WO2021103594A1 - Procédé et dispositif de détection de degré de tacitivité, serveur et support de stockage lisible - Google Patents

Procédé et dispositif de détection de degré de tacitivité, serveur et support de stockage lisible Download PDF

Info

Publication number
WO2021103594A1
WO2021103594A1 PCT/CN2020/103409 CN2020103409W WO2021103594A1 WO 2021103594 A1 WO2021103594 A1 WO 2021103594A1 CN 2020103409 W CN2020103409 W CN 2020103409W WO 2021103594 A1 WO2021103594 A1 WO 2021103594A1
Authority
WO
WIPO (PCT)
Prior art keywords
mobile terminal
words
related words
target
level
Prior art date
Application number
PCT/CN2020/103409
Other languages
English (en)
Chinese (zh)
Inventor
许剑勇
Original Assignee
深圳壹账通智能科技有限公司
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 深圳壹账通智能科技有限公司 filed Critical 深圳壹账通智能科技有限公司
Publication of WO2021103594A1 publication Critical patent/WO2021103594A1/fr

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
    • G06F16/367—Ontology

Definitions

  • This application relates to the technical field of determining related words, and in particular to a tacit degree detection method, device, server, and readable storage medium.
  • the embodiments of the present application provide a tacit degree detection method, equipment, server, and storage medium, which can more effectively determine related words, improve the accuracy and flexibility of determining related words, and effectively realize tacit degree detection.
  • an embodiment of the present application provides a tacit degree detection method, including:
  • an embodiment of the present application provides a tacit degree detection device, which includes a unit for executing the data detection method of the first aspect described above.
  • an embodiment of the present application provides a server, including a processor, an input device, an output device, and a memory.
  • the processor, input device, output device, and memory are connected to each other, wherein the memory is used for storage support
  • a computer program for a data processing device to execute the above method the computer program including program instructions, and the processor is configured to invoke the program instructions to execute the method of the above first aspect.
  • an embodiment of the present application provides a computer-readable storage medium, the computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement the above-mentioned method of the first aspect.
  • the related words related to the original keywords of the target video image are sent to multiple mobile terminals to establish the target knowledge image, and the tacit understanding between the mobile terminals is determined according to the target knowledge graph, thereby realizing an effective tacit understanding Degree detection, enhance user experience and enhance interest.
  • FIG. 1 is a schematic flowchart of a tacit degree detection method provided by an embodiment of the present application
  • FIG. 2 is a schematic flowchart of another tacit degree detection method provided by an embodiment of the present application.
  • Fig. 3 is a schematic diagram of a knowledge graph provided by an embodiment of the present application.
  • FIG. 4 is a schematic block diagram of a tacit degree detection device provided by an embodiment of the present application.
  • Fig. 5 is a schematic block diagram of a server provided by an embodiment of the present application.
  • the tacit degree detection method provided in the embodiment of the present application may be executed by a tacit degree detection device, wherein the tacit degree detection device can be applied to a server.
  • the server may be installed on smart terminals such as mobile phones, computers, tablets, and smart watches.
  • the tacit degree detection device may be installed on the server.
  • the tacit degree detection device may be spatially independent of the server.
  • the tacit degree detection device may be a component of the server, that is, the server includes a tacit degree detection device.
  • the tacit degree detection device can obtain the first detection request sent by the first mobile terminal and the second detection request sent by the second mobile terminal, and randomly select the target video image from the preset database, where,
  • the first detection request carries the terminal identification of the second mobile terminal, and the second detection request carries the terminal identification of the first mobile terminal; acquiring voice and text information, image information, and images in the target video image Text information, and determine the original keyword of the target video image according to the voice text information, image information, and image text information; determine the target associated word set related to the original keyword according to a preset associated word extraction algorithm, and compare all
  • the related words in the target related word set are sent to the first mobile terminal and the second mobile terminal for users of the first mobile terminal and the second mobile terminal to select;
  • the knowledge unit is extracted from the target related word set, and based on the The knowledge unit establishes a target knowledge graph, wherein the knowledge unit includes any one or more of entity related words, relationship related words, and attribute related words; receiving related words selected by users of the first mobile terminal and
  • FIG. 1 is a schematic flowchart of a tacit degree detection method provided by an embodiment of the present application. As shown in FIG. 1, the method may be executed by a tacit degree detection device. A specific explanation of the tacit degree detection device As mentioned earlier, I won't repeat it here. Specifically, the method of the embodiment of the present application includes the following steps.
  • S101 Obtain a first detection request sent by a first mobile terminal and a second detection request sent by a second mobile terminal, and randomly select a target video image from a preset database.
  • the tacit degree detection device may obtain the first detection request sent by the first mobile terminal and the second detection request sent by the second mobile terminal, and randomly select the target video image from the preset database.
  • the first detection request carries the terminal identification of the second mobile terminal
  • the second detection request carries the terminal identification of the first mobile terminal
  • the first detection request carries The terminal identifier of the second mobile terminal is used to instruct the first mobile terminal to request tacit detection with the second mobile terminal
  • the terminal identifier of the first mobile terminal carried in the second detection request is used to indicate the The second mobile terminal requests tacit detection with the first mobile terminal.
  • the tacit degree detection device can obtain detection requests sent by more than two mobile terminals, and the detection requests sent by each mobile terminal can carry one or more terminal identifiers.
  • the number of mobile terminals requested for detection is not specifically limited.
  • the tacit degree detection device acquires the first detection request sent by the mobile terminal 1, and the first detection request carries the terminal identification 2 of the mobile terminal 2 and the terminal identification 3 of the mobile terminal 3; and the tacit agreement
  • the second detection request sent by the mobile terminal 2 is acquired by the degree detection device, and the second detection request carries the terminal identification 1 of the mobile terminal 1 and the terminal identification 3 of the mobile terminal 3; and the tacit degree detection device acquires the mobile
  • the third detection request sent by the terminal 3 the third detection request carries the terminal identification 1 of the mobile terminal 1 and the terminal identification 2 of the mobile terminal 2.
  • the first mobile terminal, the second mobile terminal, and the third mobile terminal perform tacit degree detection with each other at the same time, so that the tacit degree detection device sends the target video image randomly selected from the preset database to the first mobile terminal and the second mobile terminal.
  • a second mobile terminal and a third mobile terminal to facilitate subsequent determination of the tacit understanding between the first mobile terminal, the second mobile terminal, and the third mobile terminal.
  • the tacit degree detection device acquires the first detection request sent by the mobile terminal 1, and the first detection request carries the terminal identification 2 of the mobile terminal 2 and the terminal identification 3 of the mobile terminal 3; and The tacit degree detection device acquires the second detection request sent by the mobile terminal 2, and the second detection request carries the terminal identification 1 of the mobile terminal 1; and the tacit degree detection device acquires the third detection sent by the mobile terminal 3.
  • Request, the third detection request carries the terminal identification 2 of the mobile terminal 2. Therefore, it can be determined that the first mobile terminal and the second mobile terminal perform tacit degree detection with each other, and the first mobile terminal and the third mobile terminal perform tacit degree detection with each other, so that the tacit degree detection device can retrieve the tacit degree from the preset database.
  • the randomly selected target video image in the mobile terminal is sent to the first mobile terminal, the second mobile terminal, and the third mobile terminal, so as to subsequently determine the tacit understanding between the first mobile terminal and the second mobile terminal, as well as the first mobile terminal and the third mobile terminal.
  • the tacit degree of the terminal is the first mobile terminal, the second mobile terminal, and the third mobile terminal.
  • S102 Acquire voice text information, image information, and image text information in the target video image, and determine the original keywords of the target video image according to the voice text information, image information, and image text information.
  • the tacit degree detection device can obtain the voice text information, image information, and image text information in the target video image, and determine the target video image according to the voice text information, image information, and image text information The original keywords.
  • the tacit degree detection device when it acquires the voice text information, image information, and image text information in the target video image, it may determine the voice corresponding to the voice information in the target video image according to voice recognition. Text information.
  • the tacit degree detection device may perform frame decomposition on the target video image to obtain multiple image frames, and determine the image information in each image frame according to image recognition.
  • the tacit degree detection device may label the fields of each image frame to obtain field image data of each image frame, and label the text information in the field image data to obtain text label information, so as to obtain text label information according to the
  • the text annotation information identifies the image text information in the field image data.
  • the tacit degree detection device may determine whether voice information exists in the target video image before determining the voice text information corresponding to the voice information in the target video image according to voice recognition.
  • the audio information is included. If it is determined that there is voice information in the target video image, the voice text information corresponding to the voice information in the target video image can be determined according to voice recognition.
  • the tacit degree detection device when it determines the voice text information corresponding to the voice information in the target video image according to voice recognition, it may obtain the voice signal of the target video image, and set it according to the first preset Time length performs windowing and framing processing on the voice signal, and splits the voice signal into multiple segments of voice frames of a second preset duration; in some embodiments, the second preset duration is less than or equal to the The first preset duration.
  • the tacit degree detection device may perform denoising processing on each segment of the speech frame of the second preset duration, and convert all the denoising processed speech frames of the second preset duration into a speech signal sequence, and
  • the voice signal sequence is input into a preset voice recognition model for recognition, thereby determining the voice text information corresponding to the voice signal sequence.
  • the tacit degree detection device when the tacit degree detection device decomposes the target video image to obtain multiple image frames, and determines the image information in each image frame according to the image recognition, it can compare all the images according to the preset period.
  • the target video image is split to obtain multiple image frames, and the image information in each image frame is determined through image recognition; in some embodiments, the image information includes, but is not limited to, the person in the target video image Information, animal information, etc.
  • the tacit degree detection device may annotate each image frame to obtain field annotation information, and determine the field in each image frame according to the field annotation information. The position of the information, and crop each image frame according to the position of the field information to obtain the field image data corresponding to the position of the field information.
  • the tacit degree detection device may obtain the text information in the field image data, and annotate the text information in the field image data to obtain the text annotation information, and then mark the text information and the text annotation information based on the OCR recognition model.
  • the field image data is processed to identify image text information in the field image data.
  • the voice text information, image information, and image text information in the target video image can be effectively determined, so as to more effectively determine the original keywords of the target video image.
  • S103 Determine a target related word set related to the original keyword according to a preset related word extraction algorithm, and send the target related word set to the first mobile terminal and the second mobile terminal.
  • the tacit degree detection device may determine the target related word set related to the original keyword according to a preset related word extraction algorithm, and send the related words in the target related word set to the first mobile terminal and the second mobile terminal.
  • a mobile terminal for users of the first mobile terminal and the second mobile terminal to select.
  • the tacit degree detection device when it determines the target related word set related to the original keyword according to a preset related word extraction algorithm, it may query the original keyword from the network according to a preset query tool. Keyword-related related data, word segmentation is performed on the related data to obtain multiple related words, and the parts of speech of the multiple related words are determined.
  • the tacit degree detection device may determine, according to the parts of speech of the multiple related words, the related words whose part of speech is a preset part of speech and the original keywords to form the target related word set.
  • the preset query tools include, but are not limited to, tools such as Baike, Zhihu, and Chinese dictionaries.
  • the preset part of speech includes but is not limited to the part of speech of nouns or adjectives
  • the tacit degree detection device can determine the part of speech as the adjective or the related word of the noun and the original keyword according to the part of speech of the multiple related words. Compose the target related word set.
  • the tacit degree detection device may obtain the frequency of occurrence of each related word in the related word set during the query process when it is determined that the related words of the preset part of speech and the original keywords form the target related word set. , And determine that the associated words with a frequency greater than a preset frequency threshold and the original keywords constitute the target associated word set.
  • the tacit degree detection device may obtain the information of each related word in the target related word set during the query process. Frequency: sort the related words in the target related word set in the order of frequency from high to low, and extract the top n related words to add to the target related word set. For example, extract the top 10 related words and add them to the target related word set.
  • the n is a positive integer.
  • the tacit degree detection device may use a query tool to perform a loop query based on the top m related words in the n related words, and add the related words queried k times in a loop to the target related word set, where, n is greater than or equal to m, and m is greater than or equal to k.
  • the tacit degree detection device can select the top 3 related words from the top 10 related words, and use the query tool to query related words based on these 3 related words, add them to the target related word set, and end after looping 3 times. . Through this kind of circular query, more and more effective related words can be determined.
  • the tacit degree detection device may determine the semantic similarity between the related words in the target related word set after determining that the related words of the preset part of speech and the original keywords constitute the target related word set
  • the multiple related words whose semantic similarity is greater than the preset similarity threshold in the target related word set are determined as repeated related words, and the multiple related words are deduplicated.
  • the semantic similarity between the same related words is 100%.
  • S104 Extract a knowledge unit from the target related word set, and establish a target knowledge graph according to the knowledge unit.
  • the tacit degree detection device may extract knowledge units from the target related word set, and establish a target knowledge graph based on the knowledge units.
  • the knowledge unit includes any one or more of entity-related words, relationship-related words, and attribute-related words; in some embodiments, the target knowledge graph consists of entity-related words, relationship-related words, and attribute-related words. Any one or more of the composition.
  • S105 Receive the first related word sent by the first mobile terminal and the second related word sent by the second mobile terminal, and determine the first related word according to the target knowledge graph, the first related word and the second related word The tacit understanding between the first mobile terminal and the second mobile terminal.
  • the tacit degree detection device may receive related words selected by users of the first mobile terminal and the second mobile terminal, and determine the first mobile terminal according to the relationship between the related words in the target knowledge graph. The tacit understanding between the terminal and the second mobile terminal.
  • the tacit agreement detection device may determine the relationship between the related words selected by the first mobile terminal and the second related words according to the relationship between the related words in the target knowledge graph and the preset related weights of the related words.
  • the tacit degree detection device may determine the tacit degree between the first mobile terminal and the second mobile terminal according to a preset correspondence between the degree of association and the tacit degree.
  • first mobile terminal and the second mobile terminal select the same related words, they are considered to have a perfect tacit degree; if the related words selected by the first mobile terminal and the second mobile terminal are not the same, then according to the target knowledge
  • the relationship between the related words in the map and the preset related weight of each related word determine the degree of association between the related word selected by the first mobile terminal and the related word selected by the second mobile terminal. Determine the tacit degree between the first mobile terminal and the second mobile terminal according to the preset correspondence between the degree of association and the tacit degree.
  • each level in the target knowledge graph may be assigned the same correlation weight, and the correlation of each level The sum of the weights is 1, and according to the association weight of each level and the number of associated words in each level, the same preset association weight is assigned to each associated word in each level, and the preset association weight of each associated word in each level is one The sum is equal to the associated weight of each level.
  • the related word selected by the first mobile terminal and the related word selected by the second mobile terminal it is determined whether the related word selected by the first mobile terminal and the related word selected by the second mobile terminal belong to the same level.
  • the difference between the preset related weight of the related word selected by the first mobile terminal and the preset related weight of the related word selected by the second mobile terminal is obtained, and the difference between 1 and the difference is determined
  • the difference is the degree of association between the associated words selected by the first mobile terminal and the associated words selected by the second mobile terminal; if the associated words selected by the first mobile terminal and the associated words selected by the second mobile terminal belong to different levels , It is determined that the difference between the correlation weights of each level in the different levels is the degree of correlation between the related words selected by the first mobile terminal and the related words selected by the second mobile terminal.
  • Figure 3 is a schematic diagram of a knowledge graph provided by an embodiment of the present application. It is assumed that the related words selected by the first mobile terminal include China 31, Shanghai 32, supermarket 33, fruit 34, and pear 341, and the second mobile terminal The selected related words include China 31, Shanghai 32, supermarket 33, fruit 34, apple 344, and the first mobile terminal and the second mobile terminal both select China 31, Shanghai 32, supermarket 33, and fruit 34, so there is no need to calculate China 31 , Shanghai 32, supermarket 33, fruit 34, only need to calculate the correlation between pear 341 and apple 344.
  • each associated word on the target knowledge graph has a preset associated weight. If the associated word is a keyword extracted from the target video image, the preset associated weight is 1, if not from For the keywords extracted from the target video image, the preset association weight can be determined according to the frequency of the associated words in the query process.
  • the keywords extracted from the target video image include China 31, Shanghai 32, supermarket 33, and fruit 34
  • the predictions of the entity related words China 31, Shanghai 32, supermarket 33, and fruit 34 can be determined.
  • the association weight is 1, if the frequency of pear 341 in the query process is 5 times, the frequency of watermelon 342 in the query process is 4 times, the frequency of Hami melon 343 in the query process is 3 times, and the frequency of apple 344 in the query process The frequency of occurrence in the process is 6 times, and the total number of queries is 7 times.
  • the preset association weight of the pear 341 is 5/7
  • the preset association weight of the watermelon 342 is 4/7
  • the cantaloupe The default correlation weight of 343 is 3/7
  • the default correlation weight of Apple 344 is 6/7.
  • the related words related to the original keywords of the target video image are sent to multiple mobile terminals to establish the target knowledge image, and the tacit understanding between the mobile terminals is determined according to the target knowledge graph, thereby realizing an effective tacit understanding Degree detection, enhance user experience and enhance interest.
  • FIG. 2 is a schematic flowchart of another tacit degree detection method provided by an embodiment of the present application. As shown in FIG. 2, the method may be executed by a tacit degree detection device. The explanation is as mentioned before, so I won't repeat it here.
  • the difference between the embodiment of the present application and the description in FIG. 1 is that the embodiment of the present application schematically illustrates how to establish a target knowledge graph. Specifically, the method of the embodiment of the present application includes the following steps.
  • S201 Acquire voice text information, image information, and image text information in a target video image, and determine original keywords of the target video image according to the voice text information, image information, and image text information.
  • the tacit degree detection device can obtain the voice text information, image information, and image text information in the target video image, and determine the original target video image according to the voice text information, image information, and image text information. Key words.
  • S202 Determine a target related word set related to the original keyword according to a preset related word extraction algorithm.
  • the tacit degree detection device may determine the target related word set related to the original keyword according to a preset related word extraction algorithm.
  • S203 Extract a knowledge unit from the target related word set, where the knowledge unit includes any one or more of entity related words, relationship related words, and attribute related words.
  • the tacit degree detection device may extract a knowledge unit from the target related word set, where the knowledge unit includes any one or more of entity related words, relationship related words, and attribute related words.
  • the attribute-related words include attributes and/or attribute values.
  • the entity related word refers to a certain specific thing that is distinguishable and exists independently, such as a certain person, a certain city, a certain kind of plant, etc., a certain kind of commodity, and so on.
  • a certain specific thing such as a certain person, a certain city, a certain kind of plant, etc., a certain kind of commodity, and so on.
  • Such as “China”, “Italy”, “Canada”, etc., entities are the most basic elements in the knowledge graph.
  • the attribute refers to the attribute that points to it from an entity related word
  • different attribute types correspond to the edges of different types of attributes, for example, "area”, “volume”, “diameter”, “longitude”, “ “Dimensions”, “Population” and “Capital” are several different attributes.
  • the attribute value mainly refers to the value of the attribute. For example, the value of the attribute of area is 3 million square kilometers, the value of the attribute of population is 5 million, and the value of the attribute is 2m in diameter.
  • the knowledge unit may also include semantic information, which refers to a collection of entity-related words with the same characteristics, such as words that summarize a category of things, such as countries, nations, animals, and books. , Mainly refers to collections, categories, object types, types of things, such as people, geography, etc.
  • the tacit degree detection device may determine, according to the entity related words and the relationship related words, the entity related words having the same semantic category are the parallel entity related words under the same semantic category.
  • S205 Determine, according to the entity related words, attribute related words, and relationship related words, that the attribute information having the same entity related words is the parallel attribute related words under the same entity related words.
  • the tacit degree detection device may determine, according to the entity-related words, attribute-related words, and relationship-related words, that the attribute information with the same entity-related words is the parallel attribute-related words under the same entity-related words.
  • S206 Establish a target knowledge graph based on the relationship among the entity related words, parallel entity related words, attribute related words, parallel attribute related words, and relationship related words.
  • the tacit degree detection device may establish a target knowledge graph based on the relationship among the entity related words, parallel entity related words, attribute related words, parallel attribute related words, and relationship related words.
