CN104462207A - Multi-piecemeal learning resource labeling method for distributed learning environment - Google Patents

Multi-piecemeal learning resource labeling method for distributed learning environment Download PDF

Info

Publication number
CN104462207A
CN104462207A CN201410610071.3A CN201410610071A CN104462207A CN 104462207 A CN104462207 A CN 104462207A CN 201410610071 A CN201410610071 A CN 201410610071A CN 104462207 A CN104462207 A CN 104462207A
Authority
CN
China
Prior art keywords
resource
fragment
education
segment
learning
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
Application number
CN201410610071.3A
Other languages
Chinese (zh)
Other versions
CN104462207B (en
Inventor
袁柳
陈健
王凡
段俊杰
陈慧君
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shaanxi Normal University
Original Assignee
Shaanxi Normal University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shaanxi Normal University filed Critical Shaanxi Normal University
Priority to CN201410610071.3A priority Critical patent/CN104462207B/en
Publication of CN104462207A publication Critical patent/CN104462207A/en
Application granted granted Critical
Publication of CN104462207B publication Critical patent/CN104462207B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00Handling natural language data
    • G06F40/10Text processing
    • G06F40/166Editing, e.g. inserting or deleting
    • G06F40/169Annotation, e.g. comment data or footnotes
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/40Information retrieval; Database structures therefor; File system structures therefor of multimedia data, e.g. slideshows comprising image and additional audio data
    • G06F16/48Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually

Abstract

The invention relates to a multi-piecemeal learning resource labeling method for a distributed learning environment. The method mainly solves the problem that due to the fact that the piecemeal of electronic learning resources cannot be labeled in an existing learning resource labeling technology, a learner cannot accurately find or locate the needed learning resources. The method comprises the steps that (1) a resource piecemeal body is defined; (2) other bodies capable of being used for resource piecemeal labeling are determined; (3) the learning resource and the learning resource piecemeal are labeled; (4) the labeled learning resource is issued. According to the method, the learning resources of different types can be described in a medium mode, meanwhile, a body and a field body are described by combining learning content, the electronic learning resource is labeled in the expression mode and the learning content together, and the method has the advantages that the labeling result is accurate and standard, accessing is easy, and the search pertinency of the learning source can be effectively improved, and the method can be used for the field of electronic learning resource management, Web semantic retrieval and the like.

Description

The multiple clips education resource mask method of Based on Distributed academic environment
Technical field
The invention belongs to web data administrative skill field, specifically based on a semantic tagger for education resource fragment body, realize the multiple clips education resource mask method of the Based on Distributed academic environment of the semantic tagger to all kinds education resource and education resource Partial Fragment.
Background technology
The prosperity that is universal and community network application of electric learning platform application makes Web to have accumulated increasing education resource, and these digitized education resources have different types, as text, video, audio frequency, image etc.Conveniently learner finds and uses these resources, and publish resource person generally can add label to education resource.By retrieval service such as search engines, the education resource using the resource tag matched with searching keyword to mark can be found.Can learner find the education resource that meets oneself demand and mutually exchange on Web, and the form of resource semantic tagger and access mode serve critical effect.Under many circumstances, the fragment of complete resource that what learner wished to obtain is not and just one or more resource.Such as, learner is only concerned about the partial page of certain lantern slide, or only needs the explanation of the specific a few minutes of seeing in an instructional video.In this case, if return to the complete resource of learner one can not meet its requirement completely.Therefore, must be at many levels multi-faceted to the mark of education resource, to improve the accuracy of education resource mark, strengthen the accessibility of education resource simultaneously.Existing following three kinds are mainly divided into the technology that education resource carries out semantic tagger.
The first, the education resource based on community network label marks.
Hend S.Al-Khalifa and Hugh C.Davis proposes the method using Folksonomy and tag set mark education resource in paper " FolksAnnotation:ASemantic Metadata Tool for Annotating Learning Resources Using Folksonomiesand Domain Ontologies " (" Proceedings of the Second International Conferenceon Innovations in Information Technology " 2006,1:5).The core procedure of the method comprises: (1) is extracted label and to be gone forward side by side column criterion from tag database; (2) produce semantic metadata to mark education resource.This is also the step usually all followed based on the semanteme marking method of label.Although label easy understand, the major defect of these class methods is the lack of standardization of label vocabulary, can cause the problems such as the semanteme of mark is unclear, ambiguity.These class methods are also only limitted to mark complete education resource simultaneously, do not support the mark to resource segment.
The second, based on the education resource semantic tagger of study body.
Relative to label, the semanteme of body form support, specification, the education resource mask method based on body can overcome label semantically Problems existing.It has the step similar with the method based on label: (1) selects the body that can be used for marking; (2) body vocabulary is used to mark.The selection of study body is the core of these class methods.Yang Xianmin and Yu Shengquan is at paper " the education resource information model under ubiquitous academic environment builds " (" Chinese audio-visual education " 2010,24 (9): 72-78) existing study body is systematically analyzed in the ability on education resource that describes with not enough, and propose study metamessage model, relatively comprehensively can express the characteristic of education resource and learner's various aspects, but still not to the ability that education resource fragment is described.Owing to lacking the semantic metadata of resource segment being carried out to specification description, therefore the current education resource mask method based on body can not meet the demand that learner only pays close attention to part education resource.
The third, the education resource based on associated data describes and issues.
The people such as Stefan Dietze and Honq Qing Yu are at paper " Linked Education:interlinkingeducational Resources and the Web of Data " (" In Proceedings of SAC " 2012, propose 366:372) and use the form of associated data (Linked Data) to be undertaken interconnected by education resource, thus provide the convenient way of a kind of education resource access.The education resource of associated data form contains some semantic informations that domain body provides, and can conduct interviews to support distributed electric learning platform to resource by URI, SPARQL language can also be used to inquire about resource simultaneously.But the granularity of inquiry is also the resource for using URI mark, and under the condition of education resource fragment not being carried out to URI mark, the education resource tissue of associated data form can not meet the demand of learner's query learning resource completely.
Summary of the invention
The object of the invention is to overcome the deficiency existing for above-mentioned technology, propose a kind of based on education resource fragment body, to realize the semantic tagger according to learner's demand, the education resource in Web being carried out to different grain size rank, thus improve the service efficiency of education resource, the multiple clips education resource mask method of the Based on Distributed academic environment freely exchanged under support the environment of social network in learning process.
To achieve these goals, the technical solution adopted in the present invention is made up of following steps:
(1) resource segment body is defined
Using the least unit of education resource fragment as mark, according to the medium type of carrying education resource, education resource is divided into continuous type media resource and discrete type media resource, if continuous type media resource, definition time fragment in its animation comprised, Voice & Video resource, and time slice is divided into time point fragment and time interval fragment; If discrete type media resource, on the image resource that it comprises definition space fragment, in non-structured text resource, define destructuring paragraph fragment, definition structure fragment in structured text resource, and be structuralized query fragment, path segments and resource description framework tlv triple query fragment by structuring fragment Further Division; Description resource segment structure is remitted as notional word with time point fragment defined above, time interval fragment, space fragment, destructuring paragraph fragment, structuralized query fragment, path segments, resource description framework tlv triple query fragment; The display effect describing resource segment is remitted as notional word by the resource of above-mentioned animation, audio frequency, video, image and non-structured text resource, structured text resource;
(2) other bodies that can be used for resource segment mark are determined
For technical standard and the body in existing multimedia, e-learning field, retrieval wherein can be used for the body describing education resource fragment medium information characteristic and education resource content characteristic, is described this education resource fragment medium information characteristic and education resource content characteristic with concept vocabulary wherein;
(3) mark of education resource and education resource fragment
User selects marking types according to object content, for complete education resource, with remitting mark for the notional word describing education resource fragment medium information characteristic and education resource content characteristic in step (2), support User Defined vocabulary mark simultaneously; For user's determined reusable resource segment according to demand, the structure of description resource segment in step (1) and the concept vocabulary of display effect is used to state resource segment, and use in step (2) and remit mark for the notional word describing education resource fragment medium information characteristic and education resource content characteristic, support User Defined vocabulary mark simultaneously;
(4) issue of the education resource after mark
According to URI naming method, the global learning resource marked in step (3) and resource segment are named, the annotation results of step (3) is converted into resource description framework tlv triple data model, the linking relationship of owl:sameAs form is set up between mutual related URI, complete the issue marking education resource, specifically:
(4.1) according to URI naming method, the global learning resource marked in step (3) and resource segment are named;
(4.2) for each named education resource or resource segment s, determine to have marked vocabulary respectively and can act on attribute p on resource segment s iand corresponding value o i, generate resource description framework tlv triple description collections { (s, p i, o i);
(4.3) the resource description framework tlv triple description collections { (s, the p that will all education resources and resource segment be generated i, o i) in any two compare, calculate its related coefficient;
(4.4) similarity threshold is set, if resource is to (s i, s j) related coefficient be more than or equal to similarity threshold, use owl:sameAs to set up linking relationship between the two, if resource is to (s i, s j) related coefficient be less than similarity threshold, then abandon establishing the link relation.
In above-mentioned steps (4.3), any two resources compare, and calculate its related coefficient, specifically:
(4.3.1) first set up " attribute-value " model of resource description framework data collection, the resource that resource description framework data is concentrated is represented by a characteristic set rfs, and each feature in rfs associates a value set, rfs k={ f 1..., f n, for each define its value set Vf simultaneously i={ vf 1..., vf k;
(4.3.2) the similarity sim_v of computational resource describing framework data centralization source attribute values,
(4.3.3) the similarity sim_f between computation attribute,
sim _ f ( f 1 , f 2 ) = Σ s = 1 k max ( sim _ v ( v s , v t ) ) ∀ t ∈ [ 1 , h ] k , Wherein v s∈ Vf 1, v t∈ Vf 2, k=|Vf 1|, h=|Vf 2|;
(4.3.4) similarity between computational resource is sim_s, namely
sim _ s ( rfs 1 , rfs 2 ) = Σ i = 1 n Σ j = 1 m sim _ f ( f i , f j ) | f i = f j n , Wherein, f i∈ rfs 1, f j∈ rfs 2, n=|rfs 1|, m=|rfs 2|.
The multiple clips education resource mask method of Based on Distributed academic environment of the present invention is the resource segment body for describing education resource fragment, concept in this body is for describing the fragment of all types of education resources that may have access to, resource segment body is combined with domain body, body vocabulary is used to carry out semantic tagger to education resource, use HTTP URI to identify to complete resource and valuable resource segment in the process of mark simultaneously, the resource that all URI of HTTP name is redefined with the form of RDF, realize being interconnected between education resource.Compared with the prior art it mainly has the following advantages:
1, because the present invention designs and define education resource fragment Description Ontology, structure according to multimedia resource defines its fragment, support mark the interested resource segment of learner and again present, overcome can only be coarse in prior art the deficiency that complete education resource is marked, make the present invention have semantic tagger accurately, be convenient to the advantage that learner exchanges and discuss.
2, because the present invention adopts the tlv triple pattern of resource description framework to represent annotation results, the form of associated data is used to name resource and resource segment, support that http protocol is to the access being marked resource simultaneously, therefore can well support sharing of education resource under distributed environment, also make the present invention while realizing semantic tagger, also have the ability integrated education resource.
3, because the present invention adopts the strategy using body vocabulary and User Defined label to mark resource simultaneously, enforceable constraint is not done to the mark behavior of user while, specification clear at guarantee markup information semanteme, make the present invention have the simple advantage of application.
4, the present invention propose resource description framework tlv triple data characteristics model and based on this model resource between correlativity evaluation algorithm can extend to the calculating of correlativity between resource in any field.
Accompanying drawing explanation
Fig. 1 is the mask method process flow diagram of embodiment 1;
Fig. 2 is the key concept figure of resource segment body;
Fig. 3 is the semantic tagger schematic diagram of image segments;
Fig. 4 is the semantic tagger schematic diagram of video segment;
Fig. 5 is for mark with hypertext education resource fragment PPT education resource fragment and associate schematic diagram;
Fig. 6 is the correlativity explanation of two education resources;
Fig. 7 is that the education resource of different platform integrates schematic diagram;
Embodiment
Below in conjunction with drawings and Examples, technical scheme of the present invention is described further, but the present invention is not limited only to following enforcement situation.
With reference to Fig. 1, the multiple clips education resource mask method of the Based on Distributed academic environment of the present embodiment is realized by following steps:
(1) design and define resource segment body
Using the least unit of education resource fragment as mark, according to the medium type of carrying education resource, education resource media are divided into two classes: continuous media and discrete type media.Continuous media mainly comprises time-based media, as animation, Voice & Video; Discrete type media mainly comprise image, text (structuring and destructuring) etc.The concept of education resource fragment is described based on this class definition, be respectively space fragment, time slice, structured text fragment, non-structured text fragment, with reference to accompanying drawing 2, wherein by " being a kind of " relationship between the related concept of tool, this contact is represented with the line segment of hollow arrow in Fig. 2, the time slice that such as time interval fragment " being a kind of " is special, the structuring fragment that path segments " being a kind of " is special.
According to common education resource type, definition can describe the concept vocabulary of education resource fragment display effect, the expression of resource segment is divided into the ImageRepresentation for Description Image fragment, for stating the AnimationRepresentation of animation fragment, for stating the AudioRepresentation of audio fragment, for describing the VideoRepresentation of video segment, for stating the StructuredTextRepresentation of structured text information, for stating the PlainTextRepresentation of non-structured text information.
For describing the concept vocabulary of resource segment architectural feature in education resource fragment Description Ontology, specifically:
Whole multiple clips resource description body is divided into three abstract concepts: resource segment (ResourceFragment), resource segment represent (FragmentRepresentation) and resource segment set (FragmentSet), the main concept that resource segment relates to comprises: space fragment (SpatialFragment) is mainly described the region in two dimensional image, uses the bottom visual signature comprising color, position, size and shape etc. to describe each region; Time slice (TemporalFragment) is mainly described continuous type media fragment, concrete is divided into again time point fragment (TimePointFragment), hour, minute, second is used to describe the concrete moment, and time interval (IntervalFragment) fragment, each time interval comprises an initial time and an end time; Structuring fragment (StructuredFragment) is mainly described the data with ad hoc structure or pattern, concrete is divided into again structuralized query fragment (SQLFragmentation) for describing the Query Result fragment in relational database, path segments (XPathFragment) is for describing the minor structure in XML file, and resource description framework tlv triple query fragment (SPARQLFragment) is for resource description framework (RDF) data slot; Destructuring fragment (PlainTextFragment) is mainly described the content presented in plain text mode.
Therefore, description resource segment structure is remitted with above-mentioned time point fragment, time interval fragment, space fragment, destructuring paragraph fragment, structuralized query fragment, path segments, resource description framework tlv triple query fragment as notional word; The display effect describing resource segment is remitted as notional word by the resource of above-mentioned animation, audio frequency, video, image and non-structured text resource, structured text resource.
(2) other bodies that can be used for resource segment mark are determined
For technical standard and the body in existing multimedia, e-learning field, retrieval wherein can be used for the body describing education resource fragment medium information characteristic and education resource content characteristic, with concept vocabulary wherein, this education resource fragment medium information characteristic and education resource content characteristic are described, realize standardization and the versatility of resource description.Concrete concept is such as: as the MPEG-7 (Moving Picture Experts Group) of Multimedia Content Description Interface; For describing body COMM (Core Ontology for Multimedia), the M3O (Multimedia metadata ontology) and Media Resource Ontology of multimedia object; For the body DIG35 of Description Image, for describing the body SAPO (Shape Acquisition and Processing Ontology) of shape and image-region, for describing body VDO (Visual Descriptor Ontology) and the VRA core 3 (VisualResource Association) of visual object; For the body Music Ontology of description audio, Kanzaki MusicVocabulary and Music Recommendation Ontology; For describing IEEE LOM (Learning Object Meta-data), SCORM (Sharable Content ObjectReference Model), IMS-LD and the IMS Common Cartridge of education resource content.
(3) mark of education resource and education resource fragment
User selects marking types according to oneself interested object content, for complete education resource, then use and remit mark for the notional word describing education resource fragment medium information characteristic and education resource content characteristic in step (2), support User Defined vocabulary mark simultaneously; And for user's determined reusable resource segment according to demand, the structure of the description resource segment in step (1) and the concept vocabulary of display effect is then used to state resource segment, and use in step (2) and remit mark for the notional word describing education resource fragment medium information characteristic and education resource content characteristic, support User Defined vocabulary mark simultaneously; User can also mark in more detail resource segment further on the basis marked complete education resource.
With reference to accompanying drawing 3, to the explanation of the image segments semantic tagger that the present invention proposes.For the image that title is " MVC.jpg ", wherein subregion can be extracted mark, this image segments belongs to " space fragment ", its position in original image can be represented with " region of right 50% ", theme " model view represents structure " can be added to this fragment, and this image segments is expressed as " MVCfragment.jpg " again.
With reference to accompanying drawing 4, to the explanation of the video segment semantic tagger that the present invention proposes.For the video telling about " binary tree " structure that URI is " http://example.com/btree.mp4 ", Partial Fragment wherein can be extracted, this fragment belongs to " time slice ", can be described by " time period " structure, start time time period of this fragment is " 40 points and 25 seconds ", end time is " 52 points and 15 seconds ", theming as " balanced binary tree " of this video segment, URI name is called " http://example.com/bbtree.mp4 ", can again be expressed as video file " balanced binary tree .mp4 " to this fragment.
(4) issue of the education resource after mark
For the education resource and the resource segment that add semantic tagger, URI is used to name it, and ensure to be conducted interviews to it by http protocol, resource description framework data model is used to redescribe marked education resource and resource segment, annotation results by above-mentioned global learning resource and resource segment is converted into resource description framework tlv triple data model, between mutual related URI, set up the linking relationship of owl:sameAs form, namely complete the issue having marked education resource.Specifically:
(4.1) according to URI naming method, the above-mentioned global learning resource that marked and resource segment are named, ensure to be conducted interviews to it by http protocol;
(4.2) for each named education resource or resource segment s, determine to have marked vocabulary respectively and can act on attribute pi on resource segment s and corresponding value oi thereof, generate resource description framework tlv triple description collections { (s, pi, oi) };
(4.3) tlv triple for two education resources and education resource fragment that belong to identical knowledge category is described as (s respectively i, p i, o i) and (s j, p j, o j), calculate the similarity between them, concrete computing method are as follows:
(4.3.1) the attribute-value model of RDF data set is first set up.The resource of a RDF data centralization can be represented by a characteristic set rfs (RDF features set), and each feature in rfsi associates a value set.Rfs can be defined as follows:
Rfs k={ f 1..., f n, for each define its value set Vf simultaneously i={ vf 1..., vf k.
(4.3.2) on this basis, the similarity of definable Resource Properties and value thereof.The similarity sim_v of property value can be defined as intuitively:
(4.3.3) on property value Similarity measures basis, the Similarity measures between attribute may be defined as:
sim _ f ( f 1 , f 2 ) = Σ s = 1 k max ( sim _ v ( v s , v t ) ) ∀ t ∈ [ 1 , h ] k , Wherein v s∈ Vf 1, v t∈ Vf 2, k=|Vf 1|, h=|Vf 2|.
(4.4.4) calculate on basis at attribute similarity, the Similarity measures between resource may be defined as:
sim _ s ( rfs 1 , rfs 2 ) = Σ i = 1 n Σ j = 1 m sim _ f ( f i , f j ) | f i = f j n , Wherein, f i∈ rfs 1, f j∈ rfs 2, n=|rfs 1|, m=|rfs 2|.
(4.4) set similarity threshold, if the related coefficient of resource to (si, sj) is less than similarity threshold, then abandon establishing the link relation, if resource is to (s i, s j) related coefficient be more than or equal to similarity threshold, use owl:sameAs to set up linking relationship between the two.
Suppose there are three education resource mark platforms being in diverse location, all provide the function of education resource fragment being carried out to semantic tagger, and the education resource belonged in identical platform can be interlinked by method of the present invention, effect as shown in Figure 5, to the mark of PPT education resource fragment and hypertext education resource fragment and the explanation that is associated, specific as follows:
URI name is called that the PPT resource " soft project introduction chapter 1 1201 " of " http://example.com/sc1201 " is a fragment belonging to " resource description framework tlv triple inquiry port ", this fragment can by inquiry " SELECT? slide WHERE{? slide a#Slide? slide dc:subject ' 1201URI ' " obtain, and be added theme " soft project, software development model ", the author of this PPT resource segment is " the soft work group of Shan Normal University ", language is " Chinese ", and publisher is " Shaanxi School of Computer Science of Normal University ".URI name is called that " http://linkedlearningresource.com/edition/1.0 " is a HTML hypertext resource, path segments " soft project introduction chapters and sections " is wherein a structured text resource segment, by XPath path "/HTML/BODY/P{59} ", it is positioned, this fragment has title " http://example2.com/seintro ", belong to hypertext format equally, and there is theme " soft project ", keyword " soft project, system development ", its language is " English ", author is " associated data subsets of resources ".Resource segment " http://example.com/sc1201 " illustrates the more deep explanation in detail of resource segment " http://example2.com/seintro ", therefore can establish the link relation " explanation " between these two resources.
Make can mutually access between the close education resource fragment of theme or field by this association.
Between the education resource of different platform, education resource more detailed on attribute can be found, further calculating relative coefficient therebetween, use as Fig. 6 the method that adopts, correlativity is greater than to two resources of set similarity threshold, its owl:SameAs can be set up associate, therefore can belong to the resource associations of different platform, realize the integration of education resource.The principle integrated different learning platform education resource is see accompanying drawing 7.Specific as follows:
See accompanying drawing 6, to the explanation of correlation calculations between two education resources.For the resource segment " frequent mode and association rule mining " that type is PDF document, its URI title is designated as " http://example.com/fpmining.pdf ", to the result that resource segment content marks be: this resource segment has theme " frequent mode, correlation rule ", affiliated field is " data management, database application technology, machine learning, large data analysis, data mining, data prediction, trend analysis ", and resource publisher is " Shaanxi School of Computer Science of Normal University ".The RDF form of this content annotation results is described below:
< " http://example.com/fpmining.pdf " theme " frequent mode " >
< " http://example.com/fpmining.pdf " theme " correlation rule " >
< " http://example.com/fpmining.pdf " field " data management " >
< " http://example.com/fpmining.pdf " field " database application technology " >
< " http://example.com/fpmining.pdf " field " machine learning " >
< " http://example.com/fpmining.pdf " field " large data analysis " >
< " http://example.com/fpmining.pdf " field " data mining " >
< " http://example.com/fpmining.pdf " field " data prediction " >
< " http://example.com/fpmining.pdf " field " trend prediction " >
< " http://example.com/fpmining.pdf " publisher " Shaanxi School of Computer Science of Normal University " >
< " http://example.com/fpmining.pdf " type " PDF " >
For the resource segment " associated data excavation " that type is " Word document ", belong to non-structured text fragment, its URI name is called " http://example2.com/ar.doc ", to the result that resource segment content marks be: this resource segment has theme " correlation rule ", affiliated field is " machine learning, large data analysis, data mining ", and resource publisher is " data mining group, Shaanxi School of Computer Science of Normal University ".The RDF form of this content annotation results is described below:
< " http://example2.com/ar.doc " theme " correlation rule " >
< " http://example2.com/ar.doc " field " machine learning " >
< " http://example2.com/ar.doc " field " large data analysis " >
< " http://example2.com/ar.doc " field " data mining " >
< " http://example2.com/ar.doc " type " Word " >
< " http://example2.com/ar.doc " publisher " data mining group " >
< " http://example2.com/ar.doc " publisher " Shaanxi School of Computer Science of Normal University " >
According to the RDF description form of resource segment " http://example.com/fpmining.pdf " and " http://example2.com/ar.doc ", be translated into the description of rfs form respectively.The rfs form of resource " http://example.com/fpmining.pdf " is:
Rfs pdf={ theme, field, type, publisher }
Vf theme={ frequent mode, correlation rule }
Vf field={ data management, database application technology, machine learning, large data analysis, data mining, data prediction, trend analysis }
Vf type={ PDF}
Vf publisher={ Shaanxi School of Computer Science of Normal University }
The rfs form of resource " http://example2.com/ar.doc " is:
Rfs dOC={ theme, field, type, publisher }
Vf theme={ correlation rule }
Vf field={ machine learning, large data analysis, data mining }
Vf type={ Word document }
Vf publisher={ data mining group, Shaanxi School of Computer Science of Normal University }
According to rfs model attributes value, can calculate " http://example.com/fpmining.pdf " and " http://example2.com/ar.doc " correlation values respectively on each attribute, occurrence is as follows:
On the basis that each attribute similarity calculates, the correlativity between these two resource segment is:
sim _ s ( PDF , DOC ) = 1 + 1 + 0 + 1 4 = 0.75
Suppose that relevance threshold be the intersegmental relative coefficient of 0.7, PDF fragment and DOC sheet is 0.75, therefore can think that these two resource segment can use owl:sameAs to associate.

Claims (3)

1. a multiple clips education resource mask method for Based on Distributed academic environment, is characterized in that being made up of following steps:
(1) resource segment body is defined
Using the least unit of education resource fragment as mark, according to the medium type of carrying education resource, education resource is divided into continuous type media resource and discrete type media resource, if continuous type media resource, definition time fragment in its animation comprised, Voice & Video resource, and time slice is divided into time point fragment and time interval fragment; If discrete type media resource, on the image resource that it comprises definition space fragment, in non-structured text resource, define destructuring paragraph fragment, definition structure fragment in structured text resource, and be structuralized query fragment, path segments and resource description framework tlv triple query fragment by structuring fragment Further Division; Description resource segment structure is remitted as notional word with time point fragment defined above, time interval fragment, space fragment, destructuring paragraph fragment, structuralized query fragment, path segments, resource description framework tlv triple query fragment; The display effect describing resource segment is remitted as notional word by the resource of above-mentioned animation, audio frequency, video, image and non-structured text resource, structured text resource;
(2) other bodies that can be used for resource segment mark are determined
For technical standard and the body in existing multimedia, e-learning field, retrieval wherein can be used for the body describing education resource fragment medium information characteristic and education resource content characteristic, is described this education resource fragment medium information characteristic and education resource content characteristic with concept vocabulary wherein;
(3) mark of education resource and education resource fragment
User selects marking types according to object content, for complete education resource, with remitting mark for the notional word describing education resource fragment medium information characteristic and education resource content characteristic in step (2), support User Defined vocabulary mark simultaneously; For user's determined reusable resource segment according to demand, the structure of description resource segment in step (1) and the concept vocabulary of display effect is used to state resource segment, and use in step (2) and remit mark for the notional word describing education resource fragment medium information characteristic and education resource content characteristic, support User Defined vocabulary mark simultaneously;
(4) issue of the education resource after mark
According to URI naming method, the global learning resource marked in step (3) and resource segment are named, the annotation results of step (3) is converted into resource description framework tlv triple data model, between mutual related URI, set up the linking relationship of owl:sameAs form, complete the issue marking education resource.
2. the multiple clips education resource mask method of Based on Distributed academic environment according to claim 1, is characterized in that described step (4) specifically:
(4.1) according to URI naming method, the global learning resource marked in step (3) and resource segment are named;
(4.2) for each named education resource or resource segment s, determine to have marked vocabulary respectively and can act on attribute p on resource segment s iand corresponding value o i, generate resource description framework tlv triple description collections { (s, p i, o i);
(4.3) the resource description framework tlv triple description collections { (s, the p that will all education resources and resource segment be generated i, o i) in any two compare, calculate its related coefficient;
(4.4) similarity threshold is set, if resource is to (s i, s j) related coefficient be more than or equal to similarity threshold, use owl:sameAs to set up linking relationship between the two, if resource is to (s i, s j) related coefficient be less than similarity threshold, then abandon establishing the link relation.
3. the multiple clips education resource mask method of Based on Distributed academic environment according to claim 1, is characterized in that any two resources compare in step (4.3), calculate its related coefficient, specifically:
(4.3.1) first set up " attribute-value " model of resource description framework data collection, the resource that resource description framework data is concentrated is represented by a characteristic set rfs, and each feature in rfs associates a value set, rfs k={ f 1..., f n, for each define its value set Vf simultaneously i={ vf 1..., vf k;
(4.3.2) the similarity sim_v of computational resource describing framework data centralization source attribute values,
(4.3.3) the similarity sim_f between computation attribute,
sim _ f ( f 1 , f 2 ) = &Sigma; s = 1 k max ( sim _ v ( v s , v t ) ) &ForAll; t &Element; [ 1 , h ] k , Wherein v s∈ Vf 1, v t∈ Vf 2, k=|Vf 1|, h=|Vf 2|;
(4.3.4) similarity between computational resource is sim_s, namely
sim _ s ( rfs 1 , rfs 2 ) = &Sigma; i = 1 n &Sigma; j = 1 m sim _ f ( f i , f j ) | f i = f j n , Wherein, f i∈ rfs 1, f j∈ rfs 2, n=|rfs 1|, m=|rfs 2|.
CN201410610071.3A 2014-11-03 2014-11-03 The multiple clips education resource mask method of Based on Distributed academic environment Expired - Fee Related CN104462207B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201410610071.3A CN104462207B (en) 2014-11-03 2014-11-03 The multiple clips education resource mask method of Based on Distributed academic environment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201410610071.3A CN104462207B (en) 2014-11-03 2014-11-03 The multiple clips education resource mask method of Based on Distributed academic environment

Publications (2)

Publication Number Publication Date
CN104462207A true CN104462207A (en) 2015-03-25
CN104462207B CN104462207B (en) 2017-07-11

Family

ID=52908243

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201410610071.3A Expired - Fee Related CN104462207B (en) 2014-11-03 2014-11-03 The multiple clips education resource mask method of Based on Distributed academic environment

Country Status (1)

Country Link
CN (1) CN104462207B (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105117130A (en) * 2015-08-28 2015-12-02 上海第九城市教育科技股份有限公司 Selective interaction system and method of multimedia courseware
CN105786943A (en) * 2015-12-08 2016-07-20 北京师范大学 Method and device for generating digital object
WO2017084238A1 (en) * 2015-11-20 2017-05-26 Chrysus Intellectual Properties Limited A method and system for analyzing a piece of text
CN110381382A (en) * 2019-07-23 2019-10-25 腾讯科技(深圳)有限公司 Video takes down notes generation method, device, storage medium and computer equipment

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101504657A (en) * 2009-03-20 2009-08-12 武汉大学 Semantic Web service oriented resource search method
CN102663010A (en) * 2012-03-20 2012-09-12 复旦大学 Personalized image browsing and recommending method based on labelling semantics and system thereof
CN103699572A (en) * 2013-11-26 2014-04-02 北京航空航天大学 Digital media content and resource integration and sharing method in cloud environment

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101504657A (en) * 2009-03-20 2009-08-12 武汉大学 Semantic Web service oriented resource search method
CN102663010A (en) * 2012-03-20 2012-09-12 复旦大学 Personalized image browsing and recommending method based on labelling semantics and system thereof
CN103699572A (en) * 2013-11-26 2014-04-02 北京航空航天大学 Digital media content and resource integration and sharing method in cloud environment

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
MIGUEL-ANGEL SICILIA 等: "Navigating learning resources through linked data: a preliminary report on the re-design of Organic.Edunet", 《PROCEEDINGS OF LINKED LEARNING 2011: THE 1ST INTERNATIONAL WORKSHOP ON ELEARNING APPROACHES FOR THE LINKED DATA AGE》 *

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105117130A (en) * 2015-08-28 2015-12-02 上海第九城市教育科技股份有限公司 Selective interaction system and method of multimedia courseware
WO2017084238A1 (en) * 2015-11-20 2017-05-26 Chrysus Intellectual Properties Limited A method and system for analyzing a piece of text
US10755594B2 (en) 2015-11-20 2020-08-25 Chrysus Intellectual Properties Limited Method and system for analyzing a piece of text
CN105786943A (en) * 2015-12-08 2016-07-20 北京师范大学 Method and device for generating digital object
CN105786943B (en) * 2015-12-08 2019-07-12 北京师范大学 A kind of method and device that digital object generates
CN110381382A (en) * 2019-07-23 2019-10-25 腾讯科技(深圳)有限公司 Video takes down notes generation method, device, storage medium and computer equipment
CN110381382B (en) * 2019-07-23 2021-02-09 腾讯科技(深圳)有限公司 Video note generation method and device, storage medium and computer equipment

Also Published As

Publication number Publication date
CN104462207B (en) 2017-07-11

Similar Documents

Publication Publication Date Title
Hyvönen Semantic portals for cultural heritage
KR20130060720A (en) Apparatus and method for interpreting service goal for goal-driven semantic service discovery
CN104462207A (en) Multi-piecemeal learning resource labeling method for distributed learning environment
Bedi et al. Focused crawling of tagged web resources using ontology
Guo et al. SOR: An optimized semantic ontology retrieval algorithm for heterogeneous multimedia big data
KR101079769B1 (en) Semantic Search Method and System for Associating with Plurality of Classifications
Heese et al. One Click Annotation.
Yadagiri et al. Semantic web and the libraries: an overview
Dhingra et al. Towards intelligent information retrieval on web
Limongelli et al. Enrichment of the dataset of joint educational entities with the web of data
Bechhofer et al. Hello Cleveland! Linked Data publication of live music archives
Yaman et al. Interlinking scigraph and dbpedia datasets using link discovery and named entity recognition techniques
Stegmaier et al. Unleashing semantics of research data
Siochos et al. Developing a formal model for mind maps
Ramulu et al. A study of semantic web mining: Integrating domain knowledge into web mining
Chakraborty et al. Linked open data representation of historical heritage of Bangladesh
Shazmeen et al. Semantic Web Mining: Benefits, Challenges and Opportunities
Yu et al. Data service generation framework from heterogeneous printed forms using semantic link discovery
Jenice Aroma et al. A Survey on need for Semantic web
Huiying et al. Ontology-based enterprise content retrieval method
Škoda et al. Explainable similarity of datasets using knowledge graph
Wu The semantic retrieval system for learning resources based on subject knowledge ontology
Greenberg Metadata and digital information [ELIS classic]
Lachake et al. A survey on personalized search: an web Information retrieval system
Said et al. Interlinking video with DBpedia using knowledge-based Word Sense Disambiguation algorithms.

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
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: 20170711

Termination date: 20191103