[go: up one dir, main page]

WO2020177592A1 - Procédé et dispositif de réponse à une question de peinture, système de réponse à une question de peinture, et support d'informations lisible - Google Patents

Procédé et dispositif de réponse à une question de peinture, système de réponse à une question de peinture, et support d'informations lisible Download PDF

Info

Publication number
WO2020177592A1
WO2020177592A1 PCT/CN2020/076780 CN2020076780W WO2020177592A1 WO 2020177592 A1 WO2020177592 A1 WO 2020177592A1 CN 2020076780 W CN2020076780 W CN 2020076780W WO 2020177592 A1 WO2020177592 A1 WO 2020177592A1
Authority
WO
WIPO (PCT)
Prior art keywords
painting
text information
answer
question
display terminal
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.)
Ceased
Application number
PCT/CN2020/076780
Other languages
English (en)
Chinese (zh)
Inventor
贾红红
胡风硕
徐楚
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
BOE Technology Group Co Ltd
Original Assignee
BOE Technology Group Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by BOE Technology Group Co Ltd filed Critical BOE Technology Group Co Ltd
Publication of WO2020177592A1 publication Critical patent/WO2020177592A1/fr
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
    • G06F3/048—Interaction techniques based on graphical user interfaces [GUI]
    • G06F3/0481—Interaction techniques based on graphical user interfaces [GUI] based on specific properties of the displayed interaction object or a metaphor-based environment, e.g. interaction with desktop elements like windows or icons, or assisted by a cursor's changing behaviour or appearance
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; 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
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/10—Text processing
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/20—Natural language analysis
    • G06F40/279—Recognition of textual entities
    • G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
    • G06F40/295—Named entity recognition
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; 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

Definitions

  • the embodiments of the present disclosure relate to a painting question answering method and device, a painting question answering system, and a readable storage medium.
  • online platforms such as online art galleries
  • online art galleries have received widespread public attention as platforms for viewing art paintings and sharing artistic creations. Users can enjoy art paintings through the online platform, and can perform operations such as inquiries and transactions on art paintings.
  • At least one embodiment of the present disclosure provides a painting question answering method, including: acquiring text information from a display terminal and acquiring the current dialogue state of the display terminal; performing semantic understanding on the text information to acquire the current intention of the user; The current dialogue state and the current intention determine a target dialogue state; and the answer data corresponding to the text information is obtained according to the target dialogue state.
  • the answer data includes painting data or extraction information used to extract the painting data.
  • the semantic understanding of the text information to obtain the current intention of the user includes: identifying the named entity of the text information; determining according to the named entity The vector to be recognized corresponding to the named entity; and based on the vector to be recognized, the intent of the standard feature vector that meets the requirements is determined as the current intent of the text information.
  • identifying the named entity of the text information includes: identifying the named entity of the text information through a named entity recognition model.
  • determining the vector to be recognized corresponding to the named entity according to the named entity includes: determining the vector corresponding to the named entity according to the named entity through a deep learning model The vector to be recognized.
  • determining the intent of the standard feature vector that meets the requirements as the current intent of the text information includes: The intent of the standard feature vector with the largest similarity of the vectors is determined as the current intent of the text information.
  • the text information is text content directly input by the user, or text content obtained after recognizing the voice information input by the user.
  • obtaining text information from the display terminal includes: receiving request data from the display terminal, and parsing the request data to obtain the text information.
  • the painting question answering method provided by at least one embodiment of the present disclosure further includes: sending the answer data to the display terminal.
  • the painting question answering method provided by at least one embodiment of the present disclosure further includes: recognizing voice information from the display terminal to obtain the text information.
  • At least one embodiment of the present disclosure also provides a painting question and answer device, including: a text information acquisition circuit configured to acquire text information from a display terminal and to acquire the current dialogue state of the display terminal; and a current intention acquisition circuit configured to In order to perform semantic understanding of the text information to obtain the current intention of the user; the target state determination circuit is configured to determine the target dialogue state according to the current dialogue state and the current intention; the answer data acquisition circuit is configured to The target dialogue state acquires answer data corresponding to the text information.
  • a text information acquisition circuit configured to acquire text information from a display terminal and to acquire the current dialogue state of the display terminal
  • a current intention acquisition circuit configured to In order to perform semantic understanding of the text information to obtain the current intention of the user
  • the target state determination circuit is configured to determine the target dialogue state according to the current dialogue state and the current intention
  • the answer data acquisition circuit is configured to The target dialogue state acquires answer data corresponding to the text information.
  • the current intention acquisition circuit includes: a named entity recognition circuit configured to recognize a named entity of the text information; and a recognition vector determination circuit configured according to The named entity determines the vector to be recognized corresponding to the named entity; the current intent determination circuit is configured to determine the intent of the standard feature vector that meets the requirements as the current intent of the text information based on the vector to be recognized.
  • At least one embodiment of the present disclosure also provides a painting question answering system, which includes a display terminal and a question answering server.
  • the display terminal is configured to send request data to the question and answer server;
  • the question and answer server is configured to obtain text information from the display terminal based on the request data, and to obtain the current dialogue state of the display terminal, And perform semantic understanding of the text information to obtain the current intention of the user; determine the target dialogue state according to the current dialogue state and the current intention; and obtain the answer data corresponding to the text information according to the target dialogue state;
  • the display terminal is also configured to receive the answer data and display painting data based on the answer data.
  • the answer data includes painting data or extraction information used to extract the painting data.
  • the painting question answering system provided by at least one embodiment of the present disclosure further includes a voice recognition server.
  • the voice recognition server is in signal connection with the display terminal, and is configured to recognize voice information from the display terminal to obtain the text information.
  • the question and answer server includes a WEB server and a semantic server; the question and answer server includes a WEB server and a semantic server; the WEB server is connected to the display terminal and the The semantic server signal connection; the WEB server is configured to receive request data from the display terminal, parse the request data to obtain the text information, and send the text information to the semantic server, and The answer data obtained by the semantic server according to the text information is returned to the display terminal via the WEB server; and the semantic server is configured to receive text information from the WEB server, and obtain the display terminal
  • the current dialogue state of the text information is understood to obtain the user’s current intention; the target dialogue state is determined according to the current dialogue state and the current intention; the text information corresponding to the text information is obtained according to the target dialogue state Answer data; and sending the answer data to the WEB server.
  • the painting question answering system provided by at least one embodiment of the present disclosure further includes a database server storing answer data.
  • the database server is configured to store the painting data.
  • the database server is in signal connection with the question-and-answer server, and is configured to respond to request information from the question-and-answer server to transfer painting data corresponding to the request information Return to the question and answer server, and the question and answer server sends the painting data as the answer data to the display terminal.
  • the database server is signally connected to the display terminal, and the display terminal is configured to use the extraction information to extract the painting data from the database server .
  • the database server is configured to respond to the extraction information sent by the display terminal, and send painting data corresponding to the extraction information to the display terminal.
  • At least one embodiment of the present disclosure further provides a painting question and answer device, including: a processor; a memory, including one or more computer program modules; wherein, the one or more computer program modules are stored in the memory and are It is configured to be executed by the processor, and the one or more computer program modules include instructions for implementing the painting question and answer method provided in the foregoing embodiment.
  • At least one embodiment of the present disclosure also provides a readable storage medium on which computer instructions are stored.
  • the computer instructions are executed by the processor, the painting question-and-answer method provided in the foregoing embodiment is implemented.
  • Fig. 1 is a block diagram of a painting question answering system according to an embodiment of the present disclosure
  • Fig. 2 is a schematic diagram showing a state transition table according to an embodiment of the present disclosure
  • FIG. 3 is a flowchart of obtaining answer data according to an embodiment of the present disclosure
  • Fig. 4 is a flowchart of obtaining current intentions according to an embodiment of the present disclosure
  • Fig. 5 is a flow chart showing obtaining current intention according to another embodiment of the present disclosure.
  • Fig. 6 is a block diagram of another painting question answering system according to an embodiment of the present disclosure.
  • FIG. 7 is a block diagram of another painting question and answer system according to an embodiment of the present disclosure.
  • Fig. 8 is a block diagram of another painting question answering system according to an embodiment of the present disclosure.
  • FIG. 9 is a block diagram of another painting question answering system according to an embodiment of the present disclosure.
  • Fig. 10 is a block diagram of a painting question answering device according to an embodiment of the present disclosure.
  • FIG. 11 is a schematic diagram of another painting question and answer device according to an embodiment of the present disclosure.
  • Fig. 12 is a schematic diagram showing a display terminal according to an embodiment of the present disclosure.
  • Fig. 13 is a schematic diagram showing a readable storage medium according to an embodiment of the present disclosure.
  • online platforms such as online art galleries
  • online art galleries have received widespread public attention as platforms for viewing art paintings and sharing artistic creations. Users can inquire, appreciate and trade various art paintings through the online platform.
  • the online platform can obtain the user's intention according to each operation of the user, and then display the corresponding painting.
  • the above solution can be applied to scenarios where each operation is not related. If several adjacent operations of the user are related, the user's intention cannot be correctly obtained, which causes the online platform to fail to display the paintings desired by the user and reduces the user experience.
  • embodiments of the present disclosure provide a painting question-and-answer method, painting question-and-answer device, and painting question-and-answer system.
  • the target dialogue state is determined in combination with the current dialogue state, so that the painting desired by the user can be obtained.
  • the technical solution provided by at least one embodiment of the present disclosure can also obtain the current intention of the user’s current operation based on the previous operation by recording the user’s conversation state, and switch from the current conversation state to the next target conversation state in combination with the current intention. That is, the technical solutions provided by the embodiments of the present disclosure can accurately obtain the intention of adjacent operations, and determine the painting desired by the user, thereby improving the user's experience of using the display terminal.
  • FIG. 1 is a block diagram of a painting question and answer system according to an embodiment of the present disclosure.
  • a painting question answering system 10 includes a display terminal 100 and a question answering server 200.
  • the display terminal 100 may be a screen-painting terminal, any applicable electronic device such as a digital photo frame, a mobile phone, a tablet computer, etc., or a specific software application program, etc.
  • the software application program is for example installed in any electronic device and can be The electronic equipment runs), they can be connected to the Internet in a wireless or wired manner, enabling users to inquire, appreciate and trade digital works (such as digital paintings), which can be displayed on the display device in the electronic equipment.
  • the display terminal 100 and the question and answer server 200 may communicate through a communication network (for example, any applicable network such as a wired local area network, a wireless local area network, a 3G/4G/5G communication network, etc.) and based on a corresponding communication protocol to transmit data.
  • the display terminal 100 is configured to send request data to the question and answer server 200.
  • the request data includes text information.
  • the text information is, for example, unstructured.
  • the text information may be text content directly input by the user, or text content obtained after recognizing the voice information input by the user.
  • the question and answer server 200 is configured to obtain text information from the display terminal 100 and the current dialogue state of the display terminal 100, and perform semantic understanding of the text information to obtain the current intention of the user, and then determine the target dialogue state according to the current dialogue state and the current intention, Obtain the answer data corresponding to the text information according to the target dialogue state.
  • the answer data includes painting data, and may also include extraction information used to extract the painting data.
  • the extraction information may be response information (including, for example, question and answer results for user questions, etc.), protocol information for communication, and encryption Information and index information corresponding to the painting data.
  • the painting data may be, for example, an image file of the painting.
  • the display terminal 100 is also configured to display painting data included in the answer data returned by the question and answer server 200.
  • the state transition table may include multiple dialogue states and multiple intents, and different dialogue states can be switched to the next dialogue state according to the corresponding intent.
  • dialogue state 1 can be switched to dialogue state 4 when the current intent is intent 1
  • dialogue state 1 can be switched to when the current intent is intent 2.
  • Switch to dialogue state 2 dialogue state 1 can be switched to dialogue state 3 when the current intention is intention 5.
  • dialogue state 2 can be switched to dialogue state 4 when the current intention is intent 3
  • dialogue state 3 can be switched to dialogue state 2 when the current intention is intent 6
  • dialogue state 3 can be switched to dialogue state when the current intention is intent 4.
  • the target dialogue state can be switched to the target dialogue state "tomorrow's how is the weather?”. It can be seen that only one input of text information by the user (ie, "tomorrow”) cannot determine the specific meaning of the user, and the text information input by the user twice (ie, "how is the weather today?" and “ What about tomorrow?”) Form associations and understand the user’s current intentions correctly.
  • the state transition table shown in FIG. 2 is only exemplary, and the embodiment of the present disclosure does not limit the number and content of the dialog states and intentions in the state transition table, and the specific switching manner between them There are no restrictions, and adjustments can be made according to actual needs. It is understandable that the state transition table can be stored in the question and answer server 200 in advance.
  • the state conversion table may be set by a technician based on actual experience, or it may be obtained by statistics or learning based on big data, which is not limited in the embodiment of the present disclosure.
  • the question and answer server 200 obtains answer data based on a state transition table (for example, the state transition table shown in FIG. 2), which may include the following operations: the question and answer server 200 obtains text information from the display terminal 100 (Corresponding to step 301), the question and answer server 200 performs semantic understanding of the text information, thereby obtaining the current intention of the user (corresponding to step 302).
  • a state transition table for example, the state transition table shown in FIG. 2
  • the question and answer server 200 obtains text information from the display terminal 100 (Corresponding to step 301), the question and answer server 200 performs semantic understanding of the text information, thereby obtaining the current intention of the user (corresponding to step 302).
  • the question and answer server 200 may pre-store the named entity recognition model and the deep learning model.
  • the question and answer server 200 can identify the named entity of the text information (corresponding to step 401).
  • the question and answer server 200 may determine the vector to be recognized corresponding to the named entity according to the named entity (corresponding to step 402).
  • the question and answer server 200 may determine the intention of the standard feature vector that meets the requirements as the current intention of the text information based on the vector to be recognized (corresponding to step 403).
  • the question and answer server 200 may recognize the named entity of the text information through a named entity recognition model.
  • Named entity recognition can refer to the recognition of entities with specific meanings in the text, such as proper nouns such as person names, organization names, and place names, and meaningful time. It is the basic task of information retrieval, question and answer systems and other technologies. For example, in “Xiao Ming is on vacation in Hawaii.”, the named entities are: “Xiao Ming-name of person", "Hawaii-name of place”. You can use language grammar-based techniques and statistical models (such as machine learning) to establish a named entity recognition system.
  • the ways to use entity detection and recognition include: (1) first perform entity detection, and then identify the detected entity, (2) combine the entity and the recognized object into a model, and obtain the position of the character for marking and Category tag.
  • question and answer server 200 may also use other models or other methods to identify the named entity of the text information, which is not limited in the embodiment of the present disclosure.
  • the question answering server 200 may determine the vector to be recognized corresponding to the named entity according to the named entity through a deep learning model. It should be noted that the question and answer server 200 may also determine the vector to be recognized corresponding to the named entity according to the named entity through other models or other methods, which is not limited in the embodiment of the present disclosure.
  • the question and answer server 200 may determine the intent of the standard feature vector with the greatest similarity to the vector to be recognized as the current intent of the text information based on the vector to be recognized. It should be noted that, in addition to the standard feature vector that has the greatest similarity to the vector to be recognized, the standard feature vector that meets the requirements can also be other standard feature vectors, depending on the actual situation. Therefore, the embodiment of the present disclosure is No specific restrictions.
  • the question and answer server 200 receives the text message "I want to see Mona Lisa", it inputs it into the named entity recognition model, and recognizes the named entity "I I want to see PICTURE".
  • the named entity recognition model performs the following operations on the received text information:
  • the named entity recognition model takes a string of characters (for example, corresponding to a sentence or paragraph in the text information) as input, and recognizes related nouns mentioned in the string (People, places and organizations).
  • the text information input into the named entity recognition model is: [I, think, look, Mo, Nai,, zhang, sun, umbrella,, female, person, O,..., O ];
  • the named entity identified by the named entity recognition model is: [O, O, O, B-PER, I-PER, O, B-PIC, I-PIC, I-PIC, I- PIC, I-PIC, O,..., O], the named entities are: character-Monet, painting- woman with parasol.
  • the named entity recognized by the named entity recognition model can determine its corresponding to-be-recognized vector after passing through the deep learning model.
  • the vector to be recognized may be a feature vector, which includes text features that are classified and extracted from named entities by a deep learning model.
  • the question and answer server 200 also includes a corpus, which includes a large number of corpora, such as "change PIC (painting)", “author's nationality”, “PERSON's painting”, etc., the question and answer server 200 Input the corpus into the deep learning model to obtain multiple standard feature vectors.
  • the standard feature vector may be a feature vector that includes text features classified and extracted from the corpus through a deep learning model.
  • the action of obtaining the standard feature vector can be completed in advance, or can be completed in real time, and can be set according to a specific scenario, which is not limited in the embodiment of the present disclosure.
  • sentences with different representations but with the same purpose can be classified into the same intent through the deep learning model. For example: “I want to see the Mona Lisa”, “Help me change the Mona Lisa”, “Show me to switch to the Mona Lisa”, etc., can be classified as the same intention "The user wants to change the PIC (Mona Lisa) Take a look”.
  • the question and answer server 200 may obtain the cosine similarity between the vector to be recognized and multiple standard feature vectors, and use the intent of the standard feature vector with the greatest similarity as the current intent of the vector to be recognized, that is, the text message "I "I want to see the Mona Lisa”'s current intention "to change the PIC”.
  • the question and answer server 200 can also obtain the current dialogue state of the display terminal 100 (corresponding to step 301), so that the question and answer server 200 can determine the target dialogue state according to the current dialogue state and current intention (corresponding step 303)
  • the question and answer server 200 can obtain the answer data corresponding to the text information according to the target dialogue state (corresponding to step 304).
  • the question and answer server 200 can determine the answer data according to the target dialogue state, combined with the current intention and text information.
  • the current intention is "Look at the PIC”
  • the slot information corresponding to the PIC has been determined to be "Mona Lisa” in the text information, so it can be determined that the answer data is the painting data "Mona Lisa” and the text template "I have recommended Leonardo's Mona Lisa for you.”
  • the question and answer server 200 sends the determined answer data to the display terminal 100, and the display terminal 100 can receive and display the painting data based on the answer data, and can also display corresponding text templates to prompt the user as needed.
  • the text information from the display terminal 100 may be text content directly input by the user, or text content obtained after recognizing the voice information input by the user.
  • a painting question and answer system 10 may include a voice recognition server 300 in addition to a display terminal 100 and a question and answer server 200.
  • the voice recognition server 300 is signally connected to the display terminal 100 and is configured to recognize voice information from the display terminal 100 and return the recognized voice result (text information) to the display terminal 100.
  • the question and answer server 200 and the speech recognition server 300 may be independent servers in a server cluster, or may be independent service processes running in a server, that is, the question and answer server 200 may refer to a server or Refers to the service process, the voice recognition server 300 is similar to this, and the embodiment of the present disclosure does not limit this.
  • the question and answer server 200 and the speech recognition server 300 may also be virtual servers and run in any physical device, private cloud or public cloud.
  • the specific implementation of the question and answer server 200 and the voice recognition server 300 is not limited in the embodiment of the present disclosure.
  • the display terminal 100, the question and answer server 200, and the voice recognition server 300 may all be deployed in the same local area network or wide area network.
  • the voice recognition process does not need to be completed by the display terminal 100, which can reduce the data processing volume and data storage volume of the display terminal 100, which is beneficial to improve the data processing efficiency.
  • the question and answer server 200 may further include a WEB server 201 and a semantic server 202.
  • the WEB server 201 may be a conversion server, which is signally connected to the display terminal 100 and the semantic server 202 respectively.
  • the WEB server 201 is configured to receive request data from the display terminal 100, parse the request data to obtain text information, send the text information to the semantic server 202, and send the answer data determined by the semantic server 202 according to the text information via the WEB server 201 returns to the display terminal 100.
  • the request data from the display terminal 100 may include the device ID of the display terminal 100, text information recognized by the voice recognition server 300 or text information input by the user, currently displayed painting data, etc. Therefore, the WEB server 201 is required to request data Parse to obtain the text information contained in the request data.
  • the specific process for the semantic server 202 to determine the answer data according to the text information may refer to the content of the embodiments shown in FIG. 4 and FIG. 5, which will not be repeated here.
  • the semantic server 202 can send the answer data to the WEB server, and the WEB server 201 returns to the display terminal 100.
  • the WEB server 201 can convert the data format between the display terminal 100 and the semantic server 202, thereby ensuring data transmission efficiency.
  • the semantic server 202 can accurately identify the current intention of the text information, and then combine the current intention and the answer data determined by the current corresponding state, and return the answer data to the display terminal 100 via the WEB server 201, thereby improving the processing efficiency and recognition efficiency of answer data .
  • a painting question and answer system 10 may include a database server 400 that stores painting data in addition to a display terminal 100 and a question and answer server 200.
  • the database server 400 is in signal connection with the question and answer server 200, and is configured to respond to the request information of the question and answer server 200 and return painting data corresponding to the request information to the question and answer server 200.
  • the query efficiency of painting data can be improved by setting a database server. It should be noted that when the painting question and answer system 10 does not include the database server 400, the painting data can be directly stored on the question and answer server 200 or in other storage devices provided separately.
  • the database server 400 may also be signally connected to the displayed terminal 100 and configured to send corresponding painting data to the display terminal 100 in response to an acquisition request sent by the display terminal 100.
  • the answer data sent by the question and answer server 200 to the display terminal 100 does not include painting data, and the answer data may include extraction information, such as obtaining permission.
  • the display terminal 100 is configured to extract painting data from the database server 400 using extraction information from the question and answer server 200.
  • the database server 400 is configured to respond to the extraction information sent by the display terminal 100 and send painting data corresponding to the extraction information to the display terminal 100.
  • the display terminal 100 sends an acquisition request including the acquisition license to the database server 400 (that is, sends the extracted information to the database server 400), and the database server 400 directly sends the painting data to the display terminal 100 after verifying the acquisition license.
  • the efficiency of obtaining painting data can be improved, the load of the question and answer server 200 can be reduced, and the load of the system composed of multiple servers such as the question answer server 200 and the database server 400 can be balanced.
  • the current intention of the current operation can be obtained on the basis of the previous operation, and combined with the current intention to switch from the current dialog state to the next target dialog state, that is, it can be accurate in this embodiment.
  • the intent of the adjacent operation is obtained, and the painting desired by the user is determined, without having to go back to the initial state after viewing and reselect the painting, thereby improving the user's experience of using the display terminal.
  • the embodiment of the present disclosure also provides a painting question answering method, which can be seen in FIG. 3, including:
  • Step 301 acquiring text information from the display terminal, and acquiring the current dialogue state of the display terminal;
  • Step 302 Perform semantic understanding on the text information to obtain the current intention of the user
  • Step 303 Determine the target dialogue state according to the current dialogue state and the current intention
  • Step 304 Obtain response data corresponding to the text information according to the target dialogue state.
  • the current intention of the current operation can be obtained on the basis of the previous operation, and combined with the current intention to switch from the current dialogue state to the next target dialogue state, that is, the relative information can be accurately obtained in this embodiment.
  • the intent of the neighbor operation determines the work desired by the user, thereby enhancing the user’s experience of using the display terminal.
  • step 302 includes:
  • Step 401 Identify the named entity of the text information
  • Step 402 Determine a vector to be recognized corresponding to the named entity according to the named entity;
  • Step 403 Based on the vector to be recognized, the intent of the standard feature vector that meets the requirements is determined as the current intent of the text information.
  • the named entity of the text information can be recognized by the named entity recognition model, the vector to be recognized corresponding to the text information can be determined by the deep learning model, and then the similarity between the vector to be recognized and the standard feature vector can be used, The current intention of the text information can be obtained, and the recognition efficiency can be improved.
  • the intention of the standard feature vector having the greatest similarity with the vector to be recognized may be determined as the current intention of the text information.
  • the painting question answering method provided by the embodiment of the present disclosure can be applied to a question and answer server. Since the painting question answering system has described the process of determining answer data by the question answering server in detail, the content of the painting question answering method shown in FIGS. 3 and 4 can be seen in the figure. The content shown in Figure 3 and Figure 4 will not be repeated here.
  • the painting question and answer device 90 may include a text information acquisition circuit 91, a current intention acquisition circuit 92, a target state determination circuit 93, and an answer data acquisition circuit 94.
  • the text information obtaining circuit 91 may be configured to obtain text information from the display terminal 100 and obtain the current conversation state of the display terminal 100.
  • the current intention obtaining circuit 92 may be configured to perform semantic understanding of the text information to obtain the current intention of the user.
  • the target state determination circuit 93 may be configured to determine the target dialogue state according to the current dialogue state and the current intention.
  • the answer data acquisition circuit 94 may be configured to acquire answer data corresponding to the text information according to the target dialogue state.
  • the current intention acquisition circuit 92 may include a named entity recognition circuit 921, a recognition vector determination circuit 922, and a current intention determination circuit 923.
  • the named entity recognition circuit 921 may be configured to recognize a named entity of text information
  • the recognition vector determination circuit 922 may be configured to determine the vector to be recognized corresponding to the named entity according to the named entity
  • the current intention determination circuit 923 may be configured to For the vector to be recognized, the intent of the standard feature vector that meets the requirements is determined as the current intent of the text information.
  • the circuit 923 may be implemented by hardware such as a processor, a controller, etc., software capable of implementing related functions, or a combination of the two, and the embodiments of the present disclosure do not limit their specific implementation manners.
  • the painting question and answer device 90 may also include more circuits, and is not limited to the text information acquisition circuit 91, the current intention acquisition circuit 92, the target state determination circuit 93, and the answer data acquisition circuit. 94. This may be determined according to actual requirements, and the embodiments of the present disclosure do not limit this.
  • the painting question answering device 90 provided by the embodiment of the present disclosure can implement the aforementioned painting question answering method, and can also achieve similar technical effects as the aforementioned painting question answering system 10, which will not be repeated here.
  • FIG. 11 is a schematic diagram of another painting question and answer device provided by an embodiment of the disclosure.
  • the painting question answering device 901 includes a processor 310 and a memory 320.
  • the memory 320 is configured to store non-transitory computer readable instructions (for example, one or more computer program modules).
  • the processor 310 is configured to execute non-transitory computer-readable instructions, and when the non-transitory computer-readable instructions are executed by the processor 310, one or more steps in the painting question answering method described above can be executed.
  • the memory 320 and the processor 310 may be interconnected by a bus system and/or other forms of connection mechanisms (not shown).
  • the memory 320 and the processor 310 may be provided on the server side (or the cloud), such as the aforementioned question and answer server 200, to execute one or more steps in the painting question answering method described in FIGS. 3-4.
  • the processor 310 may be a central processing unit (CPU), a digital signal processor (DSP), or other forms of processing units with data processing capabilities and/or program execution capabilities, such as field programmable gate arrays (FPGA), etc.;
  • the central processing unit (CPU) may be an X86 or ARM architecture.
  • the processor 310 may be a general-purpose processor or a special-purpose processor, and may control other components in the painting question and answer device 901 to perform desired functions.
  • the memory 320 may include any combination of one or more computer program products, and the computer program product may include various forms of computer-readable storage media, such as volatile memory and/or nonvolatile memory.
  • Volatile memory may include random access memory (RAM) and/or cache memory (cache), for example.
  • the non-volatile memory may include, for example, read only memory (ROM), hard disk, erasable programmable read only memory (EPROM), portable compact disk read only memory (CD-ROM), USB memory, flash memory, etc.
  • One or more computer program modules may be stored on the computer-readable storage medium, and the processor 310 may run one or more computer program modules to implement various functions of the painting question and answer device 901.
  • the computer-readable storage medium may also store various application programs and various data, various data used and/or generated by the application programs, and the like.
  • various application programs and various data various data used and/or generated by the application programs, and the like.
  • FIG. 12 is a schematic block diagram of a display terminal 100 according to an embodiment of the present disclosure.
  • the display terminal 100 can be applied to, for example, the painting question answering method provided by the embodiment of the present disclosure. It should be noted that the display terminal 100 shown in FIG. 12 is only an example, which does not bring any limitation to the function and scope of use of the embodiments of the present disclosure.
  • the display terminal 100 may include a processing device (such as a central processing unit, a graphics processor, etc.) 610, which may be loaded into a random access device according to a program stored in a read-only memory (ROM) 620 or from a storage device 680.
  • the programs in the memory (RAM) 630 execute various appropriate actions and processes.
  • various programs and data required for the operation of the display terminal 100 are also stored.
  • the processing device 610, the ROM 620, and the RAM 630 are connected to each other through a bus 640.
  • An input/output (I/O) interface 650 is also connected to the bus 640.
  • the following devices can be connected to the I/O interface 650: including input devices 660 such as touch screen, touch pad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.; including, for example, liquid crystal display (LCD), speakers, vibration An output device 670 such as a device; a storage device 680 such as a magnetic tape, a hard disk, etc.; and a communication device 690.
  • the communication device 690 may allow the display terminal 100 to perform wireless or wired communication with other electronic devices to exchange data.
  • FIG. 12 shows the display terminal 100 having various devices, it should be understood that it is not required to implement or have all the illustrated devices, and the display terminal 100 may alternatively implement or have more or fewer devices.
  • the embodiment of the present disclosure also provides a readable storage medium on which computer instructions are stored, and when the instructions are executed by a processor, the steps of the embodiments shown in FIGS. 3 to 4 are implemented.
  • FIG. 13 is a schematic diagram of a readable storage medium provided by an embodiment of the present disclosure. As shown in FIG. 13, the storage medium 500 is used to store non-transitory computer readable instructions 510. For example, when the non-transitory computer-readable instructions 510 are executed by a computer, one or more steps in the painting question answering method described above can be executed.
  • the storage medium may be any combination of one or more computer-readable storage media.
  • a computer-readable storage medium contains a computer-readable program for obtaining text information from the display terminal and the current dialog state of the display terminal Code
  • another computer-readable storage medium contains computer-readable program code that performs semantic understanding of text information to obtain the user’s current intention
  • another computer-readable storage medium contains a method for determining the target dialogue state according to the current dialogue state and the current intention
  • a computer-readable storage medium containing computer-readable program code for obtaining answer data corresponding to the text information according to the target dialogue state can also be stored in the same computer-readable medium, which is not limited in the embodiments of the present disclosure.
  • the computer can execute the program code stored in the computer storage medium, and execute, for example, the painting question and answer method provided by any embodiment of the present disclosure.
  • the storage medium may include a memory card of a smart phone, a storage component of a tablet computer, a hard disk of a personal computer, random access memory (RAM), read only memory (ROM), erasable programmable read only memory (EPROM), The portable compact disk read-only memory (CD-ROM), flash memory, or any combination of the above storage media can also be other suitable storage media.
  • the readable storage medium may also be the memory 320 in FIG. 11, and the relevant description may refer to the foregoing content, which will not be repeated here.
  • the readable storage medium can be applied to the question and answer server, and the technician can make a selection according to specific scenarios, which is not limited here.
  • the term “plurality” refers to two or more, unless specifically defined otherwise.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computational Linguistics (AREA)
  • Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • General Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Human Computer Interaction (AREA)
  • Data Mining & Analysis (AREA)
  • Mathematical Physics (AREA)
  • Databases & Information Systems (AREA)
  • Acoustics & Sound (AREA)
  • Multimedia (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • User Interface Of Digital Computer (AREA)
  • Information Transfer Between Computers (AREA)

Abstract

La présente invention concerne un procédé et un dispositif de réponse à une question de peinture, un système de réponse à une question de peinture, et un support d'informations lisible. Le procédé de réponse à une question de peinture comprend les étapes consistant à : obtenir des informations de texte à partir d'un terminal d'affichage, et obtenir un état de conversation actuel du terminal d'affichage ; effectuer une analyse sémantique sur les informations de texte de façon à obtenir une intention actuelle d'un utilisateur ; déterminer un état de conversation cible en fonction de l'état de conversation actuel et de l'intention actuelle ; et obtenir, en fonction de l'état de conversation cible, des données de réponse correspondant aux informations de texte.
PCT/CN2020/076780 2019-03-05 2020-02-26 Procédé et dispositif de réponse à une question de peinture, système de réponse à une question de peinture, et support d'informations lisible Ceased WO2020177592A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910163293.8A CN111666006B (zh) 2019-03-05 2019-03-05 画作问答方法及装置、画作问答系统、可读存储介质
CN201910163293.8 2019-03-05

Publications (1)

Publication Number Publication Date
WO2020177592A1 true WO2020177592A1 (fr) 2020-09-10

Family

ID=72338172

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/076780 Ceased WO2020177592A1 (fr) 2019-03-05 2020-02-26 Procédé et dispositif de réponse à une question de peinture, système de réponse à une question de peinture, et support d'informations lisible

Country Status (2)

Country Link
CN (1) CN111666006B (fr)
WO (1) WO2020177592A1 (fr)

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112183075A (zh) * 2020-10-10 2021-01-05 网易（杭州）网络有限公司 一种对话中的槽提取方法、装置、电子设备及存储介质
CN112231556A (zh) * 2020-10-13 2021-01-15 中国平安人寿保险股份有限公司 基于对话场景的用户画像方法、装置、设备及介质
CN112288584A (zh) * 2020-10-29 2021-01-29 泰康保险集团股份有限公司 保险报案处理方法、装置、计算机可读介质及电子设备
CN112365892A (zh) * 2020-11-10 2021-02-12 杭州大搜车汽车服务有限公司 人机对话方法、装置、电子装置及存储介质
CN113158690A (zh) * 2021-03-15 2021-07-23 京东数科海益信息科技有限公司 对话机器人的测试方法和装置
CN114049973A (zh) * 2021-11-15 2022-02-15 阿里巴巴（中国）有限公司 对话质检方法、电子设备、计算机存储介质及程序产品
CN114189740A (zh) * 2021-10-27 2022-03-15 杭州摸象大数据科技有限公司 视频合成对话构建方法、装置、计算机设备及存储介质
CN114663114A (zh) * 2022-03-22 2022-06-24 平安科技（深圳）有限公司 会话管理方法、计算机及存储介质

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106164909A (zh) * 2014-04-01 2016-11-23 微软技术许可有限责任公司 自然语言输入的任务完成
CN106663129A (zh) * 2016-06-29 2017-05-10 深圳狗尾草智能科技有限公司 一种基于状态机上下文敏感多轮对话管理系统及方法
CN107193978A (zh) * 2017-05-26 2017-09-22 武汉泰迪智慧科技有限公司 一种基于深度学习的多轮自动聊天对话方法及系统
CN107273477A (zh) * 2017-06-09 2017-10-20 北京光年无限科技有限公司 一种用于机器人的人机交互方法及装置
US20180060303A1 (en) * 2016-08-24 2018-03-01 Microsoft Technology Licensing, Llc Multi-turn cross-domain natural language understanding systems, building platforms, and methods
CN110334201A (zh) * 2019-07-18 2019-10-15 中国工商银行股份有限公司 一种意图识别方法、装置及系统
CN110532361A (zh) * 2019-08-09 2019-12-03 深圳追一科技有限公司 用户意图的识别方法、装置、计算机设备和存储介质

Family Cites Families (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101075301A (zh) * 2007-06-28 2007-11-21 腾讯科技（深圳）有限公司 虚拟宠物系统和虚拟宠物聊天方法、装置
AU2015236010A1 (en) * 2014-03-26 2016-11-10 Unanimous A.I. LLC Methods and systems for real-time closed-loop collaborative intelligence
US9965705B2 (en) * 2015-11-03 2018-05-08 Baidu Usa Llc Systems and methods for attention-based configurable convolutional neural networks (ABC-CNN) for visual question answering
US20180131642A1 (en) * 2016-11-04 2018-05-10 Microsoft Technology Licensing, Llc Conversation runtime
US11157490B2 (en) * 2017-02-16 2021-10-26 Microsoft Technology Licensing, Llc Conversational virtual assistant
CN107301213A (zh) * 2017-06-09 2017-10-27 腾讯科技（深圳）有限公司 智能问答方法及装置
CN107357875B (zh) * 2017-07-04 2021-09-10 北京奇艺世纪科技有限公司 一种语音搜索方法、装置及电子设备
CN107943860B (zh) * 2017-11-08 2020-10-27 北京奇艺世纪科技有限公司 模型的训练方法、文本意图的识别方法及装置
CN108345692B (zh) * 2018-03-16 2020-07-31 北京京东尚科信息技术有限公司 一种自动问答方法和系统
CN108804532B (zh) * 2018-05-03 2020-06-26 腾讯科技（深圳）有限公司 一种查询意图的挖掘和查询意图的识别方法、装置
US10305826B1 (en) * 2018-05-03 2019-05-28 Progressive Casualty Insurance Company Intelligent conversational systems
CN108733779B (zh) * 2018-05-04 2022-10-04 百度在线网络技术（北京）有限公司 文本配图的方法和装置
CN108897867B (zh) * 2018-06-29 2021-02-09 北京百度网讯科技有限公司 用于知识问答的数据处理方法、装置、服务器和介质

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106164909A (zh) * 2014-04-01 2016-11-23 微软技术许可有限责任公司 自然语言输入的任务完成
CN106663129A (zh) * 2016-06-29 2017-05-10 深圳狗尾草智能科技有限公司 一种基于状态机上下文敏感多轮对话管理系统及方法
US20180060303A1 (en) * 2016-08-24 2018-03-01 Microsoft Technology Licensing, Llc Multi-turn cross-domain natural language understanding systems, building platforms, and methods
CN107193978A (zh) * 2017-05-26 2017-09-22 武汉泰迪智慧科技有限公司 一种基于深度学习的多轮自动聊天对话方法及系统
CN107273477A (zh) * 2017-06-09 2017-10-20 北京光年无限科技有限公司 一种用于机器人的人机交互方法及装置
CN110334201A (zh) * 2019-07-18 2019-10-15 中国工商银行股份有限公司 一种意图识别方法、装置及系统
CN110532361A (zh) * 2019-08-09 2019-12-03 深圳追一科技有限公司 用户意图的识别方法、装置、计算机设备和存储介质

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112183075A (zh) * 2020-10-10 2021-01-05 网易（杭州）网络有限公司 一种对话中的槽提取方法、装置、电子设备及存储介质
CN112183075B (zh) * 2020-10-10 2023-06-30 网易（杭州）网络有限公司 一种对话中的槽提取方法、装置、电子设备及存储介质
CN112231556A (zh) * 2020-10-13 2021-01-15 中国平安人寿保险股份有限公司 基于对话场景的用户画像方法、装置、设备及介质
CN112231556B (zh) * 2020-10-13 2023-07-25 中国平安人寿保险股份有限公司 基于对话场景的用户画像方法、装置、设备及介质
CN112288584A (zh) * 2020-10-29 2021-01-29 泰康保险集团股份有限公司 保险报案处理方法、装置、计算机可读介质及电子设备
CN112288584B (zh) * 2020-10-29 2024-05-17 泰康保险集团股份有限公司 保险报案处理方法、装置、计算机可读介质及电子设备
CN112365892A (zh) * 2020-11-10 2021-02-12 杭州大搜车汽车服务有限公司 人机对话方法、装置、电子装置及存储介质
CN113158690A (zh) * 2021-03-15 2021-07-23 京东数科海益信息科技有限公司 对话机器人的测试方法和装置
CN113158690B (zh) * 2021-03-15 2024-05-17 京东科技信息技术有限公司 对话机器人的测试方法和装置
CN114189740A (zh) * 2021-10-27 2022-03-15 杭州摸象大数据科技有限公司 视频合成对话构建方法、装置、计算机设备及存储介质
CN114049973A (zh) * 2021-11-15 2022-02-15 阿里巴巴（中国）有限公司 对话质检方法、电子设备、计算机存储介质及程序产品
CN114663114A (zh) * 2022-03-22 2022-06-24 平安科技（深圳）有限公司 会话管理方法、计算机及存储介质

Also Published As

Publication number Publication date
CN111666006A (zh) 2020-09-15
CN111666006B (zh) 2022-01-14

Similar Documents

Publication Publication Date Title
WO2020177592A1 (fr) Procédé et dispositif de réponse à une question de peinture, système de réponse à une question de peinture, et support d'informations lisible
US10460040B2 (en) Language model using reverse translations
CN105592343B (zh) 针对问题和回答的显示装置和方法
US11024300B2 (en) Electronic device and control method therefor
WO2016206210A1 (fr) Procédé et dispositif d'envoi d'informations
CN107612814A (zh) 用于生成候选回复信息的方法和装置
US20180239812A1 (en) Method and apparatus for processing question-and-answer information, storage medium and device
CN107918634A (zh) 智能问答方法、装置及计算机可读存储介质
CN107222757A (zh) 一种语音搜索方法、机顶盒、存储介质、服务器和系统
CN103019407B (zh) 输入法应用方法、自动问答处理方法及电子设备、服务器
WO2018153130A1 (fr) Appareil et procédé de traduction
CN114528588B (zh) 跨模态隐私语义表征方法、装置、设备及存储介质
CN117312641A (zh) 智能获取信息的方法、装置、设备及存储介质
WO2019109663A1 (fr) Procédé et appareil de recherche interlingue, et appareil de recherche interlingue
JP2018072508A (ja) 音声入力装置、音声入力方法
CN114637831A (zh) 基于语义分析的数据查询方法及其相关设备
CN109640142B (zh) 一种内容推荐方法及装置、设备、存储介质
CN117236340A (zh) 问答方法、装置、设备和介质
WO2025026402A1 (fr) Procédé et appareil de réponse à une question, dispositif électronique et support de stockage
CN107608799A (zh) 一种用于执行交互指令的方法、设备及存储介质
CN113570687A (zh) 一种文件处理方法及装置
WO2023061276A1 (fr) Procédé et appareil de recommandation de données, dispositif électronique et support de stockage
CN110895924B (zh) 一种文档内容朗读方法、装置、电子设备及可读存储介质
TWI682386B (zh) 整合式語音辨識系統及方法
CN113435523B (zh) 预测内容点击率的方法、装置、电子设备以及存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20765986

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20765986

Country of ref document: EP

Kind code of ref document: A1

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 06.05.2022)