  • the tacit degree detection device may determine that the entity related words and parallel entity related words are nodes of the first level, determine that the attribute related words and the parallel attribute related words are nodes of the second level, and determine according to the relationship related words The corresponding relationship between the nodes in the first level and the nodes in the second level, and according to the corresponding relationship between the nodes in the first level and the nodes in the second level, the first level The nodes in are connected with the nodes in the second level to obtain the target knowledge graph.
  • different entity related words have different relationships.
  • the relationship is a function that maps k nodes (entity related words, semantic classes, attributes, attribute values) to Boolean values.
  • the relationship information in the target knowledge graph includes semantic category-relationship-entity, such as country-China, ethnicity-Han, etc., entity-relationship-entity, such as China-capital-Beijing, China-municipalities-Chongqing, etc., entity-attribute -Attribute values, such as Chongqing-population-30 million, Shanghai-area-500,000 square kilometers, etc.
  • the semantic category that can be determined includes China 31
  • Entity-related words belonging to the same semantic category China 31 include Shanghai 32
  • entity-related word supermarket 33 is the lower attribute of entity-related word Shanghai 32
  • entity-related word fruit 34 is the lower attribute of entity-related word supermarket 33, and belongs to the same entity-related word fruit 34.
  • Related words include pear 341, watermelon 342, cantaloupe 343, and apple 344.
  • the relationship between the entity-related word supermarket 33 and the entity-related word fruit 34 is sell 35.
  • the lower attributes of the apple 344 include apple cider vinegar 3441 and dried apple 3442, so it can Establish the target knowledge graph as shown in Figure 3.
  • the embodiment of this application determines the entity related words with the same semantic category as the parallel entity related words, and determines the parallel attribute related words with the same entity related words according to the entity related words, the attribute related words and the relation related words included in the knowledge unit, so as to determine the parallel attribute related words according to the entity related words and the parallel words.
  • the relationship between entity-related words, attribute-related words, parallel attribute-related words, and relationship-related words establishes an effective target knowledge graph in order to improve the effectiveness of tacit degree detection.
  • FIG. 4 is a schematic block diagram of a tacit degree detection device provided by an embodiment of the present application.
  • the tacit degree detection device of this embodiment includes: an acquiring unit 401, a determining unit 402, a sending unit 403, a establishing unit 404, and a receiving unit 405;
  • the acquiring unit 401 is configured to acquire a first detection request sent by a first mobile terminal and a second detection request sent by a second mobile terminal, and randomly select a target video image from a preset database, wherein the first detection
  • the request carries the terminal identification of the second mobile terminal for requesting tacit detection with the second mobile terminal
  • the second detection request carries the terminal identification of the first mobile terminal for requesting communication with the second mobile terminal.
  • the first mobile terminal performs tacit degree detection
  • the determining unit 402 is configured to obtain voice text information, image information, and image text information in the target video image, and determine the original keywords of the target video image according to the voice text information, image information, and image text information;
  • the sending unit 403 is configured to determine a target related word set related to the original keyword according to a preset related word extraction algorithm, and send the target related word set to the first mobile terminal and the second mobile terminal for the User selection of the first mobile terminal and the second mobile terminal;
  • the establishment unit 404 is configured to extract knowledge units from the target related word set, and build a target knowledge graph based on the knowledge units, wherein the knowledge unit includes any one or more of entity related words, relationship related words, and attribute related words ;
  • the receiving unit 405 is configured to receive the first related words sent by the first mobile terminal and the second related words sent by the second mobile terminal, and according to the target knowledge graph, the first related words and the second related words , Determining the degree of tacit understanding between the first mobile terminal and the second mobile terminal.
  • the sending unit 403 determines the target related word set related to the original keyword according to a preset related word extraction algorithm to obtain the voice text information, image information, and image text information in the target video image, it is specifically used for :
  • the part of speech is a related word of a preset part of speech and the original keywords form the target related word set.
  • the sending unit 403 determines that the part-of-speech is the related word of the preset part-of-speech and the original keywords form the target related word set, it is further used to:
  • a plurality of related words whose semantic similarity in the target related word set is greater than a preset similarity threshold is determined as repeated related words, and the multiple related words are deduplicated.
  • the establishing unit 404 establishes the target knowledge graph according to the knowledge unit, it is specifically configured to:
  • attribute related words, and relationship related words determine that the attribute information with the same entity related words is the parallel attribute related words under the same entity related words
  • a target knowledge graph is established.
  • the establishing unit 404 establishes the target knowledge graph according to the relationship among the entity related words, parallel entity related words, attribute related words, parallel attribute related words, and relation related words, it is specifically used for:
  • the nodes in the first level and the nodes in the second level are connected to obtain the target knowledge graph.
  • the receiving unit 405 determines the tacit degree between the first mobile terminal and the second mobile terminal according to the relationship between the related words in the target knowledge graph, it is specifically used to:
  • the preset association weight is determined according to the frequency of occurrence of the associated word in the query process
  • the receiving unit 405 determines the related words selected by the first mobile terminal and the related words selected by the second mobile terminal according to the relationship between the related words in the target knowledge graph and the preset correlation weight of each related word Before the degree of association, it was also used for:
  • each related word in each level is assigned the same preset correlation weight, and the sum of the preset correlation weight of each related word in each level is equal to the correlation of each level Weights;
  • the receiving unit 405 determines the relationship between the related words selected by the first mobile terminal and the related words selected by the second mobile terminal according to the relationship between the related words in the target knowledge graph and the preset correlation weight of each related word.
  • relevance it is specifically used for:
  • the preset association weights of the associated words selected by the first mobile terminal and the associated words selected by the second mobile terminal are acquired And determine the difference between 1 and the difference to be the degree of relevance between the related word selected by the first mobile terminal and the related word selected by the second mobile terminal;
  • the difference between the correlation weights of each level in the different levels is determined as the related word selected by the first mobile terminal and the first 2.
  • the related words related to the original keywords of the target video image are sent to multiple mobile terminals to establish the target knowledge image, and the tacit understanding between the mobile terminals is determined according to the target knowledge graph, thereby realizing an effective tacit understanding Degree detection, enhance user experience and enhance interest.
  • FIG. 5 is a schematic block diagram of a server according to an embodiment of the present application.
  • the server in this embodiment as shown in the figure may include: one or more processors 501; one or more input devices 502, one or more output devices 503, and a memory 504.
  • the aforementioned processor 501, input device 502, output device 503, and memory 504 are connected via a bus 505.
  • the memory 504 is configured to store a computer program including program instructions
  • the processor 501 is configured to execute the program instructions stored in the memory 504.
  • the processor 501 is configured to call the program instructions to execute:
  • the processor 501 determines the target associated word set related to the original keyword according to a preset associated word extraction algorithm to obtain the voice text information, image information, and image text information in the target video image, it is specifically used for :
  • the part of speech is a related word of a preset part of speech and the original keywords form the target related word set.
  • the processor 501 determines that the part-of-speech is the related word of the preset part-of-speech and the original keywords form the target related word set, it is further used to:
  • a plurality of related words whose semantic similarity in the target related word set is greater than a preset similarity threshold is determined as repeated related words, and the multiple related words are deduplicated.
  • the processor 501 establishes a target knowledge graph according to the knowledge unit, it is specifically configured to:
  • attribute related words, and relationship related words determine that the attribute information with the same entity related words is the parallel attribute related words under the same entity related words
  • a target knowledge graph is established.
  • the processor 501 establishes the target knowledge graph according to the relationship among the entity related words, parallel entity related words, attribute related words, parallel attribute related words, and relation related words, it is specifically used for:
  • the nodes in the first level and the nodes in the second level are connected to obtain the target knowledge graph.
  • the processor 501 determines the tacit degree between the first mobile terminal and the second mobile terminal according to the relationship between the related words in the target knowledge graph, it is specifically configured to:
  • the preset association weight is determined according to the frequency of occurrence of the associated word in the query process
  • the processor 501 determines the related word selected by the first mobile terminal and the related word selected by the second mobile terminal according to the relationship between the related words in the target knowledge graph and the preset related weight of each related word Before the degree of association, it was also used for:
  • each related word in each level is assigned the same preset correlation weight, and the sum of the preset correlation weight of each related word in each level is equal to the correlation of each level Weights;
  • the processor 501 determines the relationship between the related words selected by the first mobile terminal and the related words selected by the second mobile terminal according to the relationship between the related words in the target knowledge graph and the preset correlation weight of each related word.
  • relevance it is specifically used for:
  • the preset association weights of the associated words selected by the first mobile terminal and the associated words selected by the second mobile terminal are acquired And determine the difference between 1 and the difference to be the degree of relevance between the related word selected by the first mobile terminal and the related word selected by the second mobile terminal;
  • the difference between the correlation weights of each level in the different levels is determined as the related word selected by the first mobile terminal and the first 2.
  • the related words related to the original keywords of the target video image are sent to multiple mobile terminals to establish the target knowledge image, and the tacit understanding between the mobile terminals is determined according to the target knowledge graph, thereby realizing an effective tacit understanding Degree detection, enhance user experience and enhance interest.
  • An embodiment of the present application also provides a computer-readable storage medium, the computer-readable storage medium stores a computer program, and when the program instructions are executed by a processor, the processor executes:
  • the terminal identifier of the mobile terminal is used to request tacit detection with the second mobile terminal, and the second detection request carries the terminal identifier of the first mobile terminal and is used to request tacit understanding with the first mobile terminal Degree detection;
  • the computer-readable storage medium may be the internal storage unit of the data detection device described in any of the foregoing embodiments, such as the hard disk or memory of the data detection device.
  • the computer-readable storage medium may also be an external storage device of the data detection device, such as a plug-in hard disk equipped on the data detection device, a smart memory card (SmarS Media Card, SMC), and a secure digital (Secure DigiSal) ,SD) card, flash card (Flash Card), etc.
  • the computer-readable storage medium may also include both an internal storage unit of the data detection device and an external storage device.
  • the computer-readable storage medium is used to store the computer program and other programs and data required by the data detection device.
  • the computer-readable storage medium can also be used to temporarily store data that has been output or will be output.
  • the computer-readable storage medium may be non-volatile or volatile.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Animal Behavior & Ethology (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

L'invention concerne un procédé et un dispositif de détection de degré de tacitivité, ainsi qu'un serveur et un support de stockage lisible. Le procédé consiste à : acquérir une première demande de détection envoyée par un premier terminal mobile et une seconde demande de détection envoyée par un second terminal mobile, puis sélectionner de manière aléatoire une image vidéo cible à partir d'une base de données prédéfinie (S101); déterminer un mot-clé d'origine en fonction d'informations de texte vocal, d'informations d'image et d'informations de texte d'image acquises à partir de l'image vidéo cible sélectionnée de manière aléatoire (S102); envoyer les mots associés d'un ensemble de mots associés cible relatif au mot-clé d'origine au premier terminal mobile et au second terminal mobile (S103); établir un graphe de connaissances cible selon les unités de connaissances extraites de l'ensemble de mots associés cible (S104); et recevoir un premier mot associé envoyé par le premier terminal mobile et un second mot associé envoyé par le second terminal mobile, puis déterminer le degré de tacitivité entre le premier terminal mobile et le second terminal mobile selon le graphe de connaissances cible, le premier mot associé et le second mot associé (S105). Par conséquent, la précision et la flexibilité de détermination des mots associés sont améliorées, l'expérience de l'utilisateur est améliorée, et le niveau d'intérêt est augmenté.
PCT/CN2020/103409 2019-11-25 2020-07-22 Procédé et dispositif de détection de degré de tacitivité, serveur et support de stockage lisible WO2021103594A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201911178888.7 2019-11-25
CN201911178888.7A CN111125369A (zh) 2019-11-25 2019-11-25 一种默契度检测方法、设备、服务器及可读存储介质

Publications (1)

Publication Number Publication Date
WO2021103594A1 true WO2021103594A1 (fr) 2021-06-03

Family

ID=70496734

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/103409 WO2021103594A1 (fr) 2019-11-25 2020-07-22 Procédé et dispositif de détection de degré de tacitivité, serveur et support de stockage lisible

Country Status (2)

Country Link
CN (1) CN111125369A (fr)
WO (1) WO2021103594A1 (fr)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111125369A (zh) * 2019-11-25 2020-05-08 深圳壹账通智能科技有限公司 一种默契度检测方法、设备、服务器及可读存储介质
CN112151035B (zh) * 2020-10-14 2023-08-11 珠海格力电器股份有限公司 一种语音控制方法、装置、电子设备及可读存储介质

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107766498A (zh) * 2017-10-19 2018-03-06 北京百度网讯科技有限公司 用于生成信息的方法和装置
CN109033132A (zh) * 2018-06-05 2018-12-18 中证征信（深圳）有限公司 利用知识图谱计算文本和主体相关度的方法以及装置
US20190005121A1 (en) * 2017-06-29 2019-01-03 Beijing Baidu Netcom Science And Technology Co., Ltd. Method and apparatus for pushing information
CN109684553A (zh) * 2018-12-26 2019-04-26 北京百度网讯科技有限公司 用于获取信息的方法及装置
CN109801103A (zh) * 2019-01-14 2019-05-24 海南英赛德信息系统有限公司 信息投放方法及装置、存储介质及电子设备
CN111125369A (zh) * 2019-11-25 2020-05-08 深圳壹账通智能科技有限公司 一种默契度检测方法、设备、服务器及可读存储介质

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20190005121A1 (en) * 2017-06-29 2019-01-03 Beijing Baidu Netcom Science And Technology Co., Ltd. Method and apparatus for pushing information
CN107766498A (zh) * 2017-10-19 2018-03-06 北京百度网讯科技有限公司 用于生成信息的方法和装置
CN109033132A (zh) * 2018-06-05 2018-12-18 中证征信（深圳）有限公司 利用知识图谱计算文本和主体相关度的方法以及装置
CN109684553A (zh) * 2018-12-26 2019-04-26 北京百度网讯科技有限公司 用于获取信息的方法及装置
CN109801103A (zh) * 2019-01-14 2019-05-24 海南英赛德信息系统有限公司 信息投放方法及装置、存储介质及电子设备
CN111125369A (zh) * 2019-11-25 2020-05-08 深圳壹账通智能科技有限公司 一种默契度检测方法、设备、服务器及可读存储介质

Also Published As

Publication number Publication date
CN111125369A (zh) 2020-05-08

Similar Documents

Publication Publication Date Title
US11227118B2 (en) Methods, devices, and systems for constructing intelligent knowledge base
CN109101620B (zh) 相似度计算方法、聚类方法、装置、存储介质及电子设备
US9460117B2 (en) Image searching
WO2020140373A1 (fr) Procédé de reconnaissance d'intention, dispositif de reconnaissance et support d'informations lisible par ordinateur
CN109670163B (zh) 信息识别方法、信息推荐方法、模板构建方法及计算设备
CN111522927B (zh) 基于知识图谱的实体查询方法和装置
US20230057010A1 (en) Term weight generation method, apparatus, device and medium
WO2022105119A1 (fr) Procédé de génération de corpus d'apprentissage pour un modèle de reconnaissance d'intention, et dispositif associé
WO2019148712A1 (fr) Procédé de détection de site web d'hameçonnage, dispositif, équipement informatique et support de stockage
CN111708942B (zh) 多媒体资源推送方法、装置、服务器及存储介质
WO2020206910A1 (fr) Procédé de poussée d'informations de produit, appareil, dispositif informatique, et support de stockage
WO2021103594A1 (fr) Procédé et dispositif de détection de degré de tacitivité, serveur et support de stockage lisible
WO2021073410A1 (fr) Procédé de tri et de recommandation, appareil et dispositif pour fournir une preuve légale et support de stockage
CN113660541A (zh) 新闻视频的摘要生成方法及装置
CN114598597B (zh) 多源日志解析方法、装置、计算机设备及介质
CN108875050B (zh) 面向文本的数字取证分析方法、装置和计算机可读介质
CN110347900B (zh) 一种关键词的重要度计算方法、装置、服务器及介质
CN105512270B (zh) 一种确定相关对象的方法和装置
CN110442696B (zh) 查询处理方法及装置
CN111752922A (zh) 一种建立知识数据库、实现知识查询的方法及装置
CN114547257B (zh) 类案匹配方法、装置、计算机设备及存储介质
CN109101595B (zh) 一种信息查询方法、装置、设备及计算机可读存储介质
CN113741864A (zh) 基于自然语言处理的语义化服务接口自动设计方法与系统
US10803115B2 (en) Image-based domain name system
CN116610782B (zh) 文本检索方法、装置、电子设备及介质

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: 20894157

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: 20894157

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 23.09.2022)

122 Ep: pct application non-entry in european phase

Ref document number: 20894157

Country of ref document: EP

Kind code of ref document: A1