CN113204636B - Knowledge graph-based user dynamic personalized image drawing method - Google Patents
Knowledge graph-based user dynamic personalized image drawing method Download PDFInfo
- Publication number
- CN113204636B CN113204636B CN202110025972.6A CN202110025972A CN113204636B CN 113204636 B CN113204636 B CN 113204636B CN 202110025972 A CN202110025972 A CN 202110025972A CN 113204636 B CN113204636 B CN 113204636B
- Authority
- CN
- China
- Prior art keywords
- user
- data
- knowledge graph
- entities
- entity
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 16
- 239000013598 vector Substances 0.000 claims abstract description 32
- 238000005516 engineering process Methods 0.000 claims abstract description 16
- 230000004927 fusion Effects 0.000 claims abstract description 8
- 238000012545 processing Methods 0.000 claims abstract description 4
- 238000012216 screening Methods 0.000 claims abstract description 4
- 230000006399 behavior Effects 0.000 claims description 45
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000000605 extraction Methods 0.000 claims description 4
- 238000013135 deep learning Methods 0.000 claims description 3
- 238000012804 iterative process Methods 0.000 claims description 3
- 239000003086 colorant Substances 0.000 claims description 2
- 238000003384 imaging method Methods 0.000 claims 1
- 230000000007 visual effect Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013481 data capture Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/335—Filtering based on additional data, e.g. user or group profiles
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—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; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Databases & Information Systems (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Image Analysis (AREA)
- User Interface Of Digital Computer (AREA)
Abstract
The invention discloses a user dynamic personalized image drawing method based on a knowledge graph, which comprises the following steps: user data of a user to be sketched is acquired in real time, and a named entity recognition model is input to acquire entities in the user data and corresponding entity relations; classifying the data of the entity based on the knowledge fusion technology to obtain multidimensional behavior data of the user; after field screening is carried out on the behavior data, a knowledge graph oriented to the user portrait is constructed; the entity in the knowledge graph is expressed by word vectors, and Euclidean distance between words is calculated; determining entities with similar semantics in the knowledge graph and the relevance between words and the entities according to the Euclidean distance, and constructing a user behavior tag table according to the relevance; updating a user behavior tag table according to iterative processing of real-time collected data, and constructing a dynamic personalized image of a user. By the technical scheme, the dynamic personalized portrait of the user is obtained, and the accuracy and the real-time performance of the portrait of the user are ensured.
Description
Technical Field
The invention relates to the technical field of user portraits, in particular to a user dynamic personalized portraits method based on a knowledge graph.
Background
At present, with the wide application of big data technology, big data is utilized in different fields to perform corresponding analysis on users thereof so as to obtain the characteristics and attributes of the user population thereof, thereby further expanding the users. In this regard, portrait technologies for the user population have been paid attention to and developed.
However, the existing portrait method for users adopts the same algorithm for different crowds and different users, the adopted corpus and data are data from multiple channels and multiple fields, the portrait of the users can only be simply reflected, the personalized portrait of the user crowds in different fields can not be systematically, systematically and accurately reflected, and the user portrait can not be updated rapidly and dynamically due to the rapid updating and changing speeds of the interest points, the attention points and the like of the users, so that the precision and the instantaneity of the portrait of the users are influenced.
Disclosure of Invention
Aiming at the problems, the invention provides a user dynamic personalized image method based on a knowledge graph, which can excavate and analyze entities, entity relations, attributes and the like in user data by constructing the knowledge graph on the data acquired by the user image, and can organically combine the user image technology by utilizing visual technology image display, in addition, semantic similar entities in the data of the current user are obtained through the calculation of Euclidean distance in a word vector space, so that the feature label which can most represent the current user is obtained through vector similarity, and the dynamic personalized image of the user is obtained through the real-time acquisition and update of the user data, thereby ensuring the accuracy and the real-time performance of the user image.
In order to achieve the above object, the present invention provides a method for dynamically personalizing an image of a user based on a knowledge graph, comprising: collecting user data of a user to be portrait in real time; inputting the user data into a named entity recognition model to obtain entities in the user data and corresponding entity relations; classifying the data of the entity based on a knowledge fusion technology to obtain multidimensional behavior data of the user; performing field screening on the behavior data to construct a knowledge graph oriented to the user portrait; the entity in the knowledge graph is expressed by a word vector, and Euclidean distance between words is calculated in a word vector space; determining the entity with similar semantics in the knowledge graph and the correlation between the word and the entity according to the Euclidean distance, and constructing a user behavior tag table according to the correlation; updating the user behavior tag table according to the iterative processing of the real-time collected data, and constructing the dynamic personalized image of the user.
In the above technical solution, preferably, the acquiring, in real time, user data of a user to be portrait specifically includes: and carrying out real-time dynamic grabbing on the multi-terminal, multi-system and multi-field data of the user through a data grabbing tool.
In the foregoing technical solution, preferably, the inputting the user data into a named entity recognition model to obtain the entity and the corresponding entity relationship in the user data specifically includes: inputting the user data into a named entity recognition model based on deep learning; and carrying out entity recognition and relation extraction on the user data by the named entity recognition model to obtain entity relations between the entities in the user data and the corresponding entities.
In the foregoing technical solution, preferably, the classifying the data of the entity based on the knowledge fusion technology to obtain the multidimensional behavior data of the user specifically includes: and classifying the entities obtained by the named entity recognition model by adopting a knowledge fusion technology based on a cross-modal shared subspace learning theory, and obtaining multi-dimensional entity classification as behavior data of the user.
In the above technical solution, preferably, the calculating the euclidean distance between words in the word vector space by using the word vector representation for the entity in the knowledge graph specifically includes: using a depth semantic model to represent the entities in the knowledge graph by word vectors to form a word vector space; and calculating Euclidean distances between related words in the word vector space by utilizing semantic similarity and logical correlation between entities of the knowledge graph.
In the above technical solution, preferably, the determining, according to the euclidean distance, the entity with similar semantics in the knowledge graph and the correlation between the word and the entity, and constructing the user behavior tag table according to the correlation specifically includes: obtaining entities with similar semantic meaning of word representations obtained through requirement word generalization in the knowledge graph according to the Euclidean distance; and obtaining the label with the entity similar to the label semantic by using vector similarity calculation, and constructing a user behavior label table.
In the above technical solution, preferably, the updating the user behavior tag table and constructing the dynamic personalized image of the user according to the iterative process of collecting data in real time specifically includes: updating the behavior data of the user according to the data acquired in real time; updating the knowledge graph according to the updated behavior data; updating the user behavior tag table according to the updated knowledge graph; and constructing a dynamic personalized image of the user according to the updated user behavior label table.
In the above technical solution, preferably, the feature labels in the dynamic personalized image are highlighted in a preset manner according to the occurrence frequency, where the preset manner includes increasing the display word size of the feature labels, highlighting the feature labels with different colors, and displaying the feature labels in a central area of the dynamic personalized image or between the feature labels and other labels with a density smaller than that between common labels.
Compared with the prior art, the invention has the beneficial effects that: the knowledge graph is constructed on the data acquired by the user portrait, so that the entity, entity relation, attribute and the like in the user data can be mined and analyzed, the visual technology image display is utilized, the knowledge graph and the user portrait technology can be organically combined, in addition, the semantic similar entity in the data of the current user can be obtained through the calculation of the Euclidean distance in the word vector space, the feature label which can be the most representative of the current user can be obtained through the vector similarity, and the user data can be acquired and updated in real time, so that the dynamic personalized portrait of the user can be obtained, and the accuracy and the real-time performance of the user portrait can be ensured.
Drawings
Fig. 1 is a flow chart of a knowledge-based user dynamic personalized image method according to an embodiment of the invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The invention is described in further detail below with reference to the attached drawing figures:
as shown in fig. 1, the method for dynamically personalizing an image of a user based on a knowledge graph provided by the present invention includes: collecting user data of a user to be portrait in real time; inputting the user data into a named entity recognition model to acquire entities in the user data and corresponding entity relations; classifying the data of the entity based on the knowledge fusion technology to obtain multidimensional behavior data of the user; after field screening is carried out on the behavior data, a knowledge graph oriented to the user portrait is constructed; the entity in the knowledge graph is expressed by a word vector, and Euclidean distance between words is calculated in a word vector space; determining entities with similar semantics in the knowledge graph and the relevance between words and the entities according to the Euclidean distance, and constructing a user behavior tag table according to the relevance; updating a user behavior tag table according to iterative processing of real-time collected data, and constructing a dynamic personalized image of a user.
In the embodiment, knowledge graph construction is carried out on the data acquired by the user portrait, so that entities, entity relations, attributes and the like in the user data can be mined and analyzed, visual technology image display is utilized, the knowledge graph can be organically combined with the user portrait technology, in addition, semantic similar entities in the data of the current user are obtained through calculation of Euclidean distances in a word vector space, so that feature tags which can represent the current user most are obtained through vector similarity, and then the user data are acquired and updated in real time, so that dynamic personalized portraits of the user are obtained, and the accuracy and the instantaneity of the user portrait are ensured.
Specifically, analyzing specific demand information, and performing semantic generalization to obtain word representations close to the semantics of the demand information. Then, semantic similarity and logic correlation between entities provided by the knowledge graph are utilized, euclidean distance between related words is calculated in a word vector space, word representations obtained through generalization of required words are represented, entities with similar semantics are searched in the knowledge graph, and the correlation between the words and the entities can be obtained through the representation of the Euclidean distance. After obtaining the semantically related knowledge entities, the similarity calculation of the vectors is also utilized to obtain a tag table with related entities similar to the known user behavior tags semantically. And obtaining the strength of the correlation with the user corresponding to the label according to the magnitude of the similarity value, and generating the user behavior label capable of representing the user characteristics.
Further, attribute values of relative entities are extracted through collecting a large amount of real user data in real time, including mobile phone APP behavior data, browser search word data, game live broadcast and other entertainment consumption data of the user, and descriptions of the entities are enriched. In addition, new entities or new entity attributes are found through the search logs, and coverage rate of the knowledge graph is continuously expanded. Based on the method, the dynamic updating of the personalized image of the user is realized, and the real-time performance and the accuracy of the image of the user are ensured.
In the above embodiment, preferably, the acquiring, in real time, the user data of the user of the desired portrait specifically includes: the data capture tool is used for carrying out real-time dynamic capture on multi-terminal, multi-system and multi-field data of the user, such as mobile phone APP behavior data of the user, browser search word data, entertainment consumption data such as game live broadcast and the like.
In the foregoing embodiment, preferably, inputting the user data into the named entity recognition model to obtain the entity and the corresponding entity relationship in the user data specifically includes: inputting user data into a named entity recognition model based on deep learning; and carrying out entity recognition, relation extraction and attribute extraction technology on the user data by using the named entity recognition model to obtain entity relations between the entities in the user data and the corresponding entities.
In the foregoing embodiment, preferably, classifying the data of the entity based on the knowledge fusion technology to obtain the multidimensional behavior data of the user specifically includes: and classifying the entities obtained by the named entity recognition model by adopting a knowledge fusion technology based on a cross-modal shared subspace learning theory to obtain multidimensional entity classification which is taken as behavior data of users, wherein the behavior data comprise social attributes, living habits, consumption behaviors and the like.
In the above embodiment, preferably, the entity in the knowledge graph is represented by a word vector, and calculating the euclidean distance between words in the word vector space specifically includes: using a depth semantic model to represent entities in the knowledge graph by word vectors to form a word vector space; the Euclidean distance between related words is calculated in the word vector space by utilizing the semantic similarity and the logical correlation between the entities of the knowledge graph.
In the foregoing embodiment, preferably, determining, according to the euclidean distance, the entity with similar semantics and the correlation between the terms and the entity in the knowledge graph, and constructing the user behavior label table according to the correlation specifically includes: obtaining entities with similar semantic meaning of word representations obtained through generalization of the required words in the knowledge graph according to the Euclidean distance; and obtaining the label with the entity similar to the label semantic by using vector similarity calculation, and constructing a user behavior label table.
Specifically, the method comprises the steps of representing the words into word vector representation by using a deep semantic model, calculating Euclidean distances among related words by using semantic similarity and logical correlation between entities provided by a knowledge graph, searching for entities with similar semantics in the knowledge graph by using word representation obtained by generalization of required words by calculating Euclidean distances among related words in a word vector space, and obtaining the correlation between the words and the entities by using the Euclidean distance representation. After the semantically related knowledge entity is obtained, the similarity of the vector is also utilized to calculate and obtain a label table with the related entity similar to the known user behavior label semantically. And obtaining the strength of the correlation with the user corresponding to the label through combination calculation according to the magnitude of the similarity value, and generating the user behavior label capable of representing the user characteristics.
In the foregoing embodiment, preferably, updating the user behavior tag table and constructing the dynamic personalized image of the user according to the iterative process of collecting the data in real time specifically includes: updating behavior data of a user according to data acquired in real time; updating the knowledge graph according to the updated behavior data; updating the user behavior tag table according to the updated knowledge graph; and constructing a dynamic personalized image of the user according to the updated user behavior label table.
In the above embodiment, it is preferable that the feature label in the dynamic personalized image is highlighted in a preset manner according to the appearance frequency, and the preset manner includes increasing the display word size of the feature label, highlighting the feature label in a different color, and displaying the feature label in the center area of the dynamic personalized image or between the feature label and other labels with a density smaller than that between ordinary labels.
The above is only a preferred embodiment of the present invention, and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (4)
1. The user dynamic personalized image drawing method based on the knowledge graph is characterized by comprising the following steps of:
collecting user data of a user to be portrait in real time;
inputting the user data into a named entity recognition model based on deep learning, and performing entity recognition and relation extraction on the user data by the named entity recognition model to obtain entity relations between entities in the user data and corresponding entities;
carrying out data classification on the entity obtained by the named entity recognition model by adopting a knowledge fusion technology based on a cross-modal shared subspace learning theory to obtain multi-dimensional entity classification as behavior data of the user;
performing field screening on the behavior data to construct a knowledge graph oriented to the user portrait;
the method comprises the steps of representing entities in a knowledge graph by word vectors, calculating Euclidean distance between words in a word vector space, determining entities with similar semantics in the knowledge graph and correlation between the words and the entities according to the Euclidean distance, and constructing a user behavior tag table according to the correlation, wherein the method specifically comprises the following steps:
analyzing the demand information and performing semantic generalization to obtain word representations close to the semantics of the demand information;
by utilizing semantic similarity and logical correlation between entities provided by the knowledge graph, calculating Euclidean distance between related words in a word vector space, representing all words obtained by the generalization of the required words, searching for entities with similar semantics in the knowledge graph, and obtaining correlation between all word representations obtained by the generalization of the required words and the entities through the representation of Euclidean distance;
after obtaining the knowledge entity related to the semantics, obtaining a tag table of the knowledge entity related to the semantics and the known user behavior tag semantics by using similarity calculation of vectors, obtaining the strength of the correlation with the user corresponding to the tag according to the magnitude of the similarity value, and generating a user behavior tag capable of representing the user characteristics;
updating the user behavior tag table according to the iterative processing of the real-time collected data, and constructing the dynamic personalized image of the user.
2. The knowledge-based user dynamic personalized imaging method according to claim 1, wherein the real-time acquisition of user data of the user to be imaged specifically comprises:
and carrying out real-time dynamic grabbing on the multi-terminal, multi-system and multi-field data of the user through a data grabbing tool.
3. The knowledge-based user dynamic personalized image method according to claim 1, wherein updating the user behavior tag table and constructing the user dynamic personalized image according to the iterative process of collecting data in real time comprises:
updating the behavior data of the user according to the data acquired in real time;
updating the knowledge graph according to the updated behavior data;
updating the user behavior tag table according to the updated knowledge graph;
and constructing a dynamic personalized image of the user according to the updated user behavior label table.
4. The knowledge-graph-based user dynamic personalized image method according to claim 1, wherein feature labels in the dynamic personalized image are highlighted in a preset manner according to the occurrence frequency, the preset manner comprising increasing the display word size of the feature labels, highlighting the feature labels in different colors, and displaying the feature labels in a center region of the dynamic personalized image or between the feature labels and other labels with a density smaller than that between common labels.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110025972.6A CN113204636B (en) | 2021-01-08 | 2021-01-08 | Knowledge graph-based user dynamic personalized image drawing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110025972.6A CN113204636B (en) | 2021-01-08 | 2021-01-08 | Knowledge graph-based user dynamic personalized image drawing method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113204636A CN113204636A (en) | 2021-08-03 |
| CN113204636B true CN113204636B (en) | 2023-12-05 |
Family
ID=77025168
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110025972.6A Active CN113204636B (en) | 2021-01-08 | 2021-01-08 | Knowledge graph-based user dynamic personalized image drawing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113204636B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114153716B (en) * | 2022-02-08 | 2022-05-06 | 中国电子科技集团公司第五十四研究所 | Real-time portrait generation method for people and nobody objects under semantic information exchange network |
| CN114625975B (en) * | 2022-05-16 | 2022-08-09 | 山东省科院易达科技咨询有限公司 | Knowledge graph-based customer behavior analysis system |
| CN116882495B (en) * | 2023-02-27 | 2024-02-09 | 中央民族大学 | Dynamic knowledge graph-based leading edge gene technology safe image drawing method |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107729490A (en) * | 2017-10-18 | 2018-02-23 | 华南农业大学 | Clothes personalized designs method and apparatus based on multi-source dynamic knowledge collection of illustrative plates |
| WO2018036239A1 (en) * | 2016-08-24 | 2018-03-01 | 慧科讯业有限公司 | Method, apparatus and system for monitoring internet media events based on industry knowledge mapping database |
| CN108694223A (en) * | 2018-03-26 | 2018-10-23 | 北京奇艺世纪科技有限公司 | The construction method and device in a kind of user's portrait library |
| CN108763555A (en) * | 2018-06-01 | 2018-11-06 | 北京奇虎科技有限公司 | Representation data acquisition methods and device based on demand word |
| CN110245874A (en) * | 2019-03-27 | 2019-09-17 | 中国海洋大学 | A kind of Decision fusion method based on machine learning and knowledge reasoning |
| CN111125524A (en) * | 2019-12-20 | 2020-05-08 | 四川文轩教育科技有限公司 | Data analysis and data push method based on knowledge graph |
-
2021
- 2021-01-08 CN CN202110025972.6A patent/CN113204636B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018036239A1 (en) * | 2016-08-24 | 2018-03-01 | 慧科讯业有限公司 | Method, apparatus and system for monitoring internet media events based on industry knowledge mapping database |
| CN107729490A (en) * | 2017-10-18 | 2018-02-23 | 华南农业大学 | Clothes personalized designs method and apparatus based on multi-source dynamic knowledge collection of illustrative plates |
| CN108694223A (en) * | 2018-03-26 | 2018-10-23 | 北京奇艺世纪科技有限公司 | The construction method and device in a kind of user's portrait library |
| CN108763555A (en) * | 2018-06-01 | 2018-11-06 | 北京奇虎科技有限公司 | Representation data acquisition methods and device based on demand word |
| CN110245874A (en) * | 2019-03-27 | 2019-09-17 | 中国海洋大学 | A kind of Decision fusion method based on machine learning and knowledge reasoning |
| CN111125524A (en) * | 2019-12-20 | 2020-05-08 | 四川文轩教育科技有限公司 | Data analysis and data push method based on knowledge graph |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113204636A (en) | 2021-08-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113204636B (en) | Knowledge graph-based user dynamic personalized image drawing method | |
| Ding et al. | A long video caption generation algorithm for big video data retrieval | |
| CN105824959B (en) | Public opinion monitoring method and system | |
| Kuanar et al. | Video key frame extraction through dynamic Delaunay clustering with a structural constraint | |
| CN110750656B (en) | Multimedia detection method based on knowledge graph | |
| Perret et al. | Evaluation of hierarchical watersheds | |
| CN112738556B (en) | Video processing method and device | |
| CN101305368A (en) | Semantic visual search engine | |
| WO2019011936A1 (en) | Method for evaluating an image | |
| CN111324765A (en) | Fine-grained sketch image retrieval method based on depth cascade cross-modal correlation | |
| CN112559684A (en) | Keyword extraction and information retrieval method | |
| dos Santos Belo et al. | Summarizing video sequence using a graph-based hierarchical approach | |
| Vijayarani et al. | Multimedia mining research-an overview | |
| CN115131698B (en) | Video attribute determining method, device, equipment and storage medium | |
| CN113806588A (en) | Method and device for searching video | |
| Sun et al. | Remote sensing image interpretation with semantic graph-based methods: A survey | |
| Tran et al. | Lifelog Moment Retrieval with Visual Concept Fusion and Text-based Query Expansion. | |
| CN117420998A (en) | Client UI interaction component generation method, device, terminal and medium | |
| Wu et al. | A new bag-of-words model using multi-cue integration for image retrieval | |
| CN108763560B (en) | Three-dimensional human body motion retrieval method based on graph model | |
| Devis et al. | Sketch based image retrieval using transfer learning | |
| Hybridised | OntoKnowNHS: Ontology Driven Knowledge Centric Novel Hybridised Semantic Scheme for Image Recommendation Using Knowledge Graph | |
| Noorian et al. | St-sem: A multimodal method for points-of-interest classification using street-level imagery | |
| Ning et al. | Optimized shape semantic graph representation for object understanding and recognition in point clouds | |
| Zhang et al. | Interactive mobile visual search for social activities completion using query image contextual model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |