CN110413873A - A kind of knowledge network construction method, device and electronic equipment - Google Patents
A kind of knowledge network construction method, device and electronic equipment Download PDFInfo
- Publication number
- CN110413873A CN110413873A CN201910501104.3A CN201910501104A CN110413873A CN 110413873 A CN110413873 A CN 110413873A CN 201910501104 A CN201910501104 A CN 201910501104A CN 110413873 A CN110413873 A CN 110413873A
- Authority
- CN
- China
- Prior art keywords
- subject
- bibliography
- keyword
- knowledge
- data
- 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.)
- Pending
Links
- 238000010276 construction Methods 0.000 title claims abstract description 53
- 238000000034 method Methods 0.000 claims abstract description 36
- 230000015654 memory Effects 0.000 claims abstract description 19
- 230000003542 behavioural effect Effects 0.000 claims description 13
- 238000005516 engineering process Methods 0.000 claims description 12
- 238000004458 analytical method Methods 0.000 claims description 11
- 238000000605 extraction Methods 0.000 claims description 11
- 238000012545 processing Methods 0.000 claims description 7
- 230000011218 segmentation Effects 0.000 claims description 6
- 238000007418 data mining Methods 0.000 claims description 5
- 238000012800 visualization Methods 0.000 claims description 3
- 210000002230 centromere Anatomy 0.000 claims description 2
- 239000000284 extract Substances 0.000 claims description 2
- 241000208340 Araliaceae Species 0.000 claims 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 claims 1
- 235000003140 Panax quinquefolius Nutrition 0.000 claims 1
- 235000008434 ginseng Nutrition 0.000 claims 1
- 238000004590 computer program Methods 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 9
- 239000002245 particle Substances 0.000 description 8
- 238000012986 modification Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 238000012216 screening Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 235000006508 Nelumbo nucifera Nutrition 0.000 description 1
- 240000002853 Nelumbo nucifera Species 0.000 description 1
- 235000006510 Nelumbo pentapetala Nutrition 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000012806 monitoring device Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/20—Education
- G06Q50/205—Education administration or guidance
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Educational Administration (AREA)
- Educational Technology (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- General Business, Economics & Management (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of knowledge network construction method, device and electronic equipments.The knowledge network construction method, comprising: the Bibliographical Information according to involved in network learning procedure obtains corresponding primary bibliography data, carries out subject division to the primary bibliography data;Each section's purpose knowledge node information is determined according to the primary bibliography data;The subject incidence coefficient between the knowledge node incidence coefficient in subject and subject is determined according to the knowledge node information and constructs the knowledge network.The knowledge network construction device includes data acquisition module, knowledge node module, network struction module and info push module.The electronic equipment including memory, processor and stores the computer program that can be run on a memory and on a processor, and the processor realizes the knowledge network construction method when executing described program.
Description
Technical field
The present invention relates to Web education field, a kind of knowledge network construction method, device and electronic equipment are particularly related to.
Background technique
Study is not limited solely in school, but the every aspect being embodied in life, more and more people pass through net
The mode of network study makes great efforts to enrich oneself, and improves respective Specialized Quality.However, people are by various study nets
When station is learnt, usually have no way of doing it in face of the numerous kens of subject.People using study website learn by oneself be usually
Oneself uncomprehending ken, in connection and the subject that new ken cannot easily learn between each subject
The relationship of each knowledge point also can not just formulate reasonable learning planning, this allows for e-learning and gets half the result with twice the effort.
Summary of the invention
In view of this, it is an object of the invention to propose that one kind can carry out knowledge comb to the numerous kens of subject
Reason, knowledge network construction method, device and the electronics formulating the reasonable learning planning of system convenient for people, improving learning efficiency
Equipment.
Based on above-mentioned purpose, the present invention provides a kind of knowledge network construction methods, comprising:
Bibliographical Information involved in acquisition network learning procedure simultaneously obtains corresponding primary bibliography data, to described original
Bibliography data carries out subject division;
According to each subject, the primary bibliography data determine that the knowledge node that each subject is included is believed accordingly
Breath;
Determine that the knowledge node incidence coefficient in subject is associated with system with the subject between subject according to the knowledge node information
Number constructs the knowledge network according to the knowledge node incidence coefficient and the subject incidence coefficient;
According to the behavioral data of student user, the concern subject of the student user is determined, in conjunction with the knowledge network,
E-learning advisory information is generated, Xiang Suoshu student user sends the e-learning advisory information.
Optionally, the Bibliographical Information according to involved in network learning procedure obtains corresponding primary bibliography data, right
The primary bibliography data carry out subject division, comprising:
Acquire the behavioral data of student user;
The Bibliographical Information involved in network science process is determined using data mining technology according to the behavioral data;
According to the Bibliographical Information, the electronic text document of corresponding bibliography is obtained as the primary bibliography data;
Classify to the primary bibliography data, all primary bibliography data are incorporated into corresponding subject.
It is optionally, described that according to each subject, the primary bibliography data determine that each subject is included accordingly
Knowledge node information, comprising:
Keyword extraction is carried out to the corresponding primary bibliography data of bibliography each in the subject, obtains bibliography keyword;
Synonymous near synonym merging is carried out to the bibliography keywords all in the subject, obtains subject keyword;
The weighted value for calculating the subject keyword screens the subject keyword according to the weighted value, obtains
To the knowledge node information.
Optionally, described that keyword extraction is carried out to the corresponding primary bibliography data of bibliography each in the subject, obtain book
Mesh keyword, comprising:
To each bibliography, the primary bibliography data are classified accordingly, obtain catalogue data, title data and text number
According to;
Word segmentation processing is carried out to the catalogue data and the title data using semantic analysis technology, is extracted respectively
To catalogue keyword and title keyword;
De-redundancy operation is carried out to the textual data, using semantic analysis technology to the text number after de-redundancy
According to word segmentation processing is carried out, extraction obtains text keyword, while marking the location information of the text keyword;
Synonymous near synonym are carried out with the text keyword to the catalogue keyword, the title keyword to merge, and are obtained
To initial bibliography keyword;
According to data category and the word frequency in affiliated data category belonging to the initial bibliography keyword, determine
The catalogue word frequency factor, the title word frequency factor and the text word frequency factor of the initial bibliography keyword, according to the initial bibliography
The location information of keyword determines steric factor;
According to the catalogue word frequency factor of the initial bibliography keyword, the title word frequency factor, the positive cliction
The frequency factor and the steric factor calculate the bibliography weight of the initial bibliography keyword:
valuei=α × mfi+β×bfi+λ×zfi×pi
Wherein, valueiIndicate the bibliography weight of the initial bibliography keyword i, mfiIndicate the initial bibliography
The catalogue word frequency factor of keyword i, bfiIndicate the title word frequency factor of the initial bibliography keyword i, zfiTable
Show the text word frequency factor of the initial bibliography keyword i, piIndicate the orientation of the initial bibliography keyword i
The factor;α, β, λ respectively indicate the catalogue word frequency adjustment parameter of the initial bibliography keyword i, heading frequency adjustment parameter with just
Cliction frequency adjustment parameter;
The initial bibliography keyword is screened according to the bibliography weight, obtains the bibliography keyword.
Optionally, the weighted value for calculating the subject keyword, comprising:
According to the subject keyword in difference belonging in bibliography the corresponding bibliography weight determine that the subject is crucial
The weight of word;
The authority factor of the subject keyword is determined according to the authority of the affiliated bibliography of subject keyword;
According to the weight and the authority factor, the weighted value of the subject keyword is calculated:
Wherein, K indicates the weighted value of the subject keyword j, and M expression includes the book of the subject keyword j
Mesh sum, vmIndicate the weight accordingly of the subject keyword j described in m-th of bibliography, qmExpression includes
The authority factor of m-th of bibliography of the subject keyword j,ρ is respectively the weight of the subject keyword j
Adjustment parameter and authoritative adjustment parameter.
Optionally, described that knowledge node incidence coefficient in each subject is determined according to the knowledge node information, packet
It includes:
Represent described subject itself with central node, according to the weighted value calculate the corresponding knowledge node with it is described
The knowledge node incidence coefficient of central node:
βAttr(C, x)=- log2p(x)
Wherein, βAttr(C, x) indicates that the incidence coefficient of the knowledge node x and the central node C, p (x) indicate institute
State the knowledge node x weighted value accordingly.
Optionally, the subject incidence coefficient determined according to the knowledge node information between all subjects, comprising:
Knowledge node described in different section's purposes is compared, if comprising identical in two difference subjects
The knowledge node then illustrates that two differences subject has association;
The subject described presence two differences of association, according to the identical knowledge node respectively in the different subjects
In the corresponding weighted value, calculate subject incidence coefficient described in two differences section's purpose;
Wherein, dis (A, B) indicates that the subject incidence coefficient of subject A and subject B, n indicate subject A and subject B institute
The sum for the identical knowledge node for including, alIndicate first of identical knowledge node corresponding weighted value, b in subject AlTable
Show first of identical knowledge node corresponding weighted value in subject B;
Wherein, ωlIndicate weighting coefficient, the weighting coefficient is defined as:
albl=-log2ωl
Optionally, the knowledge network construction method further include:
According to the knowledge node incidence coefficient and the subject incidence coefficient, draw respectively knowledge node network with
Subject network, by the knowledge node network and the subject network by the knowledge network visualization display.
Based on above-mentioned purpose, the present invention also provides a kind of knowledge network construction devices, comprising:
Data acquisition module obtains corresponding primary bibliography for the Bibliographical Information according to involved in network learning procedure
Data carry out subject division to the primary bibliography data;
Knowledge node module, for the primary bibliography data to determine each subject institute accordingly according to each subject
The knowledge node information for including;
Network struction module, for determined according to the knowledge node information knowledge node incidence coefficient in subject with
Subject incidence coefficient between subject constructs the knowledge according to the knowledge node incidence coefficient and the subject incidence coefficient
Network;
Info push module determines the concern subject of the student user for the behavioral data according to student user,
In conjunction with the knowledge network, e-learning advisory information is generated, Xiang Suoshu student user sends the e-learning recommendation letter
Breath.
Based on above-mentioned purpose, the present invention also provides a kind of electronic equipment, including memory, processor and it is stored in
On reservoir and the computer program that can run on a processor, the processor realize the knowledge knowledge network when executing described program
Network construction method.
From the above it can be seen that a kind of knowledge network construction method, device and electronic equipment provided by the invention,
By obtaining primary bibliography data involved in network learning procedure and analyzing the primary bibliography data, determination is known
Know nodal information, and according to the knowledge node information architecture knowledge network, e-learning advisory information is sent to accordingly
Member.The knowledge network that building obtains can show knowledge point in connection relationship and subject between subject explicitly
Connection relationship, understand the knowledge train of thought in its field of interest clearly convenient for people, enable people to know according to
Know network and make scientific and reasonable learning planning, improves learning efficiency.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, below will to embodiment or
Attached drawing needed to be used in the description of the prior art is briefly described, it should be apparent that, the accompanying drawings in the following description is only
Some embodiments of the present invention, for those of ordinary skill in the art, without creative efforts, also
Other drawings may be obtained according to these drawings without any creative labor.
Fig. 1 is a kind of knowledge network schematic diagram of construction method provided by the embodiment of the present invention;
Fig. 2 is acquisition primary bibliography data in a kind of knowledge network construction method provided by the embodiment of the present invention and draws
Branch purpose method schematic diagram;
Fig. 3 is the side that knowledge node information is determined in a kind of knowledge network construction method provided by the embodiment of the present invention
Method schematic diagram;
Fig. 4 is the method that bibliography keyword is extracted in a kind of knowledge network construction method provided by the embodiment of the present invention
Schematic diagram;
Fig. 5 is the weight that subject keyword is calculated in a kind of knowledge network construction method provided by the embodiment of the present invention
The method schematic diagram of value;
Fig. 6 is a kind of knowledge node network that knowledge network construction method is drawn provided by the embodiment of the present invention;
Fig. 7 is a kind of subject network that knowledge network construction method is drawn provided by the embodiment of the present invention;
Fig. 8 is a kind of knowledge network construction device structural schematic diagram provided by the embodiment of the present invention;
Fig. 9 is a kind of knowledge network building electronic equipment schematic diagram provided by the embodiment of the present invention.
Specific embodiment
To make the objectives, technical solutions, and advantages of the present invention clearer, below in conjunction with specific embodiment, and join
According to attached drawing, the present invention is described in more detail.
It should be noted that all statements for using " first " and " second " are for distinguishing in the embodiment of the present invention
Two non-equal entities of same names or non-equal parameter, it is seen that " first " " second " only for statement convenience, no
It is interpreted as the restriction to the embodiment of the present invention, subsequent embodiment no longer illustrates this one by one.
In one aspect of the invention, a kind of knowledge network construction method is provided.
As shown in Figure 1, a kind of knowledge network construction method provided by some alternative embodiments of the invention, comprising:
S1: Bibliographical Information involved in acquisition network learning procedure simultaneously obtains corresponding primary bibliography data, to the original
Beginning bibliography data carries out subject division;
S2: according to each subject, the primary bibliography data determine knowledge node that each subject is included accordingly
Information;
S3: determine that the knowledge node incidence coefficient in subject and the subject between subject close according to the knowledge node information
Number is contacted, the knowledge network is constructed according to the knowledge node incidence coefficient and the subject incidence coefficient
S4: according to the behavioral data of student user, the concern subject of the student user is determined, in conjunction with the knowledge knowledge network
Network, generates e-learning advisory information, and Xiang Suoshu student user sends the e-learning advisory information.
The knowledge network construction method obtains former according to Bibliographical Information of the institute user involved in network learning procedure
Beginning Bibliographical Information after dividing to the primary bibliography information by subject, determines the knowledge node letter that each subject is included
Breath, using the knowledge node information architecture knowledge network, and accordingly generates e-learning advisory information and is sent to student.Institute
It states knowledge network construction method and characterizes the ken that student is related in network learning procedure with primary bibliography data,
The primary bibliography data comprehensive and accurate can cover all knowledge points in ken, by primary bibliography data
Classification and analysis, extract the knowledge node information in involved ken, the knowledge network thus constructed and life
At e-learning advisory information well-known can reflect the knowledge point train of thought of corresponding ken, thus aspect
Members quickly comprehensively understand corresponding ken, and help formulates scientific and reasonable learning planning, students can be made to learn
Efficiency greatly improves.
As shown in Fig. 2, in a kind of knowledge network construction method provided by some alternative embodiments of the invention, institute
It states Bibliographical Information involved in acquisition network learning procedure and obtains corresponding primary bibliography data, to the primary bibliography number
S1 is divided according to subject is carried out, comprising:
S101: the behavioral data of acquisition student user;
S102: the book involved in network learning procedure is determined using data mining technology according to the behavioral data
Mesh information;
S103: according to the Bibliographical Information, the electronic text document of corresponding bibliography is obtained as the primary bibliography number
According to;
S104: classifying to the primary bibliography data, and all primary bibliography data are incorporated into corresponding section
Mesh.
In the knowledge network construction method, using data mining technology to the behavioral data of collected institute user
Data mining analysis is carried out, so that it is determined that Bibliographical Information involved in network learning procedure, obtains in this manner
The case where Bibliographical Information is comprehensive in further detail, omits without bibliography, the bibliography letter got when determining Bibliographical Information
It ceases more comprehensive, it will be able to obtain more primary bibliography data, characterize student in net with all primary bibliography data
The ken being related in network learning process is properer.
In the knowledge network construction method, the electronic text document of corresponding bibliography is obtained as former according to Bibliographical Information
Beginning bibliography data, the operations such as semantic analysis later can directly carry out on the basis of electronic text document, e-text
The primary bibliography data of document format are also more convenient storage, modification.
It will be apparent to a skilled person that there are largely not accordingly for a subject in real work study
With bibliography, in order to extract the knowledge point of related knowledge domain, with bibliography be minimum analysis object be it is unscientific, often
There is data redudancy height, the ineffective problem of resource utilization, therefore in the knowledge network construction method of the application
In, choosing subject is minimum analytical unit, and the primary bibliography number is incorporated into corresponding subject.
It, can be using " subject classification and code national standard " and " high when classifying to the primary bibliography data
Equal universities and colleges' discipline classification and the criteria for classifying " at least one of be used as classification foundation." subject classification and code national standard " is
According to disciplinary study object, research characteristic, research method, the derivation source of subject, five aspect such as research purpose, target is to
The standard that section is divided is that classification foundation classifies to the initial data with " subject classification and code national standard ",
It may insure that taxo-nomic science classification results are more scientific and system." subject classification and code national standard " object of classification is to learn
Section is different from profession and industry, and " institution of higher learning's discipline classification and the criteria for classifying " is more biased towards in profession and industry, is tied simultaneously
When closing using " subject classification and code national standard " and " institution of higher learning's discipline classification and the criteria for classifying ", profession can be taken into account
Property and professional, so that the finally obtained knowledge network made is more applicable for e-learning.
In assorting process, the International Standard Book Number (ISBN) of corresponding bibliography is determined according to the Bibliographical Information, according to book
Purpose International Standard Book Number can be with the book classification information of the corresponding bibliography of quick obtaining, can in conjunction with the book classification information
Greatly improve the classification effectiveness and accuracy of the primary bibliography data.
As shown in figure 3, in a kind of knowledge network construction method provided by some alternative embodiments of the invention, institute
State that the primary bibliography data determine knowledge node information S2 that each subject is included accordingly according to each subject, packet
It includes:
S201: carrying out keyword extraction to the corresponding primary bibliography data of bibliography each in the subject, obtains bibliography key
Word;
S202: synonymous near synonym merging is carried out to the bibliography keywords all in the subject, obtains subject key
Word;
S203: calculating the weighted value of the subject keyword, is sieved according to the weighted value to the subject keyword
Choosing, obtains the knowledge node information.
In the knowledge network construction method, bibliography is extracted from the corresponding primary bibliography data of each bibliography first
Keyword merges to calculate subject keyword after obtaining subject keyword again to the bibliography keyword of different bibliographys
Corresponding weighted value, finally screening obtains subject knowledge node information.
It will be apparent to a skilled person that the keyword extracted from the primary bibliography data can be used
To state the knowledge point in corresponding bibliography, the significance level of different knowledge points is different, uses institute in the knowledge network construction method
Bibliography weight is stated to be expressed in the significance level of different knowledge points in same bibliography.
One subject includes multiple and different bibliographys, and the bibliography keyword of multiple bibliographys often will appear identical feelings
Condition, or there is the statement difference for the bibliography keyword for representing the same knowledge point, it is directed in the knowledge network construction method
The bibliography keyword is carried out synonymous near synonym merging by afore-mentioned, and the statement of the unified bibliography keyword obtains section
Mesh keyword, in this way can little data redundancy, avoid the same possible resultant error of knowledge point different expression.
In the knowledge network construction method, the weighted value of subject keyword is calculated, according to the weighted value to the section
Mesh keyword is screened, and the subject keyword is to characterize the knowledge point in subject, and the significance level of different knowledge points is not
Together, the subject keyword is screened according to preset weighted value threshold value, by the lower subject keyword of weighted value
That is the lower knowledge point of significance level filters out, final to retain the high subject keyword, that is, high knowledge point of significance level of weighted value.
The weighted value threshold value can be adjusted, to change the screening ratio of subject keyword.
As shown in figure 4, in a kind of knowledge network construction method provided by some alternative embodiments of the invention, institute
It states and keyword extraction is carried out to the corresponding primary bibliography data of bibliography each in the subject, obtain bibliography keyword S201, wrap
It includes:
S2011: to each bibliography, the primary bibliography data are classified accordingly, obtain catalogue data, title data with
Textual data;
S2012: word segmentation processing is carried out to the catalogue data and the title data using semantic analysis technology, respectively
Extraction obtains catalogue keyword and title keyword;
S2013: de-redundancy operation is carried out to the textual data, using semantic analysis technology to described in after de-redundancy
Textual data carries out word segmentation processing, and extraction obtains text keyword, while marking the location information of the text keyword;
S2014: synonymous near synonym are carried out to the catalogue keyword, the title keyword and the text keyword
Merge, obtains initial bibliography keyword;
S2015: according to data category and the word in affiliated data category belonging to the initial bibliography keyword
Frequently, the catalogue word frequency factor, the title word frequency factor and the text word frequency factor for determining the initial bibliography keyword, according to described
The location information of initial bibliography keyword determines steric factor;
S2016: according to the catalogue word frequency factor of the initial bibliography keyword, the title word frequency factor, described
The text word frequency factor and the steric factor calculate the bibliography weight of the initial bibliography keyword:
valuei=α × mfi+β×bfi+λ×zfi×pi
Wherein, valueiIndicate the bibliography weight of the initial bibliography keyword i, mfiIndicate the initial bibliography
The catalogue word frequency factor of keyword i, bfiIndicate the title word frequency factor of the initial bibliography keyword i, zfiTable
Show the text word frequency factor of the initial bibliography keyword i, piIndicate the orientation of the initial bibliography keyword i
The factor;α, β, λ respectively indicate the catalogue word frequency adjustment parameter of the initial bibliography keyword i, heading frequency adjustment parameter with just
Cliction frequency adjustment parameter;
S2017: screening the initial bibliography keyword according to the bibliography weight, and it is crucial to obtain the bibliography
Word.
It is corresponding to bibliography included by the subject using semantic analysis technology in the knowledge network construction method
Primary bibliography data carry out keyword extraction, obtain initial bibliography keyword, then determine corresponding regulatory factor and adjusting respectively
Parameter calculates this bibliography weight of initial bibliography keyword, obtains bibliography keyword by screening.
It will be apparent to a skilled person that same keyword is in variety classes text in primary bibliography data
When in this, significance level is different: under normal circumstances, the significance level of keyword is higher than mark in the title of the key words in catalogue
Significance level, higher than the significance level of keyword in text.In the embodiment of the present application, first by the primary bibliography data
It is divided into catalogue data, title data and textual data, targetedly these three data are handled differently, are used respectively
Catalogue word frequency adjustment parameter, heading frequency adjustment parameter characterize crucial in corresponding kind of class text to text word frequency adjustment parameter
The significance level of word.The importance of the different initial bibliography keywords of calculating that in this manner can be more accurate,
The importance for characterizing the initial bibliography keyword in the embodiment of the present application with the bibliography weight, with according to bibliography weight
The bibliography keyword that filters out indicates the knowledge point in the bibliography, is also more bonded actual conditions.
As shown in figure 5, in a kind of knowledge network construction method provided by some alternative embodiments of the invention, institute
State the weighted value S203 for calculating the subject keyword, comprising:
S2031: according to the subject keyword in difference belonging in bibliography the corresponding bibliography weight determine the section
The weight of mesh keyword;
S2301: according to the authority of the affiliated bibliography of subject keyword determine the subject keyword it is authoritative because
Son;
S2033: according to the weight and the authority factor, the weighted value of the subject keyword is calculated:
Wherein, K indicates the weighted value of the subject keyword j, and M expression includes the book of the subject keyword j
Mesh sum, vmIndicate the weight accordingly of the subject keyword j described in m-th of bibliography, qmExpression includes
The authority factor of m-th of bibliography of the subject keyword j,ρ is respectively the weight of the subject keyword j
Adjustment parameter and authoritative adjustment parameter.
In the knowledge network construction method, institute is determined according to the weight of the subject keyword and authority factor
The weighted value of subject keyword is stated, the weighted value screens to obtain the section subject keyword for after
Purpose knowledge node.
One subject may include multiple bibliographys, and same subject keyword appears in multiple and different bibliographys, In
The bibliography weight has differences subject keyword described in different bibliographys accordingly, these different bibliography weights all can be right
The subject keyword impacts section's purpose importance;It on the other hand, include the identical subject keyword
Different bibliographys authority it is different, bibliography described in the subject keyword it is authoritative also to the subject keyword for
Section's purpose importance impacts.In the embodiment of the present application, comprehensively consider it is above two can be to the subject keyword weight
The factor that the property wanted impacts describes aforementioned two kinds of influence factors with weight adjustment parameter and authoritative adjustment parameter respectively, leads to
It is more scientific accurate to cross the weighted value that such mode determines, but also closing later according to the weighted value to the subject
The result that keyword is screened can more reflect actual conditions.
It is described according in a kind of knowledge network construction method provided by some alternative embodiments of the invention
Knowledge node information determines the knowledge node incidence coefficient in each subject, comprising:
Represent described subject itself with central node, according to the weighted value calculate the corresponding knowledge node with it is described
The knowledge node incidence coefficient of central node:
βAttr(C, x)=- log2p(x)
Wherein, βAttr(C, x) indicates that the incidence coefficient of the knowledge node x and the central node C, p (x) indicate institute
State the knowledge node x weighted value accordingly.
In the knowledge network construction method, according to the weighted value calculation knowledge node incidence coefficient of the knowledge node,
The weighted value is determined according to weight adjustment parameter and authoritative adjustment parameter, to state the important of the corresponding knowledge node
Property, the value of more important knowledge node, the knowledge node incidence coefficient is bigger, the connection of corresponding knowledge node and central node
System is just closer.
It is described according in a kind of knowledge network construction method provided by some alternative embodiments of the invention
Knowledge node information determines the subject incidence coefficient between all subjects, comprising:
Knowledge node described in different section's purposes is compared, if comprising identical in two difference subjects
The knowledge node then illustrates that two differences subject has association;
The subject described presence two differences of association, according to the identical knowledge node respectively in the different subjects
In respective weights value, calculate subject incidence coefficient described in two differences section's purpose;
Wherein, dis (A, B) indicates that the subject incidence coefficient of subject A and subject B, n indicate subject A and subject B institute
The sum for the identical knowledge node for including, alIndicate first of identical knowledge node corresponding weighted value, b in subject AlTable
Show first of identical knowledge node corresponding weighted value in subject B;
Wherein, ωlIndicate weighting coefficient, the weighting coefficient is defined as:
albl=-log2ωl
In the knowledge node construction method, the quantity of identical knowledge node according to present in different subjects and institute
It states knowledge node and calculates the subject incidence coefficient between different subjects in the weighted value in different subjects.In two different subjects
Existing identical knowledge node is more, then illustrates that the two different section's purposes connections are closer, identical knows in addition, described
The the weighted value of knowledge node the big also to illustrate that the two different section's purposes connections are closer, comprehensively considers in the embodiment of the present application
The factor of subject incidence coefficient between above two influence subject, determining subject incidence coefficient is stated in this way
Incidence relation between corresponding subject, the subject incidence coefficient value is bigger, then illustrates that the connection between two subjects is closer.
A kind of knowledge network construction method provided by some alternative embodiments of the invention, further includes:
According to the knowledge node incidence coefficient and the subject incidence coefficient, draw respectively knowledge node network with
Subject network, by the knowledge node network and the subject network by the knowledge network visualization display.
As shown in fig. 6, being the knowledge node network of higher mathematics.In a kind of optional embodiment, centromere is used
Point represents described subject itself, is indicated with the great circle in center;Knowledge node in the subject also uses round spot table
Show, the radius of circle of knowledge node round spot and with central node great-circle distance according to the weighted value of the knowledge node determine, institute
It is bigger to state the more big then round spot radius of weighted value, it is also closer at a distance from central node great circle.It draws and obtains in such a manner
The knowledge node network in, can understand and easily find out the significance levels of all knowledge nodes in subject.
As shown in fig. 7, being the subject network of part subject.Each node indicates a subject, each subject in figure
Node is all connected at least one other subject node, and the wire length between subject node indicates that the connection of corresponding two section purpose is tight
Close degree.
In some alternative embodiments, in order to realize above-mentioned expression effect, the knowledge node network is being drawn
It when figure is with subject network, is specifically drawn using the library JAVAScript of E-CHARTS, using the pass of power guidance layout
It is figure classification, and selects the physical model of coulomb repulsion and Hu Ke elastic force, by each knowledge node and each section in network
Mesh design of node is the entity particle with energy.Such as when drawing knowledge node network, the model of the repulsion is library
Human relations repulsion, i.e., and interparticle distance is from the mechanical model at inverse square, and the weighted value of the knowledge node is denoted as to the electricity of particle
Lotus amount;The model of the gravitation takes Hooke elastic force, length is initialized as to the range index of knowledge node, and use a system
One coefficient of elasticity;In addition, the iteration finite termination in order to make system, is added a fixed retardation coefficient.It is randomly generated and is
The position of each particle in system goes out the stress condition of particle according to physical model calculating and calculates particle within the unit time
The position occurred next time.Until the location variation of particle each in certain primary system is respectively less than a definite value, stop iteration,
And think that system has reached a relatively steady state at this time.The collective effect of particle in system in repulsion and tractive force
Under, since random unordered layout of handling affairs, gradually tend to balance orderly state.The energy of entire physical system simultaneously
It is constantly consuming, after the iteration of limited times, relative displacement no longer occurs between particle, thinks that system reaches at this time
One relatively steady state finally determines a stable knowledge node network as a result,.
Based on appeal purpose, the present invention also provides a kind of knowledge network construction devices.
As shown in figure 8, a kind of knowledge network construction device, feature provided by some alternative embodiments of the invention
It is, comprising:
Data acquisition module 1 obtains corresponding primary bibliography for the Bibliographical Information according to involved in network learning procedure
Data carry out subject division to the primary bibliography data;
Knowledge node module 2, for the primary bibliography data to determine each subject institute accordingly according to each subject
The knowledge node information for including, the knowledge node information include knowledge node and corresponding weighted value;
Network struction module 3, for determined according to the knowledge node information knowledge node incidence coefficient in subject with
Subject incidence coefficient between subject constructs the knowledge according to the knowledge node incidence coefficient and the subject incidence coefficient
Network;
Info push module 4 determines the concern subject of the student user for the behavioral data according to student user,
In conjunction with the knowledge network, e-learning advisory information is generated, Xiang Suoshu student user sends the e-learning recommendation letter
Breath.
The device of above-described embodiment has corresponding method real for realizing method corresponding in previous embodiment
The beneficial effect of example is applied, details are not described herein.
Based on above-mentioned purpose, the present invention also provides a kind of electronic equipments for executing the knowledge network construction method.
As shown in figure 9, the electronic equipment includes:
One or more processors 501 and memory 502, in Fig. 9 by taking a processor 501 as an example.
The electronic equipment for executing the knowledge network construction method can also include: input unit 503 and output dress
Set 503.
Processor 501, memory 502, input unit 503 and output device 503 can pass through bus or other modes
It connects, in Fig. 9 for being connected by bus.
Memory 502 is used as a kind of non-volatile computer readable storage medium storing program for executing, can be used for storing non-volatile software journey
Sequence, non-volatile computer executable program and module, such as the knowledge network construction method pair in the embodiment of the present application
Program instruction/the module answered.Processor 501 is by running the non-volatile software program being stored in memory 502, instruction
And module realizes the knowledge of above method embodiment thereby executing the various function application and data processing of server
Network establishing method.
Memory 502 may include storing program area and storage data area, wherein storing program area can store operation system
Application program required for system, at least one function;Storage data area can be stored according to the execution knowledge network construction method
Device use created data etc..In addition, memory 502 may include high-speed random access memory, can also wrap
Include nonvolatile memory, for example, at least a disk memory, flush memory device or other non-volatile solid state memories
Part.In some embodiments, it includes the memory remotely located relative to processor 501 that memory 502 is optional, these are long-range
Memory can pass through network connection to member user's behavior monitoring device.The example of above-mentioned network includes but is not limited to interconnect
Net, intranet, local area network, mobile radio communication and combinations thereof.
Input unit 503 can receive the number or character information of input, and generates and construct with the execution knowledge network
The related key signals input of the user setting and function control of square law device.Output device 503 may include the display such as display screen
Equipment.
One or more of modules are stored in the memory 502, when by one or more of processors
When 501 execution, the knowledge network construction method in above-mentioned any means embodiment is executed.It is described to execute the knowledge network structure
The embodiment of the device of construction method, technical effect are same or similar with aforementioned any means embodiment.
It should be understood by those ordinary skilled in the art that: the discussion of any of the above embodiment is exemplary only, not
It is intended to imply that the scope of the present disclosure (including claim) is limited to these examples;Under thinking of the invention, above embodiments
Or can also be combined between the technical characteristic in different embodiments, step can be realized with random order, and be existed such as
Many other variations of the upper different aspect of the invention, for simplicity, they are not provided in details.
In addition, to simplify explanation and discussing, and in order not to obscure the invention, in provided attached drawing
It can show or can not show and be connect with the well known power ground of integrated circuit (IC) chip and other components.In addition,
Device can be shown in block diagram form, to avoid obscuring the invention, and this has also contemplated following facts, i.e.,
The details of embodiment about these block diagram arrangements be height depend on will implementing platform of the invention (that is, these are thin
Section should be completely within the scope of the understanding of those skilled in the art).Detail (for example, circuit) is being elaborated to describe
In the case where exemplary embodiment of the present invention, it will be apparent to those skilled in the art that can there is no this
Implement the present invention in the case where a little details or in the case that these details change.Therefore, these descriptions should be by
It is considered illustrative rather than restrictive.
Although having been incorporated with specific embodiments of the present invention, invention has been described, according to retouching for front
It states, many replacements of these embodiments, modifications and variations will be apparent for those of ordinary skills.Example
Such as, discussed embodiment can be used in other memory architectures (for example, dynamic ram (DRAM)).
What the embodiment of the present invention was intended to cover fall within the broad range of appended claims all such replaces
It changes, modifications and variations.Therefore, all within the spirits and principles of the present invention, any omission for being made, modification, equivalent replacement,
Improve etc., it should all be included in the protection scope of the present invention.
Claims (10)
1. a kind of knowledge network construction method characterized by comprising
Bibliographical Information involved in acquisition network learning procedure simultaneously obtains corresponding primary bibliography data, to the primary bibliography number
According to progress subject division;
According to each subject, the primary bibliography data determine knowledge node information that each subject is included accordingly;
The subject incidence coefficient between the knowledge node incidence coefficient in subject and subject, root are determined according to the knowledge node information
The knowledge network is constructed according to the knowledge node incidence coefficient and the subject incidence coefficient;
According to the behavioral data of student user, the concern subject of the student user is determined, in conjunction with the knowledge network, generate net
Network learns advisory information, and Xiang Suoshu student user sends the e-learning advisory information.
2. the method according to claim 1, wherein Bibliographical Information involved in the acquisition network learning procedure
And corresponding primary bibliography data are obtained, subject division is carried out to the primary bibliography data, comprising:
Acquire the behavioral data of student user;
The Bibliographical Information involved in network learning procedure is determined using data mining technology according to the behavioral data;
According to the Bibliographical Information, the electronic text document of corresponding bibliography is obtained as the primary bibliography data;
Classify to the primary bibliography data, all primary bibliography data are incorporated into corresponding subject.
3. the method according to claim 1, wherein described according to each subject primary bibliography data accordingly
Determine the knowledge node information that each subject is included, comprising:
Keyword extraction is carried out to the corresponding primary bibliography data of bibliography each in the subject, obtains bibliography keyword;
Synonymous near synonym merging is carried out to the bibliography keywords all in the subject, obtains subject keyword;
The weighted value for calculating the subject keyword screens the subject keyword according to the weighted value, obtains institute
State knowledge node information.
4. according to the method described in claim 3, it is characterized in that, described to the corresponding primary bibliography of bibliography each in the subject
Data carry out keyword extraction, obtain bibliography keyword, comprising:
To each bibliography, the primary bibliography data are classified accordingly, obtain catalogue data, title data and textual data;
Word segmentation processing is carried out to the catalogue data and the title data using semantic analysis technology, extracts obtain catalogue respectively
Keyword and title keyword;
De-redundancy operation is carried out to the textual data, the textual data after de-redundancy is carried out using semantic analysis technology
Word segmentation processing, extraction obtains text keyword, while marking the location information of the text keyword;
Synonymous near synonym are carried out with the text keyword to the catalogue keyword, the title keyword to merge, and are obtained just
Beginning bibliography keyword;
According to data category and the word frequency in affiliated data category belonging to the initial bibliography keyword, determine described first
The catalogue word frequency factor, the title word frequency factor and the text word frequency factor of beginning bibliography keyword, according to the initial bibliography keyword
Location information determine steric factor;
According to the catalogue word frequency factor, the title word frequency factor, the text word frequency of the initial bibliography keyword because
The sub and steric factor calculates the bibliography weight of the initial bibliography keyword:
valuei=α × mfi+β×bfi+λ×zfi×pi
Wherein, valueiIndicate the bibliography weight of the initial bibliography keyword i, mfiIndicate the initial bibliography keyword
The catalogue word frequency factor of i, bfiIndicate the title word frequency factor of the initial bibliography keyword i, zfiIndicate described first
The text word frequency factor of beginning bibliography keyword i, piIndicate the steric factor of the initial bibliography keyword i;α,β,λ
The catalogue word frequency adjustment parameter, heading frequency adjustment parameter and text word frequency for respectively indicating the initial bibliography keyword i are adjusted
Parameter;
The initial bibliography keyword is screened according to the bibliography weight, obtains the bibliography keyword.
5. according to the method described in claim 4, it is characterized in that, the weighted value for calculating the subject keyword, comprising:
According to the subject keyword in difference belonging in bibliography the corresponding bibliography weight determine the subject keyword
Weight;
The authority factor of the subject keyword is determined according to the authority of the affiliated bibliography of subject keyword;
According to the weight and the authority factor, the weighted value of the subject keyword is calculated:
Wherein, K indicates the weighted value of the subject keyword j, and M expression includes that the bibliography of the subject keyword j is total
Number, vmIndicate the weight accordingly of the subject keyword j described in m-th of bibliography, qmExpression includes the section
The authority factor of m-th of bibliography of mesh keyword j,ρ is respectively that the weight of the subject keyword j adjusts ginseng
Several and authoritative adjustment parameter.
6. according to the method described in claim 3, it is characterized in that, described determine each subject according to the knowledge node information
Interior knowledge node incidence coefficient, comprising:
Described subject itself is represented with central node, the corresponding knowledge node and the centromere are calculated according to the weighted value
The knowledge node incidence coefficient of point:
βAttr(C, x)=- log2 p(x)
Wherein, βAttr(C, x) indicates that the incidence coefficient of the knowledge node x and the central node C, p (x) indicate the knowledge
The node x weighted value accordingly.
7. the method according to claim 1, wherein described determine all subjects according to the knowledge node information
Between subject incidence coefficient, comprising:
Knowledge node described in different section's purposes is compared, if described knowing in two difference subjects comprising identical
Know node, then illustrates that two differences subject has association;
There is two differences of association subject, according to the identical knowledge node phase in the different subjects respectively
Weighted value is answered, subject incidence coefficient described in two differences section's purpose is calculated;
Wherein, dis (A, B) indicates the subject incidence coefficient of subject A and subject B, and n indicates subject A and subject B is included
The sum of identical knowledge node, alIndicate first of identical knowledge node corresponding weighted value, b in subject AlIt indicates first
Identical knowledge node corresponding weighted value in subject B;
Wherein, ωlIndicate weighting coefficient, the weighting coefficient is defined as:
albl=-log2ωl
8. the method according to claim 1, wherein further include:
According to the knowledge node incidence coefficient and the subject incidence coefficient, knowledge node network and subject net are drawn respectively
Network figure, by the knowledge node network and the subject network by the knowledge network visualization display.
9. a kind of knowledge network construction device characterized by comprising
Data acquisition module obtains corresponding primary bibliography data for the Bibliographical Information according to involved in network learning procedure,
Subject division is carried out to the primary bibliography data;
Knowledge node module, for the primary bibliography data to determine that each subject is included accordingly according to each subject
Knowledge node information;
Network struction module, for being determined between the knowledge node incidence coefficient in subject and subject according to the knowledge node information
Subject incidence coefficient, the knowledge network is constructed according to the knowledge node incidence coefficient and the subject incidence coefficient;
Info push module determines the concern subject of the student user, in conjunction with institute for the behavioral data according to student user
Knowledge network is stated, e-learning advisory information is generated, Xiang Suoshu student user sends the e-learning advisory information.
10. a kind of electronic equipment including memory, processor and stores the calculating that can be run on a memory and on a processor
Machine program, which is characterized in that the processor realizes the side as described in claim 1 to 8 any one when executing described program
Method.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910501104.3A CN110413873A (en) | 2019-06-11 | 2019-06-11 | A kind of knowledge network construction method, device and electronic equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910501104.3A CN110413873A (en) | 2019-06-11 | 2019-06-11 | A kind of knowledge network construction method, device and electronic equipment |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110413873A true CN110413873A (en) | 2019-11-05 |
Family
ID=68358958
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910501104.3A Pending CN110413873A (en) | 2019-06-11 | 2019-06-11 | A kind of knowledge network construction method, device and electronic equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110413873A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115186674A (en) * | 2022-06-20 | 2022-10-14 | 成都飞机工业(集团)有限责任公司 | Aviation failure case management method, device, equipment and storage medium |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105205180A (en) * | 2015-10-27 | 2015-12-30 | 无锡天脉聚源传媒科技有限公司 | Knowledge map evaluation method and device |
CN108052672A (en) * | 2017-12-29 | 2018-05-18 | 北京师范大学 | Promote structural knowledge map construction system and method using group study behavior |
CN108629497A (en) * | 2018-04-25 | 2018-10-09 | 北京比特智学科技有限公司 | Course content Grasping level evaluation method and device |
US20180349511A1 (en) * | 2017-06-06 | 2018-12-06 | Diffeo, Inc. | Knowledge operating system |
CN109829059A (en) * | 2019-01-18 | 2019-05-31 | 平安科技(深圳)有限公司 | Recommend method, apparatus, equipment and the storage medium of knowledge point |
-
2019
- 2019-06-11 CN CN201910501104.3A patent/CN110413873A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105205180A (en) * | 2015-10-27 | 2015-12-30 | 无锡天脉聚源传媒科技有限公司 | Knowledge map evaluation method and device |
US20180349511A1 (en) * | 2017-06-06 | 2018-12-06 | Diffeo, Inc. | Knowledge operating system |
CN108052672A (en) * | 2017-12-29 | 2018-05-18 | 北京师范大学 | Promote structural knowledge map construction system and method using group study behavior |
CN108629497A (en) * | 2018-04-25 | 2018-10-09 | 北京比特智学科技有限公司 | Course content Grasping level evaluation method and device |
CN109829059A (en) * | 2019-01-18 | 2019-05-31 | 平安科技(深圳)有限公司 | Recommend method, apparatus, equipment and the storage medium of knowledge point |
Non-Patent Citations (1)
Title |
---|
张亚龙: "学科知识的可视化技术研究与实现", 《中国优秀博硕士学位论文全文数据库(硕士)信息科技辑》 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115186674A (en) * | 2022-06-20 | 2022-10-14 | 成都飞机工业(集团)有限责任公司 | Aviation failure case management method, device, equipment and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Kadoić et al. | A new method for strategic decision-making in higher education | |
Asanbe et al. | Teachers’ performance evaluation in higher educational institution using data mining technique | |
CN106528656A (en) | Student history and real-time learning state parameter-based course recommendation realization method and system | |
CN106384319A (en) | Teaching resource personalized recommending method based on forgetting curve | |
CN109360459A (en) | A kind of Training Management method, Training Management device and electronic equipment | |
CN113918806A (en) | Method for automatically recommending training courses and related equipment | |
CN110263181A (en) | The method for digging of the structure of knowledge and the planing method of learning path | |
Al-Sarem | Building a decision tree model for academic advising affairs based on the algorithm C 4-5 | |
CN105260963A (en) | Subject accomplishment evaluation system | |
Langrall | The status of probability in the elementary and lower secondary school mathematics curriculum: The rise and fall of probability in school mathematics in the United States | |
CN110413873A (en) | A kind of knowledge network construction method, device and electronic equipment | |
Siebert-Evenstone et al. | Cause and because: Using epistemic network analysis to model causality in the next generation science standards | |
Chaudhary et al. | Student future prediction using machine learning | |
Fernandes et al. | Educational data mining: Discovery standards of academic performance by students in public high schools in the federal district of brazil | |
CN108550019A (en) | A kind of resume selection method and device | |
CN105608067A (en) | Automatic knowledge extraction method and apparatus for network teaching system | |
AMIN et al. | Group decision support system model to determine prospective participants for lecturer strengthening activities | |
CN109977197A (en) | Electronic exercise processing method, device, equipment and storage medium | |
CN106780225A (en) | A kind of cloud educational system and educational data output intent | |
Devi et al. | Impact of socio-economic factors on students’ academic performance: A case study of Jawahar Navodaya Vidyalaya | |
Reichel et al. | Analysis of student behavior in virtual learning environment depending on student assessments | |
CN111090751A (en) | Teaching recommendation method, system, storage medium and terminal based on knowledge graph | |
Shan | Computer-based outdoor sport sustainable development using wavelet neural network | |
CN109447865A (en) | Learning content recommendation method and system | |
Hawari | Integration of soft systems methodology and system dynamics modelling for supply and demand analysis of the rising cost in higher education |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191105 |
|
RJ01 | Rejection of invention patent application after publication |