CN102982076B - Based on the various dimensions content mask method in semantic label storehouse - Google Patents
Based on the various dimensions content mask method in semantic label storehouse Download PDFInfo
- Publication number
- CN102982076B CN102982076B CN201210424525.9A CN201210424525A CN102982076B CN 102982076 B CN102982076 B CN 102982076B CN 201210424525 A CN201210424525 A CN 201210424525A CN 102982076 B CN102982076 B CN 102982076B
- Authority
- CN
- China
- Prior art keywords
- tag
- storehouse
- resource
- label
- mark
- 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
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of various dimensions content mask method based on semantic label storehouse, comprising: set up semantic label storehouse; Configure extendible resource category; Set up multistage, configurable content mark dimension; By resource according to content characteristic partition dimension, set up multi-level content dimension; Set up the corresponding relation that configurable, revisable resource category and content mark dimension; The resource content carried out based on semantic label storehouse marks; Temporary labels process; Based on the resource retrieval in semantic label storehouse; User inputs term, and system is mated automatically in extension tag storehouse: if the match is successful, and system retrieves corresponding picture according to the label for labelling code of correspondence; If mate unsuccessful, term can mate with resource description information by system, simultaneity factor by this term stored in temporary labels storehouse.Effectively raise degree of accuracy and the efficiency of resource mark, for resource retrieval and data analysis are had laid a good foundation.
Description
Technical field
The present invention relates to data mining, data analysis and knowledge reasoning field, designed and Implemented one and various dimensions, semantization, structurized mask method are carried out to resource content.
Background technology
In recent years, along with the high speed development of economic society, the quantity of resource increases sharply, and the development of resource mark is relatively slow, and the search problem of resource becomes increasingly conspicuous.Resources for research mask method effectively can solve the management and retrieval problem of resource, improve the utilization rate of resource, meet resource in efficiency, use and managerial requirement, this is by the research and development to China's present stage intelligent dimension, play positive impetus to the reasonable efficiency utilization of resource.
At present, resource mask method has a lot, mainly can be divided into the semanteme marking method of method, the character injecting method based on resource content, the label for labelling method based on resource content and the body based on resource specific area marked based on Resource Properties.
Method based on Resource Properties mark realizes mainly through the mode marking corresponding value for the attributive character of resource.This mode is simple and easy to use, can describe preferably, can be used as the basic data of resource retrieval, but attribute information is only the fraction of information contained by resource, lacks the description to resource content semantic information the important attribute information of resource; Attribute item need be determined when system, and not easily revise, extendability is poor; The simple text of non-standardization is mated the semantic ambiguity caused and is difficult to avoid.
Character injecting method based on resource content mainly extracts the generic features of resource or field correlated characteristic marks resource.This mode generally uses computing machine automatic business processing, and takes full advantage of the abundant content information that resource itself comprises, good in domain specific application, but resource content becomes the problem needing solution badly how to utilize the feature of resource effectively to represent.
Label for labelling method based on resource content mainly marks resource with label.This mode breaches the limitation of attribute labeling, disclose content and the theme feature of resource, but common socialized label exists, and definition is not strict, variable, unwatched deficiency, make the subjectivity of label for labelling strong, polysemant and synonym easily cause semanteme to obscure, annotating efficiency is low, and retrieval is mated with the word of mark and is difficult to coincide.
Semanteme marking method based on the body of resource specific area mainly carries out resource mark by the ontology in semantic net.Originally isolated resource relationship is got up by this mode, increases the degree of coupling between different resource, and resource ontology is that standardization mark provides formalization basis, and resource after mark is corresponding with domain body, can realize the intelligent retrieval of resource; But the structure of domain body not overnight just can complete, and the category that resource relates to is very extensive, relies on body completely and carries out the general of resource and mark completely not have actual operation at present.
Summary of the invention
The object of the invention is to propose a kind of various dimensions content mask method based on semantic label storehouse, to reach higher resource annotating efficiency, improve the precision of mark, for efficient resource retrieval lays the foundation.
The concrete steps that a kind of various dimensions content mask method based on semantic label storehouse of the present invention realizes are described below:
(1) semantic label storehouse is set up; Semantic label storehouse refers to the semantic system of the label be made up of canonical tag storehouse, extension tag storehouse, temporary labels storehouse, label correlation database and label data analysis, and wherein extension tag storehouse comprises the content in canonical tag storehouse.
The formal label of mark resource is stored, i.e. canonical tag in canonical tag storehouse.Canonical tag is only had just to be assigned with mark code.Canonical tag adopts grouping-hierarchy management: first press the grouping of word category division, then to often organizing canonical tag layering, builds the tag set of a tree structure, and is that each canonical tag distributes a mark code automatically.Be polysemant label with this label of the different representation of word, represent that this group label is synonym set of tags with code dissenting words.In addition, mark code can be used to be mapped by the label of different language, to realize multi-language tag expansion.
Extension tag and whole canonical tag is stored in extension tag storehouse.Extension tag refers to a series of expansion words of certain canonical tag corresponding, and itself does not have mark code.Extension tag and resource do not have direct correlation relation, but have indirect association relation by the canonical tag of its correspondence.Extension tag is bound to have incidence relation with certain or multiple canonical tag, and namely can be obtained one group of extension tag of its correspondence by canonical tag, vice versa.Extension tag storehouse main application comprises two aspects: during mark resource, when indexer inputs word, system mates canonical tag corresponding to this word from extension tag storehouse, is prompted to indexer.During retrieve resources, user inputs keyword when retrieving, and system mates canonical tag corresponding to this word and mark code thereof from extension tag storehouse, and then searches the resource of this mark code correspondence.
Temporary labels is the word not belonging to canonical tag and extension tag that indexer adds temporarily in resource annotation process, does not have mark code.Due to canonical tag storehouse along with the work of resource mark is improved and expansion gradually, so indexer or other unprofessional users are when marking resource, the keyword (i.e. temporary labels) that can not have in operating specification tag library and extension tag storehouse according to actual needs marks resource.
The information such as the outgoing label degree of association, label temperature (comprehensive label is used to the frequency marking and retrieve) is mainly analyzed to obtain in label data analysis, by the semantic information of label abundantization more, is resource mark and retrieval service.Data analysis can be carried out: (1) carries out label Co-occurrence Analysis to label that certain resource marks from following three aspects; (2) label used during user search resource is recorded and analyzed; (3) to similar resource (manually arrange and automatically know method for distinguishing and determine) note label and carry out statistical study.
The result of label correlation database stored tag data analysis, for intelligent recommendation when label for labelling and retrieval.
(2) extendible resource category is configured.
Wherein, resource supports the multimedia resource kinds such as picture, audio frequency, video, and allows to carry out dynamic conditioning to it.
(3) multistage, configurable content mark dimension is set up.By resource according to content characteristic partition dimension, set up multi-level content dimension.
Wherein, content mark dimension refers to multiple gradable mark dimension, supports the mark dimension that different types of resource is corresponding different, for retraining and specification the label for labelling of resource.
(4) corresponding relation that configurable, revisable resource category and content mark dimension is set up.
(5) resource content carried out based on semantic label storehouse marks.During mark resource, indexer directly can choose canonical tag and mark from canonical tag storehouse, also index term can be inputted, system automatic benchmarking draws word and mates in extension tag storehouse: if the match is successful, in canonical tag storehouse, then obtain canonical tag and mark code thereof, set up resource and the corresponding relation marking code; If mate unsuccessful, then index term is retained this word stored in temporary labels storehouse and is marked the corresponding relation of resource.In annotation process, system carries out intelligent recommendation according to label correlation database.
(6) temporary labels process.
Tag control person audits one by one to temporary labels, adopts the processing mode that two kinds main: one is the standard according to canonical tag and extension tag, and temporary labels is directly set as canonical tag or extension tag; Two is directly delete this temporary labels.In addition, existing canonical tag or extension tag can also be selected to replace this temporary labels.
(7) based on the resource retrieval in semantic label storehouse.User inputs term, and system is mated automatically in extension tag storehouse; If the match is successful, system retrieves corresponding picture according to the label for labelling code of correspondence; If mate unsuccessful, term can mate with resource description information by system, simultaneity factor by this term stored in temporary labels storehouse.
The present invention compared with prior art, has following obvious advantage and beneficial effect:
First, the present invention, on the basis of abundant resources for research content, proposes the various dimensions mark system of resource content, and the resource content dimension of further refinement contributes to more accurate content mark and retrieval.Secondly, in order to avoid the impact that semantic ambiguity marks for resource, the present invention proposes the semantic Intelligent Support system in semantic label storehouse first in resource mark: design specifications label supports polysemant, synonym and multilingual, extension tag effectively raises the accuracy of mark and the universality of retrieval, and label correlation database further enhances excavation and the utilization of label semantic information.Again, this method is all applicable for all kinds of resource, support the personalization setting of different resource, mark dimension can be managed, can be joined, easily extensible, in semantic label storehouse, each ingredient all has good extendability, wherein the data analysis of label can adopt day by day perfect data analysis technique, obtains better analytical effect.Experiment proves that the method effectively raises degree of accuracy and the efficiency of resource mark, for resource retrieval and data analysis are had laid a good foundation.
Accompanying drawing explanation
Fig. 1 is the various dimensions content mask method process flow diagram based on semantic label storehouse;
Fig. 2 is semantic label library structure schematic diagram;
Fig. 3 is the various dimensions content mask method structural representation based on semantic label storehouse;
Fig. 4 is resource content mark process flow diagram;
Fig. 5 is resource retrieval process flow diagram.
Embodiment
Below in conjunction with Figure of description, specific embodiments of the invention are illustrated.
The present invention, based on semantic label storehouse, carries out various dimensions, semantization, structurized mark to resource content, for effective retrieval of resource and application provide safeguard.The deficiencies such as semantic label storehouse compensate for that the subjectivity that traditional society Focus label exists is strong, ambiguousness, dispersion are unordered are the label systems of an ALARA Principle, easily extensible, structuring, semantization.
Referring to shown in Fig. 1, is the various dimensions content mask method process flow diagram based on semantic label storehouse.
Sequentially comprise: (1) sets up the semantic label storehouse of picture; (2) extendible picture categories is configured; (3) multistage, configurable image content mark dimension is set up; (4) corresponding relation that configurable, revisable picture categories and image content mark dimension is set up; (5) image content based on semantic label storehouse marks; (6) temporary labels process; (7) based on the picture retrieval in semantic label storehouse.
As shown in Figure 3, based on the various dimensions content mask method structural representation in semantic label storehouse.
The method forms semantic label storehouse by canonical tag storehouse, extension tag storehouse, temporary labels storehouse, label correlation database, and basis, semantic label storehouse realizes resource mark and resource retrieval.
Below described in detail:
(1) the semantic label storehouse of picture is set up.
Set up the semantic label storehouse of picture, as shown in Figure 2, semantic label storehouse is made up of canonical tag storehouse, extension tag storehouse, temporary labels storehouse, label correlation database and label data analytical approach.
(a) canonical tag storehouse: canonical tag contains the semantic concept of set of tags, mark code, synonym label, polysemant label and multi-language tag.
First according to image content feature by labeled packet, as set of tags such as action, particular persons, region, politics, physical culture;
Secondly also mark code for each canonical tag distributes one to each set of tags layering, can be divided into the one-level labels such as China, the world, city, rural area as set of tags region, China can be divided into the secondary labels etc. such as east, western part, middle part; Mark code oneself can formulate code value rule, according to rule for each canonical tag distributes a mark code.
In canonical tag storehouse, be polysemant label with this label of the different representation of word, represent that this group label is synonym label with code dissenting words, and mark code can be used to be mapped by the label of different language, realize the support of multi-language tag." Li Na " as different labeled code can be athletic Li Na, also can be sing Li Na, the two is polysemant label, can by different mark codes, parent label even set of tags distinguish; Both " happiness " and " happiness " signs of identical mark code are synonym labels.
(b) extension tag storehouse: the extension tag in picture expanding library and picture do not have direct correlation relation, but set up indirect association relation by the canonical tag of its correspondence and picture.
As, the corresponding one group of extension tag of canonical tag " happiness " possibility, comprises " happiness " " pleasure " " excitedly " " happily " etc.Extension tag is that the expansion of canonical tag illustrates, enhances the semantic connotation of label.
(c) temporary labels storehouse: temporary labels just can use after tag control person's examination & verification.Due to canonical tag storehouse along with the work of picture mark is improved and expansion gradually, so indexer or other unprofessional users are when marking picture, the keyword (i.e. temporary labels) that can not have in operating specification tag library and extension tag storehouse according to actual needs marks picture.
(d) label correlation database: what deposit in label correlation database is semantic association relation between canonical tag, this incidence relation is drawn by label data analysis result, label correlation database is auxiliary improves intelligent recommendation function, improves mark and effectiveness of retrieval and quality.
(e) label data analysis: data analysis can be carried out from following aspect: label Co-occurrence Analysis is carried out to label that picture marks; Label used during user search picture is recorded and analyzed; To similar pictures (manually arrange and automatically know method for distinguishing and determine) note label and carry out statistical study.By data analysis and the incidence relation that excavates between canonical tag, set up the semantic association between label, improve picture mark and effectiveness of retrieval and quality.
(2) extendible picture categories is configured.
According to picture feature, picture is divided into editor's class picture and intention class picture.
(3) multistage, configurable image content mark dimension is set up.
According to picture feature, picture is divided into the one-level dimensions such as attribute layer, content layer, content layer dimension can be divided into the secondary dimension such as personage, spot for photography; Personage's dimension can be divided into three grades of dimensions such as particular persons, sex, age; The multiple levels of content dimension of picture can be set up as required.
(4) corresponding relation that configurable, revisable picture categories and image content mark dimension is set up.
Configure the corresponding relation of picture categories and dimension voluntarily, as editor's class picture can set up corresponding relation with personage, spot for photography etc.; The corresponding relation safeguarding kind and dimension can be revised according to actual needs.
(5) image content based on semantic label storehouse marks.
Picture mark is the various dimensions content annotation process based on semantic label storehouse, as shown in Figure 4, can be divided into following steps:
During mark picture, indexer directly can choose canonical tag and mark from canonical tag storehouse, also index term can be inputted, system automatic benchmarking draws word and mates in extension tag storehouse: if the match is successful, in canonical tag storehouse, then obtain canonical tag and mark code thereof, set up picture and the corresponding relation marking code; If mate unsuccessful, then index term is retained this word stored in temporary labels storehouse and is marked the corresponding relation of picture.In annotation process, system carries out intelligent recommendation according to label correlation database.
(6) temporary labels process.
Temporary labels is audited through tag control person could formally for image content mark, and one is the standard according to canonical tag and extension tag, temporary labels is directly set as canonical tag or extension tag; Two is directly delete this temporary labels.In addition, existing canonical tag or extension tag can also be selected to replace this temporary labels.
(7) based on the picture retrieval in semantic label storehouse.
Picture retrieval is based on semantic label storehouse, and the retrieval use procedure after various dimensions mark, as shown in Figure 5, can be divided into following steps:
During retrieving image, user inputs term, and system is mated automatically in extension tag storehouse: if the match is successful, then obtain the mark code of this term, and system utilizes mark code intelligent recommendation and retrieving image; If mate unsuccessful, then by term and picture description information, as keyword etc. mates, simultaneity factor by this term stored in temporary labels storehouse;
Wherein, intelligent recommendation refers to according to mark code, find all canonical tag with this mark code, then find the canonical tag with semantic association that the tag extension relation that label associates and extension tag storehouse provides provided through label correlation database is recommended, according to these label search pictures.
