CN109815338A - Relation extraction method and system in knowledge mapping based on mixed Gauss model - Google Patents
Relation extraction method and system in knowledge mapping based on mixed Gauss model Download PDFInfo
- Publication number
- CN109815338A CN109815338A CN201811625195.3A CN201811625195A CN109815338A CN 109815338 A CN109815338 A CN 109815338A CN 201811625195 A CN201811625195 A CN 201811625195A CN 109815338 A CN109815338 A CN 109815338A
- Authority
- CN
- China
- Prior art keywords
- entity
- mixed gauss
- gauss model
- entity relationship
- relationship
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000605 extraction Methods 0.000 title claims abstract description 41
- 238000013507 mapping Methods 0.000 title claims abstract description 39
- 239000013598 vector Substances 0.000 claims abstract description 45
- 238000000034 method Methods 0.000 claims abstract description 29
- 238000012549 training Methods 0.000 claims abstract description 18
- 238000012937 correction Methods 0.000 claims description 2
- 238000005086 pumping Methods 0.000 claims 1
- 230000006870 function Effects 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 239000000284 extract Substances 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 238000010225 co-occurrence analysis Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses Relation extraction method and systems in a kind of knowledge mapping based on mixed Gauss model, this method comprises: extracting entity from text, and disambiguation;Relationship characteristic extraction is carried out to the entity after disambiguation, obtains entity relationship feature vector;According to the entity relationship feature vector and mixed Gauss model, model training is carried out, entity relationship mixed Gauss model is obtained;Amendment is updated to entity relationship mixed Gauss model, obtains updated entity relationship mixed Gauss model;In knowledge mapping building process, it is based on the updated entity relationship mixed Gauss model, carries out entity relation extraction.Present invention seek to address that the problem of diversification of knowledge mapping incidence relation classification, multi-tag, realize the modeling to complex class incidence relation and feature distribution.
Description
Technical field
The invention belongs to close in knowledge mapping technical field more particularly to a kind of knowledge mapping based on mixed Gauss model
It is abstracting method and system.
Background technique
Knowledge mapping be by by the subjects such as applied mathematics, graphics, Information Visualization Technology, information science theory with
Method and the methods of meterological citation analysis, Co-occurrence Analysis combine, and the core of subject is visually shown using visual map
Core structure, developing history, Disciplinary Frontiers and whole Knowledge framework reach the modern theory of Multidisciplinary Integration purpose.It is complicated
Ken shown by data mining, information processing, knowledge measure and graphic plotting, disclose the dynamic of ken
The state rule of development provides practical, valuable reference for disciplinary study.Entity relation extraction is knowledge mapping core procedure,
Its key is the foundation of relationship training pattern.
Gaussian density function estimation is a kind of parameterized model.Gauss hybrid models (Gaussian Mixture Model,
GMM) be single Gaussian probability-density function extension, the Density Distribution of the enough smoothly approximate arbitrary shapes of gauss hybrid models.
Gauss hybrid models type has single Gauss model (Single Gaussian Model, SGM) and gauss hybrid models
(Gaussian Mixture Model, GMM) two classes.Similar to cluster, according to Gaussian probability-density function (Probability
Density Function, PDF) parameter is different, and each Gauss model is considered as a kind of classification, inputs a sample x, i.e.,
Its value can be calculated by PDF, then judge whether the sample belongs to Gauss model by a threshold value.
For knowledge mapping relationship modeling, every incidence relation might have more than one class label, this is just
So that the SGM model for being appropriate only for two category classification problems is not suitable for the application scenarios.
Summary of the invention
Technology of the invention solves the problems, such as: overcoming the deficiencies of the prior art and provide a kind of knowing based on mixed Gauss model
Know Relation extraction method and system in map, it is intended to solve the problems, such as the diversification of knowledge mapping incidence relation classification, multi-tag, it is real
The modeling to complex class incidence relation and feature distribution is showed.
In order to solve the above-mentioned technical problem, the invention discloses relationships in a kind of knowledge mapping based on mixed Gauss model
Abstracting method, comprising:
Entity, and disambiguation are extracted from text;
Relationship characteristic extraction is carried out to the entity after disambiguation, obtains entity relationship feature vector;
According to the entity relationship feature vector and mixed Gauss model, model training is carried out, obtains entity relationship mixing
Gauss model;
Amendment is updated to entity relationship mixed Gauss model, obtains updated entity relationship mixed Gauss model;
In knowledge mapping building process, it is based on the updated entity relationship mixed Gauss model, carries out entity pass
System extracts.
Preferably, entity, and disambiguation are extracted from text, comprising:
Using remote supervisory method, text is compared with the entity in existing knowledge base, entity is extracted from text, and
Disambiguation.
Preferably, relationship characteristic extraction is carried out to the entity after disambiguation, obtains entity relationship feature vector, comprising:
Using example multi-tag model modeling, feature representation is carried out between the relationship the entity after disambiguation, by entity
Between relationship be expressed as the element in vector space, i.e. entity relationship feature vector, and retain relationship type information.
Preferably, according to the entity relationship feature vector and mixed Gauss model, model training is carried out, obtains entity pass
It is mixed Gauss model, comprising:
Mixed Gauss model is called, and the mixed Gauss model is initialized, the poly- of mixed Gauss model is set
The quantity at class center;
Using determining cluster centre as prediction initial center point;
Computational entity relationship characteristic vector is clustered at a distance from each cluster centre;
According to cluster result, model training is carried out, the pass of each entity relationship feature vector and each cluster centre is converted to
Join likelihood angle value, obtains entity relationship mixed Gauss model.
Preferably, the quantity of the cluster centre of the mixed Gauss model of setting are as follows: 3~5.
Preferably, amendment is updated to entity relationship mixed Gauss model, it is high obtains updated entity relationship mixing
This model, comprising:
The new entity relationship feature vector of input is received, the quantity and new entity for constantly adjusting cluster centre are closed
It is feature vector at a distance from each cluster centre, completes the update of entity relationship mixed Gauss model, obtain updated entity
Relationship mixed Gauss model.
Preferably, in knowledge mapping building process, it is based on the updated entity relationship mixed Gauss model, is carried out
Entity relation extraction, comprising:
In knowledge mapping building process, when receiving new entity relationship, new entity relationship is expressed as entity
Relationship characteristic vector is expressed, and the updated entity relationship mixed Gauss model is based on, calculate with cluster centre away from
From, and select apart from the smallest classification;
Using the corresponding relational tags of the smallest classification of the distance selected as the label of the new entity relationship.
The invention also discloses Relation extraction systems in a kind of knowledge mapping based on mixed Gauss model, comprising:
Disambiguation module, for extracting entity, and disambiguation from text;
Extraction module obtains entity relationship feature vector for carrying out relationship characteristic extraction to the entity after disambiguation;
Model training module, for carrying out model training according to the entity relationship feature vector and mixed Gauss model,
Obtain entity relationship mixed Gauss model;
It updates correction module and obtains updated entity for being updated amendment to entity relationship mixed Gauss model
Relationship mixed Gauss model;
Relation extraction module, for being mixed based on the updated entity relationship high in knowledge mapping building process
This model carries out entity relation extraction.
The invention has the following advantages that
The present invention makes full use of advantage of the mixed Gauss model in clustering problem by extracting, and provides one kind towards knowledge
The relational learning method of map construction and update.Solve the problems, such as the diversification of knowledge mapping incidence relation classification, multi-tag, it is real
The modeling to complex class incidence relation and feature distribution is showed.
Detailed description of the invention
Fig. 1 is in the embodiment of the present invention in a kind of knowledge mapping based on mixed Gauss model the step of Relation extraction method
Flow chart;
Fig. 2 is a kind of knowledge mapping incidence relation modeling process signal based on mixed Gauss model in the embodiment of the present invention
Figure.
Specific embodiment
To make the object, technical solutions and advantages of the present invention clearer, public to the present invention below in conjunction with attached drawing
Embodiment is described in further detail.
Such as Fig. 1, in the present embodiment, Relation extraction method in the knowledge mapping based on mixed Gauss model, comprising:
Step 101, entity, and disambiguation are extracted from text.
In the present embodiment, remote supervisory method can be used, text is compared with the entity in existing knowledge base, from
Entity, and disambiguation are extracted in text.
Step 102, relationship characteristic extraction is carried out to the entity after disambiguation, obtains entity relationship feature vector.
In the present embodiment, can use example multi-tag model modeling, between the relationship the entity after disambiguation into
Relationship between entity is expressed as the element in vector space, i.e. entity relationship feature vector, and retains relationship by row feature representation
Type information.
Step 103, according to the entity relationship feature vector and mixed Gauss model, model training is carried out, entity is obtained
Relationship mixed Gauss model.
In the present embodiment, mixed Gauss model can be called, and the mixed Gauss model is initialized, is arranged
The quantity of the cluster centre of mixed Gauss model;Using determining cluster centre as prediction initial center point;Computational entity relationship
Feature vector is clustered at a distance from each cluster centre;According to cluster result, model training is carried out, each reality is converted to
Body relationship characteristic vector is associated with likelihood angle value with each cluster centre, obtains entity relationship mixed Gauss model.
Preferably, the quantity of the cluster centre of the mixed Gauss model of setting are as follows: 3~5.
Step 104, amendment is updated to entity relationship mixed Gauss model, it is high obtains updated entity relationship mixing
This model.
In the present embodiment, the new entity relationship feature vector of input is received, the quantity of cluster centre is constantly adjusted, with
And new entity relationship feature vector is completed the update of entity relationship mixed Gauss model, is obtained at a distance from each cluster centre
Updated entity relationship mixed Gauss model.
Step 105, in knowledge mapping building process, it is based on the updated entity relationship mixed Gauss model, into
Row entity relation extraction.
In the present embodiment, in knowledge mapping building process, when receiving new entity relationship, new entity is closed
System is expressed as entity relationship feature vector and is expressed, and is based on the updated entity relationship mixed Gauss model, calculate with
The distance of cluster centre, and select apart from the smallest classification;Using the corresponding relational tags of the smallest classification of the distance selected as
The label of the new entity relationship.
On the basis of the above embodiments, it is illustrated below with reference to a specific example.
Such as Fig. 2, explanation is explained in detail in conjunction with the knowledge mapping modeling example of target association relationship in a harbour, into
And show the multi-tag incidence relation modeling ability of gauss hybrid models in the present invention:
(1) entity is extracted and is disambiguated.
Based on air force, Han Deng foreign military, U.S. Japan and the United States military exercises news report acquired in open number source, long-range prison is utilized
(Distance Supervison) method is superintended and directed by filling with typical foreign military's military affairs in existing knowledge base (knowledge base)
The mode that standby entity is compared extracts main military hardware target entity involved by military exercise from public data source text information
{ei, and the disambiguations such as entity attributes, title, vocabulary of the same name are equipped according to extracting.
(2) entity relationship feature extraction.
It is now assumed that certain model military aircraft A and aircraft B involved in military exercise, uses more example multi-tag (multi-
Instance multi-label) model modeling mode, to aircraft A and aircraft B entity to (ei,ej) between relationship carry out it is special
Sign expression, the element z relationship being abstracted between entity being expressed as in vector space* i,j, and retain possible relationship as much as possible
Type information y 'k.For example, relationship type that may be present includes escorting, refueling, cooperating, relieving a garrison between aircraft A and aircraft B,
All as entity to (ei,ej) between possible relationship type label retained and recorded, convenient for subsequent relationship cluster.
(3) the entity relationship model training based on mixed Gauss model.
Firstly, disclosing report text, military reference based on history, in foreign military's equipment technology handbook and A, Type B aircraft
Relevant partial contextual information and prior information carry out just the incidence relation mixed Gauss model between aircraft A and aircraft B
Beginningization, setting mixed Gauss model cluster centre quantity is 3-5, roughly the same with possible relationship type, is expressed as predicting
Initial center point < μk,Σk>。
Then, aircraft A, B sample entity relationship feature vector z are calculated* w,jAt a distance from cluster centreAnd it is clustered.
Finally, obtaining A, B aircraft entity relationship mixed Gauss model by training, current context and priori letter are converted to
The a variety of possible feature vectors of relationship type of A, Type B aircraft and the association likelihood angle value of each cluster centre in breath
(4) entity relationship model based on mixed Gauss model updates.
Input discloses the relevant new entity relationship feature of extracted A, Type B aircraft in report and priori material from history
Vector, constantly adjustment A, the cluster centre quantity of Type B aircraft entity relationship model and position < μ in vector spacek,Σk>.Its
In, new cluster centre is calculated according to multivariate Gaussian modelsComplete A, Type B aircraft entity relationship model
Update.
(5) entity relation extraction.
In A, Type B aircraft knowledge mapping incidence relation building process based on analysis required in current task, for working as
New entity relationship relevant with A, Type B aircraft in preceding open report context, is expressed as feature vector z first* i,j, calculate with
The distance of mixed Gaussian relational model cluster centreSelection is apart from the smallest classificationThen the corresponding relational tags of the category by currently analyze in open military affairs message with A, B
Relationship type label y ' between type aircraft entityk。
On the basis of the above embodiments, the invention also discloses close in a kind of knowledge mapping based on mixed Gauss model
It is extraction system, comprising: disambiguation module, for extracting entity, and disambiguation from text;Extraction module, for pair
Entity after disambiguation carries out relationship characteristic extraction, obtains entity relationship feature vector;Model training module, for according to institute
Entity relationship feature vector and mixed Gauss model are stated, model training is carried out, obtains entity relationship mixed Gauss model;Update is repaired
Positive module obtains updated entity relationship mixed Gaussian mould for being updated amendment to entity relationship mixed Gauss model
Type;Relation extraction module, for being based on the updated entity relationship mixed Gaussian mould in knowledge mapping building process
Type carries out entity relation extraction.
For system embodiments, since it is corresponding with embodiment of the method, so be described relatively simple, correlation
Place referring to embodiment of the method part explanation.
Various embodiments are described in a progressive manner in this explanation, the highlights of each of the examples are with its
The difference of his embodiment, the same or similar parts between the embodiments can be referred to each other.
The above, optimal specific embodiment only of the invention, but scope of protection of the present invention is not limited thereto,
In the technical scope disclosed by the present invention, any changes or substitutions that can be easily thought of by anyone skilled in the art,
It should be covered by the protection scope of the present invention.
The content that description in the present invention is not described in detail belongs to the well-known technique of professional and technical personnel in the field.
Claims (8)
1. a kind of Relation extraction method in knowledge mapping based on mixed Gauss model characterized by comprising
Entity, and disambiguation are extracted from text;
Relationship characteristic extraction is carried out to the entity after disambiguation, obtains entity relationship feature vector;
According to the entity relationship feature vector and mixed Gauss model, model training is carried out, entity relationship mixed Gaussian is obtained
Model;
Amendment is updated to entity relationship mixed Gauss model, obtains updated entity relationship mixed Gauss model;
In knowledge mapping building process, it is based on the updated entity relationship mixed Gauss model, carries out entity relationship pumping
It takes.
2. Relation extraction method in the knowledge mapping according to claim 1 based on mixed Gauss model, which is characterized in that
Entity, and disambiguation are extracted from text, comprising:
Using remote supervisory method, text is compared with the entity in existing knowledge base, entity is extracted from text, and eliminate
Ambiguity.
3. Relation extraction method in the knowledge mapping according to claim 1 based on mixed Gauss model, which is characterized in that
Relationship characteristic extraction is carried out to the entity after disambiguation, obtains entity relationship feature vector, comprising:
Using example multi-tag model modeling, feature representation is carried out between the relationship the entity after disambiguation, it will be between entity
Relationship is expressed as the element in vector space, i.e. entity relationship feature vector, and retains relationship type information.
4. Relation extraction method in the knowledge mapping according to claim 3 based on mixed Gauss model, which is characterized in that
According to the entity relationship feature vector and mixed Gauss model, model training is carried out, entity relationship mixed Gauss model is obtained,
Include:
Mixed Gauss model is called, and the mixed Gauss model is initialized, is arranged in the cluster of mixed Gauss model
The quantity of the heart;
Using determining cluster centre as prediction initial center point;
Computational entity relationship characteristic vector is clustered at a distance from each cluster centre;
According to cluster result, model training is carried out, is converted to being associated with seemingly for each entity relationship feature vector and each cluster centre
Right angle value obtains entity relationship mixed Gauss model.
5. Relation extraction method in the knowledge mapping according to claim 4 based on mixed Gauss model, which is characterized in that
The quantity of the cluster centre of the mixed Gauss model of setting are as follows: 3~5.
6. Relation extraction method in the knowledge mapping according to claim 1 based on mixed Gauss model, which is characterized in that
Amendment is updated to entity relationship mixed Gauss model, obtains updated entity relationship mixed Gauss model, comprising:
The new entity relationship feature vector of input is received, quantity and the new entity relationship for constantly adjusting cluster centre are special
Vector is levied at a distance from each cluster centre, the update of entity relationship mixed Gauss model is completed, obtains updated entity relationship
Mixed Gauss model.
7. Relation extraction method in the knowledge mapping according to claim 1 based on mixed Gauss model, which is characterized in that
In knowledge mapping building process, it is based on the updated entity relationship mixed Gauss model, carries out entity relation extraction, packet
It includes:
In knowledge mapping building process, when receiving new entity relationship, new entity relationship is expressed as entity relationship
Feature vector is expressed, and the updated entity relationship mixed Gauss model is based on, and is calculated at a distance from cluster centre, and
Selection is apart from the smallest classification;
Using the corresponding relational tags of the smallest classification of the distance selected as the label of the new entity relationship.
8. Relation extraction system in a kind of knowledge mapping based on mixed Gauss model characterized by comprising
Disambiguation module, for extracting entity, and disambiguation from text;
Extraction module obtains entity relationship feature vector for carrying out relationship characteristic extraction to the entity after disambiguation;
Model training module, for carrying out model training, obtaining according to the entity relationship feature vector and mixed Gauss model
Entity relationship mixed Gauss model;
It updates correction module and obtains updated entity relationship for being updated amendment to entity relationship mixed Gauss model
Mixed Gauss model;
Relation extraction module, for being based on the updated entity relationship mixed Gaussian mould in knowledge mapping building process
Type carries out entity relation extraction.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811625195.3A CN109815338B (en) | 2018-12-28 | 2018-12-28 | Method and system for extracting relation in knowledge graph based on Gaussian mixture model |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811625195.3A CN109815338B (en) | 2018-12-28 | 2018-12-28 | Method and system for extracting relation in knowledge graph based on Gaussian mixture model |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109815338A true CN109815338A (en) | 2019-05-28 |
CN109815338B CN109815338B (en) | 2021-10-22 |
Family
ID=66602734
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811625195.3A Expired - Fee Related CN109815338B (en) | 2018-12-28 | 2018-12-28 | Method and system for extracting relation in knowledge graph based on Gaussian mixture model |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109815338B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111241360A (en) * | 2020-01-09 | 2020-06-05 | 腾讯科技(深圳)有限公司 | Information recommendation method, device, equipment and storage medium |
CN114186690A (en) * | 2022-02-16 | 2022-03-15 | 中国空气动力研究与发展中心计算空气动力研究所 | Aircraft knowledge graph construction method, device, equipment and storage medium |
CN114595337A (en) * | 2022-01-28 | 2022-06-07 | 河南大学 | Method for constructing curriculum knowledge graph based on GMM |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20140198998A1 (en) * | 2013-01-14 | 2014-07-17 | Samsung Electronics Co., Ltd. | Novel criteria for gaussian mixture model cluster selection in scalable compressed fisher vector (scfv) global descriptor |
CN104809176A (en) * | 2015-04-13 | 2015-07-29 | 中央民族大学 | Entity relationship extracting method of Zang language |
CN106055675A (en) * | 2016-06-06 | 2016-10-26 | 杭州量知数据科技有限公司 | Relation extracting method based on convolution neural network and distance supervision |
CN106156286A (en) * | 2016-06-24 | 2016-11-23 | 广东工业大学 | Type extraction system and method towards technical literature knowledge entity |
CN106570148A (en) * | 2016-10-27 | 2017-04-19 | 浙江大学 | Convolutional neutral network-based attribute extraction method |
CN106934032A (en) * | 2017-03-14 | 2017-07-07 | 软通动力信息技术(集团)有限公司 | A kind of city knowledge mapping construction method and device |
CN108241669A (en) * | 2016-12-26 | 2018-07-03 | 上海神计信息***工程有限公司 | A kind of construction method and system of adaptive text feature cluster |
-
2018
- 2018-12-28 CN CN201811625195.3A patent/CN109815338B/en not_active Expired - Fee Related
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20140198998A1 (en) * | 2013-01-14 | 2014-07-17 | Samsung Electronics Co., Ltd. | Novel criteria for gaussian mixture model cluster selection in scalable compressed fisher vector (scfv) global descriptor |
CN104809176A (en) * | 2015-04-13 | 2015-07-29 | 中央民族大学 | Entity relationship extracting method of Zang language |
CN106055675A (en) * | 2016-06-06 | 2016-10-26 | 杭州量知数据科技有限公司 | Relation extracting method based on convolution neural network and distance supervision |
CN106156286A (en) * | 2016-06-24 | 2016-11-23 | 广东工业大学 | Type extraction system and method towards technical literature knowledge entity |
CN106570148A (en) * | 2016-10-27 | 2017-04-19 | 浙江大学 | Convolutional neutral network-based attribute extraction method |
CN108241669A (en) * | 2016-12-26 | 2018-07-03 | 上海神计信息***工程有限公司 | A kind of construction method and system of adaptive text feature cluster |
CN106934032A (en) * | 2017-03-14 | 2017-07-07 | 软通动力信息技术(集团)有限公司 | A kind of city knowledge mapping construction method and device |
Non-Patent Citations (1)
Title |
---|
延浩然等: "一种改进的实体关系抽取算法—OptMultiR", 《中文信息学报》 * |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111241360A (en) * | 2020-01-09 | 2020-06-05 | 腾讯科技(深圳)有限公司 | Information recommendation method, device, equipment and storage medium |
CN111241360B (en) * | 2020-01-09 | 2023-03-21 | 深圳市雅阅科技有限公司 | Information recommendation method, device, equipment and storage medium |
CN114595337A (en) * | 2022-01-28 | 2022-06-07 | 河南大学 | Method for constructing curriculum knowledge graph based on GMM |
CN114595337B (en) * | 2022-01-28 | 2024-06-28 | 河南大学 | Method for constructing course knowledge graph based on GMM |
CN114186690A (en) * | 2022-02-16 | 2022-03-15 | 中国空气动力研究与发展中心计算空气动力研究所 | Aircraft knowledge graph construction method, device, equipment and storage medium |
CN114186690B (en) * | 2022-02-16 | 2022-04-19 | 中国空气动力研究与发展中心计算空气动力研究所 | Aircraft knowledge graph construction method, device, equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN109815338B (en) | 2021-10-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111160474B (en) | Image recognition method based on deep course learning | |
CN111598164B (en) | Method, device, electronic equipment and storage medium for identifying attribute of target object | |
CN110781663B (en) | Training method and device of text analysis model, text analysis method and device | |
Mohandes et al. | A signer-independent Arabic Sign Language recognition system using face detection, geometric features, and a Hidden Markov Model | |
CN110019843A (en) | The processing method and processing device of knowledge mapping | |
CN109815338A (en) | Relation extraction method and system in knowledge mapping based on mixed Gauss model | |
CN110457677B (en) | Entity relationship identification method and device, storage medium and computer equipment | |
CN105512627A (en) | Key point positioning method and terminal | |
CN106340297A (en) | Speech recognition method and system based on cloud computing and confidence calculation | |
CN111210111B (en) | Urban environment assessment method and system based on online learning and crowdsourcing data analysis | |
CN108959474B (en) | Entity relation extraction method | |
CN114419736A (en) | Experiment scoring method, system, equipment and readable storage medium | |
CN110929640B (en) | Wide remote sensing description generation method based on target detection | |
CN111666766A (en) | Data processing method, device and equipment | |
CN112214595A (en) | Category determination method, device, equipment and medium | |
CN113434722B (en) | Image classification method, device, equipment and computer readable storage medium | |
CN110458215A (en) | Pedestrian's attribute recognition approach based on multi-time Scales attention model | |
Feng | Mask RCNN-based single shot multibox detector for gesture recognition in physical education | |
CN112329830B (en) | Passive positioning track data identification method and system based on convolutional neural network and transfer learning | |
CN115705393A (en) | Radar radiation source grading identification method based on continuous learning | |
CN111695570B (en) | Small sample lower semantic segmentation method based on variational prototype reasoning | |
CN114581761A (en) | Remote sensing image recognition method, device, equipment and computer readable storage medium | |
CN117312140A (en) | Method and device for generating test case, electronic equipment and storage medium | |
CN115482436B (en) | Training method and device for image screening model and image screening method | |
Shao et al. | Attention GAN-based method for designing intelligent making system |
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 | ||
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20211022 |