a notebook, etc.) with a voice acquisition device and a voice playing device, or the device may be an intelligent electrical appliance (e.g., an intelligent sound box, an intelligent television) with a voice acquisition device and a voice playing device, where the voice acquisition device is used to receive voice data of a user, the voice playing device is used to play target data, and the played target data may be music, weather report, news, etc. The device may also be an intelligent appliance with a voice acquisition device and a voice playing device, such as an intelligent refrigerator and an intelligent air conditioner, and is configured to perform a corresponding operation according to target data, where the target data is a control instruction, such as: adjusting the temperature up, switching the operation mode, etc.
The application adopts a multilayer micro-service architecture and is divided into 4 layers: front-end interfaces, application services, session services, and data storage. The architecture can enable each layer of service to run independently, reduce the coupling of the service and update each layer of service independently. And the knowledge graph is combined with a relational database to realize the universal domain accurate search. And the voice conversation is combined with the search, so that the multi-parameter information extraction is realized.
The dialog data processing system of the present application can be applied to dialog interaction tasks, such as: voice shopping guide, voice kitchen guide, home robot and the like.
FIG. 2 is a flow diagram illustrating a method for session data processing in one embodiment; referring to fig. 2, the method includes the steps of:
s100: the current text is received through the interpreter, and key information in the current text is extracted.
Specifically, the current text is data generated from the received user data. The interpreter is used for analyzing the meaning of the current text, namely extracting the intention in the current text. The current text may be obtained by receiving user data sent by the smart device. For example: the user data may be user data sent through a mobile terminal such as a mobile phone or a tablet, for example: "i want to listen to a song of three", "i want to see an animation", "what weather today", may also be user data sent through the smart speaker, for example: "i want to listen to the song of lie four", may also be user data transmitted through the smart tv, for example: "i want to watch a movie". User data sent by the smart air conditioner may also be, for example: "adjust the air conditioner to the temperature of not cold or hot"; user data may be transmitted via the smart cookware, for example: regulating the fire of a kitchen gas stove to the degree of heating for cooking soup and the like.
Key information can be analyzed from the current text, for example, "i want to listen to a song of three pages" can be analyzed as "listen", "song" and "three pages", and the key information constitutes an intention. For example: "what weather today" can resolve that the key information is "today" and "weather", and the key information constitutes the intention. The key information characterizes the intent.
S200: the key information is marked by predefined dialog logic in the tracker.
Specifically, the key information is marked by historical text of the current text stored in the tracker and dialog logic, wherein the historical text and the current text are the same conversation text.
The tracker stores the historical text of the current text in the same conversation, and can track the state of the current text by combining the historical text of the current text. The tracker is used for tracking the current text, acquiring the state of the current text, and performing identification and storage operation on the current text. The defined dialog logic is for example: the dialog logic for listening to music is "play". The dialog logic is based on the actual situation and has no restrictions whatsoever. The dialog logic may specifically be defined in accordance with natural language processing techniques.
S300: and performing operation strategy matching on the marked key information through a strategy manager to obtain a target operation strategy.
Specifically, a step of finding target data desired by the user needs to be determined. The operation policy may specifically be through which steps the desired target data is found. For example, when the key information is clear, the target data is obtained by only one search. When the key information is not clear, not directly returning inaccurate target data, re-inquiring the user to obtain clear key information until the obtained key information is clear, and then giving out final accurate target data; alternatively, when the key information is ambiguous, the user is not re-queried for unambiguous key information, but the target data is inferred through the knowledge graph database.
S400: and executing the target operation strategy in a preset knowledge map database through an operator.
Specifically, the knowledge graph database may be a large data storage area in which a knowledge graph is combined with a database, and includes a user information set, a historical interaction information set, and a domain knowledge set, which are associated with each other. The user information set is stored in a user information knowledge graph and a user information database; the historical interaction information set is stored in a historical interaction data knowledge graph and a historical interaction data database; the set of domain knowledge is stored in a domain knowledge graph and a domain database. When the key information is clear, the accurate target data can be directly matched from the domain knowledge set through the key information; when the key information is ambiguous, the target data is obtained by reasoning according to the user information and the user information set, the historical interaction information set and the domain knowledge set, or a response result with conversation property is obtained according to the key information, the response result with conversation property is returned to the user to guide the user to express more accurate key information, and then accurate target data is obtained through matching according to the more accurate key information. The operator can be customized by a user, and mainly specifies key information and related operations corresponding to a target operation strategy under the dialogue logic for response. The operator is also used for storing the user information, the key information, the corresponding target operation strategy and other information carried in the current text into a knowledge map database. And also for transmitting the marked key information back to the tracker for storage.
In one embodiment, the dialog system includes a plurality of interpreters, the dialog system further includes a mode selector, and the method further includes: receiving an original text through a mode selector, wherein the original text carries an equipment identifier; determining a corresponding target interpreter according to the equipment identifier; receiving, by an interpreter, current text, comprising: the current text is received by the target interpreter, the current text being the original text or a portion of the original text.
Specifically, the dialogue data processing system may be a shared server, and various devices may interact with the shared server, and corresponding interpreters may be set according to the types of the devices, that is, one type of device shares one interpreter, and interpreters corresponding to different types of devices are different, so that the interpreters are targeted, and more accurate in parsing the original text. Specifically, the target interpreter can be matched and selected through the device identifier carried by the original text. The current text is then received by the target interpreter.
In a specific embodiment, the dialog system further includes a data receiver and a voice recognition gateway, and the method further includes: receiving user data by a data receiver; and when the user data is voice data, recognizing the voice data of the user through the voice recognition gateway to obtain an original text.
Specifically, the data receiver is connected with the interpreter through the voice recognition gateway and the mode selector in sequence. The data receiver is used for receiving user data, when the user data is voice type data, the language of the user data may be Chinese or different languages of different countries such as English, French, German, Japanese, Korean, etc., so that it is required to identify the language type, and then call different language databases according to the specific language type to convert the user data into original text in text form.
In a specific embodiment, the dialog system further includes a data type judger, and the method further includes: judging the data type of the user data through a data type judger; when the data type is a text type, taking the user data as an original text; and when the data type is the voice type, identifying the voice data of the user through the voice identification gateway to obtain an original text.
Specifically, the data receiver is connected with the interpreter through the data type judger, the voice recognition gateway and the mode selector in sequence. The data receiver is also connected with the interpreter through the data type judger and the mode selector in turn.
The data type determiner is for determining whether the user data is text type data or voice type data. If the user data is text type data, the data type judger directly sends the user data as original text to the mode selector. If the user data is voice type data, the data type judger firstly sends the user data to the voice recognition gateway, and the voice recognition gateway converts the user data into original text in a text form and then sends the original text to the mode selector.
In a specific embodiment, the preset knowledge profile database includes a domain knowledge profile database and a personal knowledge profile database, and the target operation policy is a first operation policy or a second operation policy.
Specifically, the domain knowledge profile database includes: a domain knowledge graph, a domain database, a personal knowledge graph database comprising: the system comprises a historical interactive data knowledge map, a historical interactive data database, a user information knowledge map and a user information database which are mutually associated.
The first operation policy is an operation performed when the intent characterized by the key information is clear. The second operational policy is an operation performed when the intent characterized by the key information is ambiguous.
Step S400 specifically includes:
executing the first operation strategy in the domain knowledge profile database through the operator when the target operation strategy is the first operation strategy,
and when the target operation strategy is a second operation strategy, executing the second operation strategy in the personal knowledge map database through the operator to obtain a third operation strategy, and executing the third operation strategy in the domain knowledge map database through the operator.
Specifically, the operator is used to perform a specific application operation. When the key information is clear, searching the domain knowledge map database for target data matching the key information through the operator. For example: and if the key information is that the user wants to see the western notes, searching related video data of the western notes in a domain knowledge map database.
For example, "i want to listen to the national song," the audio data of the national song is searched out.
And when the key information is ambiguous, executing a second operation strategy in the personal knowledge graph database through the operator, wherein the second operation strategy is a third operation strategy obtained by reasoning through the user information corresponding to the current text and the user information set and the historical interaction information set in the personal knowledge graph database. The third operation strategy is an operation executed after obtaining clear key information through reasoning. At this time, the key information is clarified by derivation, so that target data matching the inferred key information can be searched in the domain knowledge map database by the operator. The specific reasoning process is as follows: the characteristic data of the user can be obtained from a personal knowledge map database, and comprises the following steps: the personal knowledge map database also comprises feature data of other users, the feature data of the user is matched with the feature data of the other users, similar users closest to the feature data of the user can be obtained, and operation strategies associated with the similar users under the same key information can be used as third operation strategies corresponding to the user.
Of course, similar key information closest to the key information in the historical interaction data of the user may also be obtained according to the historical interaction data of the user, and the operation policy corresponding to the similar key information is used as the third operation policy. For example: the key information is 'i want to listen to three songs', and the three songs listened in the historical interactive data of the user can be recommended to the user because the intention of the key information is not clear. Specifically, the song with the largest playing times can be recommended to the user.
In a particular embodiment, the method further comprises: obtaining target data according to the execution target operation strategy; and displaying the target data.
Specifically, the target data is displayed after being obtained. The target data can be various data such as voice broadcast such as pictures, characters, music, videos and weather broadcast, news and instructions.
The display target data can be displayed by the terminal device by sending the target data to the terminal device. For example: the target data is presented to the user by the device in speech form or in a visual display.
For example, "i want to see the shorthand of western book", the video of "shorthand of western book" is fed back to the device, and the device plays the video of "shorthand of western book" for the user to watch.
For example, "i want to listen to the national song," the audio of the national song is fed back to the device, and the device plays the national song to the user.
For example: and adjusting the air conditioner to the non-cold and non-hot temperature, wherein the non-cold and non-hot temperature obtained by analysis is 26 ℃, the target data is a control instruction, and the control instruction is issued to the air conditioner to control the air conditioner to adjust the temperature to 26 ℃.
In one particular embodiment, the user may select a search box to determine whether it is a single voice search or a single text search; the dialog box may also be selected to determine whether a dialog voice search or a dialog text search of multiple back and forth dialogs is received. The single voice search and the single text search may be the same type of search mode, and the dialogue voice search and the dialogue text search may be the same type of search mode. The single search is realized through the single voice search or the single text search, namely, after the user data is received, the system directly obtains the target data according to the user data and returns the target data to the user. The dialogue search of both parties with questions and answers is realized through dialogue voice search or dialogue text search, and the user data can be received for multiple times until the user obtains satisfactory target data and then does not wake up the equipment any more or does not receive the user data any more in continuous time or the user closes the search function.
For example: the user data is 'I want to listen to song A of Zhang III', the key information is clear at this time, and only the audio data of Zhang III song A needs to be found through matching, and the audio data of Zhang III song A is returned to the user. When the user data is 'i want to listen to a song', if the user data is a single voice search, no response result exists, and only the target data obtained through recommendation is given. The recommendation and acquisition may specifically be to search the historical interaction data of the user to obtain target data closest to the key information, or to search target data closest to the key information in similar users most similar to the user.
For example: the user data is 'i want to listen to songs', if the dialogue voice search and the dialogue text search are performed, the key information is 'listen to songs', but the 'songs' range is too wide, so that the search intention is not clear, the response result is returned to the user at the moment, and the response result can be: "who songs you want to listen to" or "what type of songs you want to listen to? "and the like are not limited thereto. The dialogue search or interactive search with unclear key information or intention but too wide search range can set semantic analysis and dialogue rules, so that corresponding response results can be obtained, the user is guided to answer the intention more clearly, and the final target data is more in line with the user expectation. Of course, the search may also be obscured, giving the search results directly, and not asking the user for more detailed intentions.
With the arrival of the artificial intelligence era, users expect simpler and more natural ways to interact, and the existing dialogue interaction cannot perform accurate searching and reasoning functions. Therefore, the technology of the application is based on the basic function of the dialogue system, combines the knowledge map technology, solves the problems that the dialogue interaction process is complicated, the user cannot find the requirement quickly, and the like, and improves the voice dialogue interaction experience.
FIG. 3 is a block diagram showing a configuration of a dialogue data processing apparatus according to an embodiment; referring to fig. 3, the apparatus includes:
a receiving and extracting module 100, configured to receive a current text through an interpreter, and extract key information in the current text, where the current text is data generated according to received user data;
a tagging module 200 for tagging key information through predefined dialog logic in the tracker;
the matching module 300 is configured to perform operation policy matching on the marked key information through a policy manager to obtain a target operation policy;
an executing module 400, configured to execute, by an operator, a target operation policy in a preset knowledge graph database.
In one embodiment, the dialog system includes a plurality of interpreters, the dialog system further includes a mode selector, the apparatus further includes:
the text receiving module is used for receiving an original text through the mode selector, and the original text carries an equipment identifier;
the selection module is used for determining a corresponding target interpreter according to the equipment identifier;
the receiving and extracting module 100 is specifically configured to receive a current text through the target interpreter, where the current text is an original text or a partial text of the original text.
In a specific embodiment, the dialog system further includes a data receiver and a voice recognition gateway, and the apparatus further includes:
a user data receiving module for receiving user data through a data receiver;
and the conversion module is used for identifying the voice data of the user through the voice identification gateway when the user data is the voice data to obtain an original text.
In a specific embodiment, the dialog system further includes a data type judger, and the apparatus further includes:
and the judging module is used for judging the data type of the user data through the data type judger, taking the user data as an original text when the data type is a text type, and entering the converting module when the data type is a voice type.
In one embodiment, the marking module 200 is specifically configured to: and marking key information by historical text of the current text stored in the tracker and dialog logic, wherein the historical text and the current text are the same session text.
In one embodiment, the preset knowledge profile database includes a domain knowledge profile database and a personal knowledge profile database, the target operation policy is a first operation policy or a second operation policy,
the execution module 400 includes:
a first executing module for executing the first operation policy in the domain knowledge profile database through the operator when the target operation policy is the first operation policy,
and the second execution module is used for executing the second operation strategy in the personal knowledge graph database through the operator to obtain a third operation strategy when the target operation strategy is the second operation strategy, and executing the third operation strategy in the domain knowledge graph database through the operator.
In a specific embodiment, the apparatus further comprises:
the data display module is used for obtaining target data according to the execution target operation strategy; and displaying the target data.
FIG. 4 is a block diagram of a session data processing system in one embodiment; referring to fig. 4, the dialogue data processing system includes: the system comprises a data receiver, a data type judger, a voice recognition gateway, a mode selector, an interpreter, a tracker, a strategy manager and an operator, and further comprises a domain knowledge map, a domain database, a historical interaction data knowledge map, a historical interaction data database, a user information knowledge map and a user information database, wherein the domain knowledge map, the domain database, the historical interaction data knowledge map and the user information database are used for storing data; the historical interactive data knowledge map, the historical interactive data database, the user information knowledge map and the user information database form a personal knowledge map database.
The method can also be applied to terminal equipment, for example, a mobile phone which receives user data, and displays the searched target data to a client in various forms such as pictures, videos or voices through a relevant module of the mobile phone by the conversation data processing method. For example: the songs that the user wants to listen to can be played, the videos that the user wants to watch can be played through videos, the webpages that the user wants to browse can be displayed, and the like.
The method can also be applied to intelligent household appliances, intelligent shopping guides, intelligent kitchen guides and the like in intelligent homes, for example, in a voice control air conditioner, the user data is voice type data, for example, "temperature is reduced a little", the response result returned by the air conditioner is "how much degree you want to reduce", the user returns the user data again as "reduction of 1 degree", and the air conditioner automatically reduces the temperature by 1 degree on the basis of the current temperature. It may also be, for example, turning on an air conditioner, adjusting the temperature, opening a window, turning on a light, adjusting the brightness, etc.
Fig. 5 is an internal configuration diagram of a computer device in an embodiment, referring to fig. 5, the computer device includes a processor, a memory, a network interface, an input device, a voice device, and a display screen, which are connected via a system bus. Wherein the memory includes a non-volatile storage medium and an internal memory. The non-volatile storage medium of the computer device stores an operating system and may also store a computer program that, when executed by the processor, causes the processor to implement the dialogue data processing method. The internal memory may also have a computer program stored therein, which, when executed by the processor, causes the processor to perform the session data processing method. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, a key, a track ball or a touch pad arranged on the shell of the computer equipment, an external keyboard, a touch pad or a mouse and the like. The speech device of the computer equipment can comprise a speech acquisition function, a speech processing function, a speech analysis function and a speech conversion function.
Those skilled in the art will appreciate that the architecture shown in fig. 5 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
In one embodiment, the dialogue data processing apparatus provided in the present application may be implemented in the form of a computer program, which is executable on a computer device as shown in fig. 5. The memory of the computer device may store various program modules constituting the dialogue data processing apparatus, such as the reception extraction module 100, the marking module 200, the matching module 300, and the execution module 400 shown in fig. 3. The computer program constituted by the respective program modules causes the processor to execute the steps in the dialogue data processing of the respective embodiments of the present application described in the present specification.
In one embodiment, a computer device is provided, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the following steps when executing the computer program: receiving a current text through an interpreter, and extracting key information in the current text, wherein the current text is data generated according to received user data; marking key information through predefined dialogue logic in the tracker; performing operation strategy matching on the marked key information through a strategy manager to obtain a target operation strategy; and executing the target operation strategy in a preset knowledge map database through an operator.
In one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of: receiving a current text through an interpreter, and extracting key information in the current text, wherein the current text is data generated according to received user data; marking key information through predefined dialogue logic in the tracker; performing operation strategy matching on the marked key information through a strategy manager to obtain a target operation strategy; and executing the target operation strategy in a preset knowledge map database through an operator.
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 hardware related to instructions of a computer program, and the program can be stored in a non-volatile computer readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. 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), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
It is noted that, in this document, relational terms such as "first" and "second," and the like, may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
The foregoing are merely exemplary embodiments of the present invention, which enable those skilled in the art to understand or practice the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. A dialogue data processing method is applied to a dialogue system, wherein the dialogue system comprises an interpreter, a tracker, a policy manager and an operator, and the method comprises the following steps:
receiving a current text through an interpreter, and extracting key information in the current text, wherein the current text is data generated according to received user data;
marking key information through predefined dialogue logic in the tracker;
performing operation strategy matching on the marked key information through a strategy manager to obtain a target operation strategy;
and executing the target operation strategy in a preset knowledge map database through an operator.
2. The method of claim 1, wherein the dialog system includes a plurality of interpreters, the dialog system further including a mode selector, the method further comprising:
receiving an original text through a mode selector, wherein the original text carries an equipment identifier;
determining a corresponding target interpreter according to the equipment identifier;
receiving, by an interpreter, current text, comprising: the current text is received by the target interpreter, the current text being the original text or a portion of the original text.
3. The method of claim 2, wherein the dialog system further comprises a data receiver and a speech recognition gateway, the method further comprising:
receiving user data by a data receiver;
and when the user data is voice data, recognizing the voice data of the user through the voice recognition gateway to obtain an original text.
4. The method of claim 3, further comprising a data type determiner in the dialog system, the method further comprising:
judging the data type of the user data through a data type judger;
when the data type is a text type, taking the user data as an original text;
and when the data type is the voice type, identifying the voice data of the user through the voice identification gateway to obtain an original text.
5. The method of claim 1, wherein the tagging of key information by predefined dialog logic in the tracker comprises:
and marking key information by historical text of the current text stored in the tracker and dialog logic, wherein the historical text and the current text are the same session text.
6. The method of claim 1, wherein the preset knowledge profile database includes a domain knowledge profile database and a personal knowledge profile database, the target operation policy is a first operation policy or a second operation policy,
executing a target operation strategy in a preset knowledge graph database through an operator, wherein the target operation strategy comprises the following steps:
executing the first operation strategy in the domain knowledge profile database through the operator when the target operation strategy is the first operation strategy,
and when the target operation strategy is a second operation strategy, executing the second operation strategy in the personal knowledge map database through the operator to obtain a third operation strategy, and executing the third operation strategy in the domain knowledge map database through the operator.
7. The method of claim 1, wherein the method further comprises:
obtaining target data according to the execution target operation strategy;
and displaying the target data.
8. A dialogue data processing device, which is applied to a dialogue system, wherein the dialogue system comprises an interpreter, a tracker, a policy manager and an operator, and the device comprises:
the receiving and extracting module is used for receiving the current text through the interpreter and extracting key information in the current text, wherein the current text is data generated according to the received user data;
the marking module is used for marking key information through conversation logic defined in advance in the tracker;
the matching module is used for matching the operation strategies to the marked key information through the strategy manager to obtain a target operation strategy;
and the execution module is used for executing the target operation strategy in the preset knowledge map database through the operator.
9. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, causes the processor to carry out the steps of the method according to any one of claims 1 to 7.
10. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor executes the program to perform the steps of the method according to any of claims 1-7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010486061.9A CN111816173B (en) | 2020-06-01 | 2020-06-01 | Dialogue data processing method and device, storage medium and computer equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010486061.9A CN111816173B (en) | 2020-06-01 | 2020-06-01 | Dialogue data processing method and device, storage medium and computer equipment |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111816173A true CN111816173A (en) | 2020-10-23 |
CN111816173B CN111816173B (en) | 2024-06-07 |
Family
ID=72848582
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010486061.9A Active CN111816173B (en) | 2020-06-01 | 2020-06-01 | Dialogue data processing method and device, storage medium and computer equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111816173B (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112732951A (en) * | 2020-12-30 | 2021-04-30 | 青岛海信智慧生活科技股份有限公司 | Man-machine interaction method and device |
CN113098699A (en) * | 2021-03-30 | 2021-07-09 | 四川巧夺天工信息安全智能设备有限公司 | User information distribution and pairing method facing intelligent terminal |
CN113688220A (en) * | 2021-09-02 | 2021-11-23 | 国家电网有限公司客户服务中心 | Text robot dialogue method and system based on semantic understanding |
CN114610856A (en) * | 2022-03-16 | 2022-06-10 | 零犀(北京)科技有限公司 | Dialog interaction intelligent decision-making method and device based on causal graph |
CN116701662A (en) * | 2023-08-07 | 2023-09-05 | 国网浙江浙电招标咨询有限公司 | Knowledge graph-based supply chain data management method, device, equipment and medium |
Citations (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1313972A (en) * | 1998-08-24 | 2001-09-19 | Bcl计算机有限公司 | Adaptive natural language interface |
US20050010393A1 (en) * | 2001-10-30 | 2005-01-13 | Morena Danieli | Method for managing mixed initiative human-machine dialogues based on interactive speech |
CN105931638A (en) * | 2016-04-26 | 2016-09-07 | 北京光年无限科技有限公司 | Intelligent-robot-oriented dialog system data processing method and device |
CN106228983A (en) * | 2016-08-23 | 2016-12-14 | 北京谛听机器人科技有限公司 | Scene process method and system during a kind of man-machine natural language is mutual |
CN107590120A (en) * | 2016-07-07 | 2018-01-16 | 深圳狗尾草智能科技有限公司 | Artificial intelligence process method and device |
US20180025726A1 (en) * | 2016-07-22 | 2018-01-25 | International Business Machines Corporation | Creating coordinated multi-chatbots using natural dialogues by means of knowledge base |
CN108197167A (en) * | 2017-12-18 | 2018-06-22 | 深圳前海微众银行股份有限公司 | Human-computer dialogue processing method, equipment and readable storage medium storing program for executing |
CN108646918A (en) * | 2018-05-10 | 2018-10-12 | 北京光年无限科技有限公司 | Visual interactive method and system based on visual human |
CN109002510A (en) * | 2018-06-29 | 2018-12-14 | 北京百度网讯科技有限公司 | A kind of dialog process method, apparatus, equipment and medium |
CN109241250A (en) * | 2018-07-25 | 2019-01-18 | 南京瓦尔基里网络科技有限公司 | A kind of dialogue of policing rule promotes and intention method of discrimination and system |
CN110222162A (en) * | 2019-05-10 | 2019-09-10 | 天津中科智能识别产业技术研究院有限公司 | A kind of intelligent answer method based on natural language processing and knowledge mapping |
CN110313154A (en) * | 2017-02-14 | 2019-10-08 | 微软技术许可有限责任公司 | Intelligent assistant with the information discrimination based on intention |
CN110556097A (en) * | 2018-06-01 | 2019-12-10 | 声音猎手公司 | Customizing acoustic models |
CN110640764A (en) * | 2019-11-04 | 2020-01-03 | 卓谨信息科技(常州)有限公司 | Intelligent robot control system and working method thereof |
CN111026857A (en) * | 2019-12-10 | 2020-04-17 | 苏州思必驰信息科技有限公司 | Conversation state tracking method, man-machine conversation method and system |
CN111159371A (en) * | 2019-12-21 | 2020-05-15 | 华南理工大学 | Dialogue strategy method for task-oriented dialogue system |
-
2020
- 2020-06-01 CN CN202010486061.9A patent/CN111816173B/en active Active
Patent Citations (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1313972A (en) * | 1998-08-24 | 2001-09-19 | Bcl计算机有限公司 | Adaptive natural language interface |
US20050010393A1 (en) * | 2001-10-30 | 2005-01-13 | Morena Danieli | Method for managing mixed initiative human-machine dialogues based on interactive speech |
CN105931638A (en) * | 2016-04-26 | 2016-09-07 | 北京光年无限科技有限公司 | Intelligent-robot-oriented dialog system data processing method and device |
CN107590120A (en) * | 2016-07-07 | 2018-01-16 | 深圳狗尾草智能科技有限公司 | Artificial intelligence process method and device |
US20180025726A1 (en) * | 2016-07-22 | 2018-01-25 | International Business Machines Corporation | Creating coordinated multi-chatbots using natural dialogues by means of knowledge base |
CN106228983A (en) * | 2016-08-23 | 2016-12-14 | 北京谛听机器人科技有限公司 | Scene process method and system during a kind of man-machine natural language is mutual |
CN110313154A (en) * | 2017-02-14 | 2019-10-08 | 微软技术许可有限责任公司 | Intelligent assistant with the information discrimination based on intention |
CN108197167A (en) * | 2017-12-18 | 2018-06-22 | 深圳前海微众银行股份有限公司 | Human-computer dialogue processing method, equipment and readable storage medium storing program for executing |
CN108646918A (en) * | 2018-05-10 | 2018-10-12 | 北京光年无限科技有限公司 | Visual interactive method and system based on visual human |
CN110556097A (en) * | 2018-06-01 | 2019-12-10 | 声音猎手公司 | Customizing acoustic models |
CN109002510A (en) * | 2018-06-29 | 2018-12-14 | 北京百度网讯科技有限公司 | A kind of dialog process method, apparatus, equipment and medium |
CN109241250A (en) * | 2018-07-25 | 2019-01-18 | 南京瓦尔基里网络科技有限公司 | A kind of dialogue of policing rule promotes and intention method of discrimination and system |
CN110222162A (en) * | 2019-05-10 | 2019-09-10 | 天津中科智能识别产业技术研究院有限公司 | A kind of intelligent answer method based on natural language processing and knowledge mapping |
CN110640764A (en) * | 2019-11-04 | 2020-01-03 | 卓谨信息科技(常州)有限公司 | Intelligent robot control system and working method thereof |
CN111026857A (en) * | 2019-12-10 | 2020-04-17 | 苏州思必驰信息科技有限公司 | Conversation state tracking method, man-machine conversation method and system |
CN111159371A (en) * | 2019-12-21 | 2020-05-15 | 华南理工大学 | Dialogue strategy method for task-oriented dialogue system |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112732951A (en) * | 2020-12-30 | 2021-04-30 | 青岛海信智慧生活科技股份有限公司 | Man-machine interaction method and device |
CN113098699A (en) * | 2021-03-30 | 2021-07-09 | 四川巧夺天工信息安全智能设备有限公司 | User information distribution and pairing method facing intelligent terminal |
CN113688220A (en) * | 2021-09-02 | 2021-11-23 | 国家电网有限公司客户服务中心 | Text robot dialogue method and system based on semantic understanding |
CN113688220B (en) * | 2021-09-02 | 2022-05-24 | 国家电网有限公司客户服务中心 | Text robot dialogue method and system based on semantic understanding |
CN114610856A (en) * | 2022-03-16 | 2022-06-10 | 零犀(北京)科技有限公司 | Dialog interaction intelligent decision-making method and device based on causal graph |
CN116701662A (en) * | 2023-08-07 | 2023-09-05 | 国网浙江浙电招标咨询有限公司 | Knowledge graph-based supply chain data management method, device, equipment and medium |
Also Published As
Publication number | Publication date |
---|---|
CN111816173B (en) | 2024-06-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111816173B (en) | Dialogue data processing method and device, storage medium and computer equipment | |
US10922355B2 (en) | Method and apparatus for recommending news | |
WO2020192450A1 (en) | Interactive message processing method and apparatus, computer device and storage medium | |
US20200356928A1 (en) | Collaborative personal assistant system for delegating provision of services by third party task providers and method therefor | |
JP7335062B2 (en) | Voice service providing method and apparatus | |
US10650816B2 (en) | Performing tasks and returning audio and visual feedbacks based on voice command | |
US10547571B2 (en) | Message service providing method for message service linked to search service and message server and user terminal to perform the method | |
CN109165302B (en) | Multimedia file recommendation method and device | |
CN110797022B (en) | Application control method, device, terminal and server | |
KR102340228B1 (en) | Message service providing method for message service linking search service and message server and user device for performing the method | |
US20160055245A1 (en) | Systems and methods for providing information discovery and retrieval | |
US20040199394A1 (en) | Speech input system, speech portal server, and speech input terminal | |
US20180239812A1 (en) | Method and apparatus for processing question-and-answer information, storage medium and device | |
CN104199810A (en) | Intelligent service method and system based on natural language interaction | |
CN115424624B (en) | Man-machine interaction service processing method and device and related equipment | |
KR20130062799A (en) | Method for managing keyword information server | |
CN112579031A (en) | Voice interaction method and system and electronic equipment | |
CN113938755A (en) | Server, terminal device and resource recommendation method | |
CN111427444B (en) | Control method and device of intelligent device | |
CN112309387A (en) | Method and apparatus for processing information | |
CN110609956B (en) | Information search method, device, medium and equipment | |
US11776537B1 (en) | Natural language processing system for context-specific applier interface | |
CN117252730B (en) | Service subscription processing system, service subscription information processing method and device | |
KR102050377B1 (en) | A collaborative personal assistant system for delegating providing of services supported by third party task providers and method therefor | |
KR102091441B1 (en) | A collaborative personal assistant system for delegating providing of services supported by third party task providers and method therefor |
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 |