Claims (4)
1., based on the various dimensions content mask method in semantic label storehouse, it is characterized in that, comprise the following steps:
1.1 set up semantic label storehouse; Semantic label storehouse is made up of canonical tag storehouse, extension tag storehouse, temporary labels storehouse, label correlation database and label data analysis, and wherein extension tag storehouse comprises the content in canonical tag storehouse;
The extendible resource category of 1.2 configuration;
1.3 set up multistage, configurable content mark dimension; By resource according to content characteristic partition dimension, set up multi-level content dimension;
1.4 set up the corresponding relation that configurable, revisable resource category and content mark dimension;
1.5 resource contents carried out based on semantic label storehouse mark; During mark resource, directly from canonical tag storehouse, choose canonical tag to mark, also index term can be inputted, system automatic benchmarking draws word and mates in extension tag storehouse: if the match is successful, in canonical tag storehouse, then obtain canonical tag and mark code thereof, set up resource and the corresponding relation marking code; If mate unsuccessful, then index term is retained this word stored in temporary labels storehouse and is marked the corresponding relation of resource; In annotation process, system carries out intelligent recommendation according to label correlation database;
1.6 temporary labels process; Tag control person will audit temporary labels one by one, or be set as new canonical tag or extension tag, or be deleted;
1.7 based on the resource retrieval in semantic label storehouse; User inputs term, and system is mated automatically in extension tag storehouse: if the match is successful, and system retrieves corresponding picture according to the label for labelling code of correspondence; If mate unsuccessful, term can mate with resource description information by system, simultaneity factor by this term stored in temporary labels storehouse;
The formal label of mark resource is stored, i.e. canonical tag in described canonical tag storehouse; Canonical tag is only had just to be assigned with mark code; Canonical tag adopts grouping-hierarchy management: first press the grouping of word category division, then to often organizing canonical tag layering, builds the tag set of a tree structure, and is that each canonical tag distributes a mark code automatically; Be polysemant label with this label of the different representation of word, represent that this group label is synonym set of tags with code dissenting words; In addition, mark code can be used to be mapped by the label of different language, to realize multi-language tag expansion;
Extension tag and whole canonical tag is stored in described extension tag storehouse; Extension tag is a series of expansion words of certain canonical tag corresponding, and itself does not have mark code; Extension tag and resource do not have direct correlation relation, but have indirect association relation by the canonical tag of its correspondence; Extension tag is bound to have incidence relation with certain or multiple canonical tag, and namely can be obtained one group of extension tag of its correspondence by canonical tag, vice versa; Extension tag storehouse comprises two aspects: during mark resource, when indexer inputs word, system mates canonical tag corresponding to this word from extension tag storehouse, is prompted to indexer; During retrieve resources, user inputs keyword when retrieving, and system mates canonical tag corresponding to this word and mark code thereof from extension tag storehouse, and then searches the resource of this mark code correspondence;
Described temporary labels storehouse is the word not belonging to canonical tag and extension tag that indexer adds temporarily in resource annotation process, does not have mark code; The keyword mark resource do not had in operating specification tag library and extension tag storehouse according to actual needs;
The result of described label correlation database stored tag data analysis, for intelligent recommendation when label for labelling and retrieval.
2. a kind of various dimensions content mask method based on semantic label storehouse according to claim 1, is characterized in that: the extendible resource category of described configuration, supports picture, audio frequency, video multimedia resource category, and allows to carry out dynamic conditioning to it.
3. a kind of various dimensions content mask method based on semantic label storehouse according to claim 1, it is characterized in that: described setting up multistage, configurable content mark dimension, multiple gradable mark dimension, support the mark dimension that different types of resource is corresponding different, for retraining and specification the label for labelling of resource.
4. a kind of various dimensions content mask method based on semantic label storehouse according to claim 1, it is characterized in that: tag control person audits described temporary labels, adopt the standard according to canonical tag and extension tag, temporary labels is directly set as canonical tag or extension tag, or directly deletes this temporary labels; Existing canonical tag or extension tag can also be selected to replace this temporary labels.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210424525.9A CN102982076B (en) | 2012-10-30 | 2012-10-30 | Based on the various dimensions content mask method in semantic label storehouse |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210424525.9A CN102982076B (en) | 2012-10-30 | 2012-10-30 | Based on the various dimensions content mask method in semantic label storehouse |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102982076A CN102982076A (en) | 2013-03-20 |
| CN102982076B true CN102982076B (en) | 2015-08-19 |
Family
ID=47856096
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210424525.9A Active CN102982076B (en) | 2012-10-30 | 2012-10-30 | Based on the various dimensions content mask method in semantic label storehouse |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102982076B (en) |
Families Citing this family (29)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105095288B (en) * | 2014-05-14 | 2020-02-07 | 腾讯科技（深圳）有限公司 | Data analysis method and data analysis device |
| CN105279208B (en) * | 2014-07-25 | 2019-01-22 | 北京龙源创新信息技术有限公司 | A kind of data marker method and management system |
| CN104484328A (en) * | 2014-10-23 | 2015-04-01 | 安徽教育网络出版有限公司 | Educational resource semantic annotation method |
| CN105630835A (en) * | 2014-11-07 | 2016-06-01 | 中兴通讯股份有限公司 | Content management method and device |
| CN105701139A (en) * | 2015-11-26 | 2016-06-22 | 中国传媒大学 | Holographic video material indexing method |
| CN105787015A (en) * | 2016-02-23 | 2016-07-20 | 浪潮软件集团有限公司 | Method for querying data resources based on labels |
| CN106202292B (en) * | 2016-06-30 | 2022-03-18 | 中国电力科学研究院 | Standard information analysis method based on structured data model |
| CN106354857A (en) * | 2016-09-06 | 2017-01-25 | 中国传媒大学 | News tag management system |
| CN106528633B (en) * | 2016-10-11 | 2019-07-02 | 杭州电子科技大学 | A kind of video society attention rate improvement method recommended based on keyword |
| CN106919695B (en) * | 2017-03-07 | 2018-11-30 | 维沃移动通信有限公司 | A kind of information recommendation method and terminal |
| CN106919711B (en) * | 2017-03-13 | 2020-10-02 | 北京百度网讯科技有限公司 | Method and device for labeling information based on artificial intelligence |
| CN107368529A (en) * | 2017-06-13 | 2017-11-21 | 中国传媒大学 | Documentary film data content feature obtains system and tag library |
| CN107404671A (en) * | 2017-06-13 | 2017-11-28 | 中国传媒大学 | Movie contents feature obtains system and application system |
| CN107347163A (en) * | 2017-06-13 | 2017-11-14 | 中国传媒大学 | A kind of TV play content characteristic obtains system and tag library |
| CN108255907A (en) * | 2017-06-30 | 2018-07-06 | 勤智数码科技股份有限公司 | A kind of catalogue generation method, system and the management method based on the catalogue generation method |
| CN107330069B (en) * | 2017-06-30 | 2020-10-23 | 北京金山安全软件有限公司 | Multimedia data processing method and device, server and storage medium |
| CN107491479B (en) * | 2017-07-05 | 2020-11-24 | 上海大学 | Label management method based on ontology library |
| CN107992513B (en) * | 2017-10-25 | 2021-07-13 | 中兴通讯股份有限公司 | Information processing system and method for realizing information processing |
| CN108205590B (en) * | 2017-12-29 | 2022-01-28 | 北京奇元科技有限公司 | Method and device for establishing network level topological graph of interest points |
| CN108491774B (en) * | 2018-03-12 | 2020-06-26 | 北京地平线机器人技术研发有限公司 | Method and device for tracking and labeling multiple targets in video |
| CN108563642A (en) * | 2018-03-20 | 2018-09-21 | 孙跃 | A kind of Chinese knot-type cultural spreading overseas trade method and system |
| CN109522424B (en) * | 2018-10-16 | 2020-04-24 | 北京达佳互联信息技术有限公司 | Data processing method and device, electronic equipment and storage medium |
| CN109816047B (en) * | 2019-02-19 | 2022-05-24 | 北京达佳互联信息技术有限公司 | Method, device and equipment for providing label and readable storage medium |
| CN110110224A (en) * | 2019-04-16 | 2019-08-09 | 中科金联（北京）科技有限公司 | A kind of data migration method and system based on the multiple label of data |
| CN110457155B (en) * | 2019-07-31 | 2020-09-11 | 清华大学 | Sample class label correction method and device and electronic equipment |
| CN110489649B (en) * | 2019-08-19 | 2023-06-27 | 北京创鑫旅程网络技术有限公司 | Method and device for associating content with tag |
| CN111680219B (en) * | 2020-06-09 | 2023-10-20 | 深圳市雅阅科技有限公司 | Content recommendation method, device, equipment and readable storage medium |
| CN113312493A (en) * | 2021-05-25 | 2021-08-27 | 珠海沃德尔软件科技有限公司 | Label dictionary generating system and method |
| CN115600577B (en) * | 2022-10-21 | 2023-05-23 | 文灵科技(北京)有限公司 | Event segmentation method and system for news manuscript labeling |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102073692A (en) * | 2010-12-16 | 2011-05-25 | 北京农业信息技术研究中心 | Agricultural field ontology library based semantic retrieval system and method |
| CN102521242A (en) * | 2011-11-14 | 2012-06-27 | 江苏联著实业有限公司 | Automatic classification system based on OWL (Ontology of Web Language) ontology analysis |
| CN102622453A (en) * | 2012-04-20 | 2012-08-01 | 北京邮电大学 | Body-based food security event semantic retrieval system |
-
2012
- 2012-10-30 CN CN201210424525.9A patent/CN102982076B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102073692A (en) * | 2010-12-16 | 2011-05-25 | 北京农业信息技术研究中心 | Agricultural field ontology library based semantic retrieval system and method |
| CN102521242A (en) * | 2011-11-14 | 2012-06-27 | 江苏联著实业有限公司 | Automatic classification system based on OWL (Ontology of Web Language) ontology analysis |
| CN102622453A (en) * | 2012-04-20 | 2012-08-01 | 北京邮电大学 | Body-based food security event semantic retrieval system |
Non-Patent Citations (1)
| Title |
|---|
| 光谱信息数据库系统设计与开发;任利华等;《海洋测绘》;20080731;第28卷(第4期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102982076A (en) | 2013-03-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102982076B (en) | Based on the various dimensions content mask method in semantic label storehouse | |
| CN109992645B (en) | Data management system and method based on text data | |
| CN102207948B (en) | Method for generating incident statement sentence material base | |
| CN107609052A (en) | A kind of generation method and device of the domain knowledge collection of illustrative plates based on semantic triangle | |
| CN103049575B (en) | A kind of academic conference search system of topic adaptation | |
| CN104021198B (en) | The relational database information search method and device indexed based on Ontology | |
| CN105608232B (en) | A kind of bug knowledge modeling method based on graphic data base | |
| CN101593200A (en) | Chinese Web page classification method based on the keyword frequency analysis | |
| RU2010107150A (en) | IDENTIFICATION OF SEMANTIC RELATIONS IN INDIRECT SPEECH | |
| CN105718585B (en) | Document and label word justice correlating method and its device | |
| CN102622453A (en) | Body-based food security event semantic retrieval system | |
| CN103440232A (en) | Automatic sScientific paper standardization automatic detecting and editing method | |
| CN101799827A (en) | Video database management method based on layering structure | |
| CN103440233A (en) | Automatic sScientific paper standardization automatic detecting and editing system | |
| CN113190687B (en) | Knowledge graph determining method and device, computer equipment and storage medium | |
| CN106547893A (en) | A kind of photo sort management system and photo sort management method | |
| CN102880721A (en) | Implementation method of vertical search engine | |
| CN103455335A (en) | Multilevel classification Web implementation method | |
| CN104216979A (en) | Chinese technology patent automatic classification system and method for patent classification by using system | |
| Liu et al. | Document processing and retrieval: texpros | |
| CN101957860A (en) | Method and device for releasing and searching information | |
| CN112363996B (en) | Method, system and medium for establishing physical model of power grid knowledge graph | |
| CN103761312B (en) | Information extraction system and method for multi-recording webpage | |
| CN103514214B (en) | Data query method and device | |
| CN102207947A (en) | Direct speech material library generation method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |