CN104915388B - It is a kind of that method is recommended based on spectral clustering and the book labels of mass-rent technology - Google Patents
It is a kind of that method is recommended based on spectral clustering and the book labels of mass-rent technology Download PDFInfo
- Publication number
- CN104915388B CN104915388B CN201510270676.7A CN201510270676A CN104915388B CN 104915388 B CN104915388 B CN 104915388B CN 201510270676 A CN201510270676 A CN 201510270676A CN 104915388 B CN104915388 B CN 104915388B
- Authority
- CN
- China
- Prior art keywords
- term
- user
- cluster
- mrow
- msub
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000005516 engineering process Methods 0.000 title claims abstract description 21
- 230000003595 spectral effect Effects 0.000 title claims abstract description 21
- 238000000034 method Methods 0.000 title claims abstract description 18
- 238000005457 optimization Methods 0.000 claims abstract description 6
- 230000002045 lasting effect Effects 0.000 claims abstract description 5
- 239000011159 matrix material Substances 0.000 claims description 32
- 230000000694 effects Effects 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 3
- 230000000739 chaotic effect Effects 0.000 claims description 3
- 238000000354 decomposition reaction Methods 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims description 3
- 238000007689 inspection Methods 0.000 claims 1
- 230000015572 biosynthetic process Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 239000002360 explosive 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
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
Method is recommended based on spectral clustering and the book labels of mass-rent technology the invention discloses a kind of, this method is applied to digital library system, Laplacian matrixes are built by using the retrieval click logs of user, and term is clustered using spectral clustering, afterwards by using mass-rent technology, lasting optimization is carried out to the result of cluster, finally the result of optimization is applied in commending system.The present invention lifts the degree of accuracy of term clustering, so as to improve accuracy of the system in terms of label recommendations using the term of user as label by the combination of spectral clustering and mass-rent technology.
Description
Technical field
The invention belongs to the book labels recommended technology based on spectral clustering and mass-rent technology, be related to it is a kind of based on spectral clustering and
The book labels of mass-rent technology recommend method.
Background technology
With being on the increase for internet information, explosive growth is presented in information, and information, which is rationally efficiently sorted out, to be turned into
The key that information effectively utilizes.Traditional classifying method is mainly carried out by artificial mode, and on the premise of magnanimity information,
The mode that such information is sorted out is hard to carry on, thus has engendered the novel information classifying mode using label as core,
And have become the key of the Internet, applications.Among digital library system, label essentially from book information, while with
During family uses system, the term of user, books index information can also be added among system as a kind of label,
And the relation of furthered using label as tie user and books, lifting user have found the efficiency of books.
Meanwhile got growing concern for for the application of mass data, commending system.User obtains the mode of information
The vertical retrieval of domain knowledge is retrieved by full network type information, then it is continuous to current commending system, the acquisition speed of information
Accelerate, it is increasingly notable for the information personalized continuous improvement of different user, contribution of the commending system in terms of system availability.
Clustering algorithm is the key method of data mining, and clustering algorithm is used for realizing the cluster to article, user in commending system, and
Run by the iteration of algorithm to optimize the effect of cluster.
The content of the invention
It is an object of the invention to the deficiency utilized for existing commending system to term, there is provided one kind is used for digitized map
Method is recommended based on spectral clustering and the book labels of mass-rent technology on book shop.
The purpose of the present invention is achieved through the following technical solutions：A kind of books based on spectral clustering and mass-rent technology
Label recommendation method, comprise the following steps：
(1) the retrieval data and retrieval click data of user are filtered out from result collection system or Web daily records；
(2) using the retrieval data and retrieval click data of user, term-books matrix is built, according to term-figure
Book matrix obtains the Laplacian matrixes of term-term；
(3) cluster operation is carried out to Laplacian matrixes using spectral clustering, obtains the cluster result of term；
(4) lasting optimization is carried out to the cluster result obtained by step 3 using mass-rent technology；
(5) cluster result after the past retrieval record of user is optimized with step 4 is mapped, and utilizes gathering after mapping
Class formation is as label recommendations to user.
Further, the step 2 is specially：The retrieval set of words Q of all users is obtained from the retrieval data of user
={ q1,q2,…,qn, wherein n is the sum of term, and q is independent retrieval word；Examined from the retrieval click data of user
The books set B={ b that rope word is clicked on1,b2,…,bm, wherein m is the sum for clicking on books, and b is independent books；According to institute
The books set B for having the retrieval set of words Q of user and term to click on obtains term-books matrix M, for term-figure
Book matrix M each single item, is defined as follows：
Wherein IijFor the corresponding relation of this book of i-th of term and jth；For each books, if multiple retrievals
There is click behavior in word, then contact between these terms be present, built according to the contact between term to this this book
Term-term matrix D, for each single item of term-term matrix D, if contact between two terms be present
It is then 1, is otherwise 0；It is placed in by the value that each column element of term-term matrix D is added to obtain on diagonal, its
Its position is set to 0, so as to form new matrix W；Laplacian matrix Ls are obtained by formula L=D-W.
Further, the step 3 is specially：For spectral clustering, selected object function RatioCut is：
Wherein k be cluster number, AiIth cluster result is represented, | Ai| represent the term in ith cluster result
Quantity,Represent to remove AiOutside other cluster result set,Represent ith cluster result and other clusters
As a result weight sum,Calculation formula beWherein W (a, b) is cluster
As a result a and cluster result b weight；Released according to the property of Laplacian matrix Ls and minimize object function RatioCut equivalences
In minimizing Laplacian matrixes, the dimensionality reduction to Laplacian matrixes is realized thereby using the method for SVD matrix decompositions, is used
K-mean clustering algorithms complete the cluster operation to the Laplacian matrixes after dimensionality reduction.
Further, the step 4 is specially：In the cluster result for the term that step 3 is obtained corresponding to term
Selected user of the user as mass-rent, selected user is sent to by way of sending mail by the result of cluster, selectes user
Feedback be defined as：
Wherein, Query represents a term, and positive feedback represents that user thinks that the term meets place cluster result
Theme, negative-feedback represent that user thinks that the term does not meet the theme of cluster result, and the zero feedback representation term is difficult to sentence
It is disconnected whether to meet theme；Feedback information according to selected user to a cluster result, the cluster result is carried out following three kinds
The processing of different modes：
(a) feedback information for selecting user shows that the cluster can be very good to show some theme, and it is embodied in
Two aspects：On the one hand it is that negative-feedback result is less than positive feedback result, is on the other hand that the feedback information of user is not present each other
The situation of contradiction；In this case, the negative-feedback in cluster result is deleted, retains the term of positive feedback and zero feedback；
(b) feedback information for selecting user is chaotic, it is difficult to shows the quality of the Clustering Effect, it is embodied in several use
Family is different to the feedback information of identical term or even opposite；In the case of this kind, it is meant that the feedback information of current selected user
Still it is not enough to judge the cluster, thus needs to introduce new user, again mass-rent task distribution operation；
(c) feedback information for selecting user shows that the cluster does not have clear and definite theme, is embodied in selected user
Feedback in term more than 50% feedback information it is different or opposite；In this case, directly the cluster result is deleted
Remove.
The beneficial effects of the invention are as follows：This method is clustered using retrieval word information of the spectral clustering to user, and is used
The lasting optimization of result of the mass-rent technology to cluster, it is final to realize the effect that book labels recommendation is improved using term.This hair
It is bright on the basis of cluster result, it is proposed that the purpose optimized to cluster result is realized by using mass-rent technology, lead to
The feedback information for collecting multiple users to cluster result is crossed to judge and optimize the result of cluster, and the result of cluster is applied to
Among commending system.
Brief description of the drawings
Fig. 1 is the flow chart that the present invention recommends method based on spectral clustering and the book labels of mass-rent technology.
Embodiment
The present invention is described in further detail below in conjunction with the accompanying drawings.
As shown in figure 1, the present invention is a kind of to recommend method, including following step based on spectral clustering and the book labels of mass-rent technology
Suddenly：
(1) the retrieval data and retrieval click data of user are filtered out from result collection system or Web daily records；
(2) using the retrieval data and retrieval click data of user, term-books matrix is built, according to term-figure
Book matrix obtains the Laplacian matrixes of term-term；Specially：All users are obtained from the retrieval data of user
Retrieval set of words Q={ q1,q2,...,qn, wherein n is the sum of term, and q is independent retrieval word；From the Access Points of user
Hit in data and obtain the books set B={ b of term click1,b2,...,bm, wherein m is the sum for clicking on books, and b is only
Vertical books；Term-books matrix is obtained according to the retrieval set of words Q and term of all users the books set B clicked on
M, for term-books matrix M each single item, it is defined as follows：
Wherein IijFor the corresponding relation of this book of i-th of term and jth；For each books, if multiple retrievals
There is click behavior in word, then contact between these terms be present, built according to the contact between term to this this book
Term-term matrix D, for each single item of term-term matrix D, if contact between two terms be present
It is then 1, is otherwise 0；It is placed in by the value that each column element of term-term matrix D is added to obtain on diagonal, its
Its position is set to 0, so as to form new matrix W；Laplacian matrix Ls are obtained by formula L=D-W.
(3) cluster operation is carried out to Laplacian matrixes using spectral clustering, obtains the cluster result of term；Specially：
For spectral clustering, selected object function RatioCut is：
Wherein k be cluster number, AiIth cluster result is represented, | Ai| represent the term in ith cluster result
Quantity,Represent to remove AiOutside other cluster result set,Represent ith cluster result and other clusters
As a result weight sum,Calculation formula beWherein W (a, b) is cluster
As a result a and cluster result b weight；Released according to the property of Laplacian matrix Ls and minimize object function RatioCut equivalences
In minimizing Laplacian matrixes, the dimensionality reduction to Laplacian matrixes is realized thereby using the method for SVD matrix decompositions, is used
K-mean clustering algorithms complete the cluster operation to the Laplacian matrixes after dimensionality reduction.
(4) lasting optimization is carried out to the cluster result obtained by step 3 using mass-rent technology；Specially：Step 3 is obtained
To term cluster result in selected user of the user corresponding to term as mass-rent, by way of sending mail will
The result of cluster is sent to selected user, and the feedback of selected user is defined as：
Wherein, Query represents a term, and positive feedback represents that user thinks that the term meets place cluster result
Theme, negative-feedback represent that user thinks that the term does not meet the theme of cluster result, and the zero feedback representation term is difficult to sentence
It is disconnected whether to meet theme；Feedback information according to selected user to a cluster result, the cluster result is carried out following three kinds
The processing of different modes：
(a) feedback information for selecting user shows that the cluster can be very good to show some theme, and it is embodied in
Two aspects：On the one hand it is that negative-feedback result is less than positive feedback result, is on the other hand that the feedback information of user is not present each other
The situation of contradiction；In this case, the negative-feedback in cluster result is deleted, retains the term of positive feedback and zero feedback；
(b) feedback information for selecting user is chaotic, it is difficult to shows the quality of the Clustering Effect, it is embodied in several use
Family is different to the feedback information of identical term or even opposite；In the case of this kind, it is meant that the feedback information of current selected user
Still it is not enough to judge the cluster, thus needs to introduce new user, again mass-rent task distribution operation；
(c) feedback information for selecting user shows that the cluster does not have clear and definite theme, is embodied in selected user
Feedback in term more than 50% feedback information it is different or opposite；In this case, directly the cluster result is deleted
Remove.
(5) cluster result after the past retrieval record of user is optimized with step 4 is mapped, and utilizes gathering after mapping
Class formation is as label recommendations to user.
Claims (3)
1. a kind of recommend method based on spectral clustering and the book labels of mass-rent technology, it is characterised in that comprises the following steps：
(1) the retrieval data and retrieval click data of user are filtered out from result collection system or Web daily records；
(2) using the retrieval data and retrieval click data of user, term-books matrix is built, according to term-books square
Battle array obtains the Laplacian matrixes of term-term；Specially：The inspection of all users is obtained from the retrieval data of user
Rope set of words Q={ q1,q2,…,qn, wherein n is the sum of term, and q is independent retrieval word；From the retrieval hits of user
The books set B={ b of term click are obtained in1,b2,…,bm, wherein m is the sum for clicking on books, and b is independent figure
Book；Term-books matrix M is obtained according to the retrieval set of words Q and term of all users the books set B clicked on, for
Term-books matrix M each single item, is defined as follows：
Wherein IijFor the corresponding relation of this book of i-th of term and jth；For each books, if multiple terms are equal
Click behavior to this this book be present, then contact between these terms be present, retrieval is built according to the contact between term
Word-term matrix D, for each single item of term-term matrix D, it is if it contact be present between two terms
1, it is otherwise 0；It is placed in by the value that each column element of term-term matrix D is added to obtain on diagonal, Qi Tawei
Install as 0, so as to form new matrix W；Laplacian matrix Ls are obtained by formula L=D-W；
(3) cluster operation is carried out to Laplacian matrixes using spectral clustering, obtains the cluster result of term；
(4) lasting optimization is carried out to the cluster result obtained by step 3 using mass-rent technology；
(5) cluster result after the past retrieval record of user is optimized with step 4 is mapped, and utilizes the cluster knot after mapping
Structure is as label recommendations to user.
2. a kind of according to claim 1 recommend method based on spectral clustering and the book labels of mass-rent technology, it is characterised in that
The step 3 is specially：For spectral clustering, selected object function RatioCut is：
<mrow>
<mi>R</mi>
<mi>a</mi>
<mi>t</mi>
<mi>i</mi>
<mi>o</mi>
<mi>C</mi>
<mi>u</mi>
<mi>t</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>A</mi>
<mi>i</mi>
</msub>
<mo>,</mo>
<mn>...</mn>
<mo>,</mo>
<msub>
<mi>A</mi>
<mi>k</mi>
</msub>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mfrac>
<mn>1</mn>
<mn>2</mn>
</mfrac>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>k</mi>
</munderover>
<mfrac>
<mrow>
<mi>W</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>A</mi>
<mi>i</mi>
</msub>
<mo>,</mo>
<mover>
<msub>
<mi>A</mi>
<mi>i</mi>
</msub>
<mo>&OverBar;</mo>
</mover>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<mo>|</mo>
<msub>
<mi>A</mi>
<mi>i</mi>
</msub>
<mo>|</mo>
</mrow>
</mfrac>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>k</mi>
</munderover>
<mfrac>
<mrow>
<mi>c</mi>
<mi>u</mi>
<mi>t</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>A</mi>
<mi>i</mi>
</msub>
<mo>,</mo>
<mover>
<msub>
<mi>A</mi>
<mi>i</mi>
</msub>
<mo>&OverBar;</mo>
</mover>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<mo>|</mo>
<msub>
<mi>A</mi>
<mi>i</mi>
</msub>
<mo>|</mo>
</mrow>
</mfrac>
</mrow>
Wherein k be cluster number, AiIth cluster result is represented, | Ai| the term quantity in ith cluster result is represented,Represent to remove AiOutside other cluster result set,Represent ith cluster result and other cluster results
Weight sum,Calculation formula beWherein W (a, b) be cluster result a with
Cluster result b weight；Minimum object function RatioCut is released according to the property of Laplacian matrix Ls and is equivalent to minimum
Change Laplacian matrixes, realize the dimensionality reduction to Laplacian matrixes thereby using the method for SVD matrix decompositions, use K-mean
Clustering algorithm completes the cluster operation to the Laplacian matrixes after dimensionality reduction.
3. a kind of according to claim 1 recommend method based on spectral clustering and the book labels of mass-rent technology, it is characterised in that
The step 4 is specially：User corresponding to term is as the selected of mass-rent in the cluster result for the term that step 3 is obtained
User, the result of cluster is sent to selected user by way of sending mail, the feedback of selected user is defined as：
Wherein, Query represents a term, and positive feedback represents that user thinks that the term meets the master of place cluster result
Topic, negative-feedback represent that user thinks that the term does not meet the theme of cluster result, and the zero feedback representation term is difficult to judge
Whether theme is met；Feedback information according to selected user to a cluster result, following three kinds are carried out to the cluster result not
With the processing of mode：
(a) feedback information for selecting user shows that the cluster can be very good to show some theme, and it is embodied in two
Aspect：On the one hand it is that negative-feedback result is less than positive feedback result, is on the other hand that contradiction each other is not present in the feedback information of user
Situation；In this case, the negative-feedback in cluster result is deleted, retains the term of positive feedback and zero feedback；
(b) feedback information for selecting user is chaotic, it is difficult to shows the quality of the Clustering Effect, it is embodied in several users couple
The feedback information of identical term is different or even opposite；In the case of this kind, it is meant that the feedback information of current selected user is still not
It is enough to judge the cluster, thus needs to introduce new user, again mass-rent task distribution operation；
(c) feedback information for selecting user shows that the cluster does not have clear and definite theme, is embodied in the anti-of selected user
The feedback information of term in feedback more than 50% is different or opposite；In this case, directly the cluster result is deleted.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510270676.7A CN104915388B (en) | 2015-03-11 | 2015-05-26 | It is a kind of that method is recommended based on spectral clustering and the book labels of mass-rent technology |
Applications Claiming Priority (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2015101072904 | 2015-03-11 | ||
| CN201510107290 | 2015-03-11 | ||
| CN201510270676.7A CN104915388B (en) | 2015-03-11 | 2015-05-26 | It is a kind of that method is recommended based on spectral clustering and the book labels of mass-rent technology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104915388A CN104915388A (en) | 2015-09-16 |
| CN104915388B true CN104915388B (en) | 2018-03-16 |
Family
ID=54084451
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510270676.7A Active CN104915388B (en) | 2015-03-11 | 2015-05-26 | It is a kind of that method is recommended based on spectral clustering and the book labels of mass-rent technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104915388B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105426826A (en) * | 2015-11-09 | 2016-03-23 | 张静 | Tag noise correction based crowd-sourced tagging data quality improvement method |
| CN106202184B (en) * | 2016-06-27 | 2019-05-31 | 华中科技大学 | A kind of books personalized recommendation method and system towards libraries of the universities |
| CN107301199B (en) * | 2017-05-17 | 2021-02-12 | 北京融数云途科技有限公司 | Data tag generation method and device |
| CN110851706B (en) | 2019-10-10 | 2022-11-01 | 百度在线网络技术（北京）有限公司 | Training method and device for user click model, electronic equipment and storage medium |
| US11113580B2 (en) | 2019-12-30 | 2021-09-07 | Industrial Technology Research Institute | Image classification system and method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101901450A (en) * | 2010-07-14 | 2010-12-01 | 中兴通讯股份有限公司 | Media content recommendation method and media content recommendation system |
| CN102376063A (en) * | 2011-11-29 | 2012-03-14 | 北京航空航天大学 | Social-label-based method for optimizing personalized recommendation system |
| JP2013084216A (en) * | 2011-10-12 | 2013-05-09 | Ntt Docomo Inc | Fixed phrase discrimination device and fixed phrase discrimination method |
-
2015
- 2015-05-26 CN CN201510270676.7A patent/CN104915388B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101901450A (en) * | 2010-07-14 | 2010-12-01 | 中兴通讯股份有限公司 | Media content recommendation method and media content recommendation system |
| JP2013084216A (en) * | 2011-10-12 | 2013-05-09 | Ntt Docomo Inc | Fixed phrase discrimination device and fixed phrase discrimination method |
| CN102376063A (en) * | 2011-11-29 | 2012-03-14 | 北京航空航天大学 | Social-label-based method for optimizing personalized recommendation system |
Non-Patent Citations (2)
| Title |
|---|
| 基于标签和关联规则挖掘的图书组合推荐系统模型研究;李默等;《计算机应用研究》;20140831;全文 * |
| 标签技术在高效图书馆OPAC系统中的应用调查;罗琳等;《图书情报工作》;20130228;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104915388A (en) | 2015-09-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104915388B (en) | It is a kind of that method is recommended based on spectral clustering and the book labels of mass-rent technology | |
| CN101853299B (en) | Image searching result ordering method based on perceptual cognition | |
| CN103631929B (en) | A kind of method of intelligent prompt, module and system for search | |
| CN103778148B (en) | Life cycle management method and equipment for data file of Hadoop distributed file system | |
| CN102446225A (en) | Real-time search method, device and system | |
| CN104394118A (en) | User identity identification method and system | |
| CN105205689A (en) | Method and system for recommending commercial tenant | |
| CN104375992A (en) | Address matching method and device | |
| CN102708130A (en) | Scalable engine that computes user micro-segments for offer matching | |
| CN104391908B (en) | Multiple key indexing means based on local sensitivity Hash on a kind of figure | |
| CN106547864A (en) | A kind of Personalized search based on query expansion | |
| CN108171071A (en) | A kind of multiple key towards cloud computing can sort cipher text retrieval method | |
| CN103412903B (en) | The Internet of Things real-time searching method and system predicted based on object of interest | |
| CN104615734B (en) | A kind of community management service big data processing system and its processing method | |
| CN106095964A (en) | A kind of method that data are carried out visualization filing and search | |
| CN107357845A (en) | A kind of tour interest commending system and recommendation method based on Spark | |
| Tian et al. | A music recommendation system based on logistic regression and eXtreme gradient boosting | |
| CN102737027B (en) | Individuation search method and system | |
| CN107341199A (en) | A kind of recommendation method based on documentation ＆ info general model | |
| CN102737123A (en) | Multidimensional data distribution method | |
| CN106503246A (en) | Method for establishing ancient book intelligent digital document library | |
| CN106649380A (en) | Hot spot recommendation method and system based on tag | |
| CN109815309A (en) | A kind of user information recommended method and system based on personalization | |
| CN107291951A (en) | Data processing method, device, storage medium and processor | |
| CN106777395A (en) | A kind of topic based on community's text data finds system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |