CN103853834B - Text structure analysis-based Web document abstract generation method - Google Patents
Text structure analysis-based Web document abstract generation method Download PDFInfo
- Publication number
- CN103853834B CN103853834B CN201410090200.0A CN201410090200A CN103853834B CN 103853834 B CN103853834 B CN 103853834B CN 201410090200 A CN201410090200 A CN 201410090200A CN 103853834 B CN103853834 B CN 103853834B
- Authority
- CN
- China
- Prior art keywords
- text
- sentence
- cut
- semantic
- point
- 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.)
- Expired - Fee Related
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/80—Information retrieval; Database structures therefor; File system structures therefor of semi-structured data, e.g. markup language structured data such as SGML, XML or HTML
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Document Processing Apparatus (AREA)
Abstract
The invention discloses a text structure analysis-based Web document abstract generation method. The method comprises the steps of using a URL (uniform resource locator) as input, integrating the webpage main bodies of visual features and text features for extraction, partitioning the main bodies into a plurality of semantic paragraphs, and abstracting each semantic paragraph, so the generated abstract has higher coverage rate. The text structure analysis-based Web document summary generation method realizes the generation of the text abstract with better quality from a Webpage aiming at the conditions that the Webpage structure is complex, the main body is hard to identify and the Chinese automatic abstract is still positioned in the probe stage.
Description
Technical field
The present invention relates to Web page text extraction, natural language processing, Chinese Text Summarization technical field, specifically one
Plant the generation method of the Web document summary based on text retrieval conference TREC.
Background technology
At present, Internet has had become as the main source that people obtain information.In particularly user generates in recent years
Hold developing rapidly of (UGC), the information on Internet is just in explosive growth.Although search engine can require according to user
Return Search Results.But user still needs and finds the webpage being best suitable for oneself needs from search listing, especially because mutually
A large amount of search engine optimizations existing and reprint phenomenon in networking, to user fast and accurately find information bring very tired greatly
Difficult.
Automatic abstracting system is quickly to process Web document using computer, therefrom captures out Web document by certain compression ratio
Core content, user can therefrom obtain subject information and judge the value of this Web document, improve user and search for information
Efficiency.
Noise information is there is in a large number, such as advertisement, navigation bar, user function bar, associated recommendation, copyright letter in Web document
The information unrelated with theme such as breath.Web document is a kind of semi-structured information although having a fixed structure, but semanteme cannot be true
Fixed.The page that expression in html source code for the content and final rendering obtain has very big difference.JS and AJAX skill in recent years
The extensive application of art is so that web data is no longer static HTML code, and is dynamically generated, even for the behaviour of user
Make behavior and also can produce corresponding change.So how to extract and structure correct content related with theme from Web document,
There is certain difficulty.
The history of Chinese Text Summarization systematic research about more than two decades, but at present also in the exploratory stage, automatically
The result of summary also far can not be satisfactory.The method of autoabstract is broadly divided into two big class, based on the automatic abstract understanding
With the automatic abstract based on extraction.Because natural language processing technique is still without important breakthrough, so based on the method understanding simultaneously
Can not be real realize automatic abstract.
And the research history of the autoabstract technology of web oriented document is shorter, " compared with traditional text, the text of webpage
Loosely organized, title name is relatively less rigorous, and sentence terminates to be likely to be not over according with, and exist substantial amounts of with
The incoherent content of text, this brings certain difficulty to the generation of summary.”
Content of the invention
It is an object of the invention to provide a kind of generation method of the Web document summary based on text retrieval conference TREC, the method
The integrated use technology such as visual feature analysis, natural language analysis, text retrieval conference TREC, is each webpage in Search Results
Generate based on semantic, the preferable web-page summarization of quality, provide the user reference.
The object of the present invention is achieved like this：
A kind of generation method of the Web document summary based on text retrieval conference TREC, it comprises the following steps：
1) input the URL of webpage to be made a summary；
2) analyze from webpage view-based access control model to be made a summary and extract Web page text, specifically include；
2.1) using browser core, Web document is parsed and rendered；
2.2) piecemeal is carried out to webpage using Visual tree (VIPS) algorithm, obtain position, the area of each block；
2.3) participle is carried out to each block；
2.4) to each component analysis text feature；
2.5) whether comprise text to each block to give a mark；
2.6) text being higher than a certain threshold value by score is linked in sequence；
2.7) export Web document text；
3) autoabstract based on text retrieval conference TREC is carried out to the text extracting, specifically include：
3.1) by step 2) obtain Web page text；
3.2) participle and part-of-speech tagging are carried out to text；
3.3) carry out Text Pretreatment：Basic structure in identification text, that is, identify article title, sentence completion, paragraph
Cutting；
3.4) text is carried out with semantic section cutting, the position being changed by text retrieval conference TREC identification semanteme, as
The mark of semantic section cutting；
3.5) to each semantic section, using the promotion method of TFIDF, to importance in the semantic section that is located for each sentence
Measured, then required according to abstract word number, extract the sentence that some sentences can represent this semantic section theme；
3.6) each sentence is linked in sequence, exports digest.
Described step 2.4) in text feature be number of words, font size, assertive sentence quantity, non-assertive sentence quantity and text fragment
Quantity.
Described step 2.5) described in judge whether each block comprises text and given a mark, using below equation calculate marking
Score value：
Wherein S represents assertive sentence quantity, and N represents non-assertive sentence quantity, and P obtains according to block size and position calculation
One value, x1, y1Represent the coordinate in the block upper left corner, x2, y2Represent the coordinate in the block lower right corner.
Described step 3.4) in semanteme change position analysis identification be：
3.4-1) subordinate sentence is carried out to document D, between the adjacent sentence of each two, be cut-point undetermined；
3.4-2) each cut-point undetermined is given a mark, its formula is：
Wherein, R (si, sj) represent sentence siWith sentence sjSentence between semantic relevancy；piRepresent cut-point in sentence siWith
si-1Between, if Q is (pi) ＞ Q (pi-1) and Q (pi) ＞ Q (pi+1), p is describediIt is the maximum point of cut-point weights, so piIt is
Cut-point between semantic section in the text.A is an adjustable empirical parameter, represents that the semanteme when identifying cut-point divides
The scope of analysis, that is, represent and consider each a sentence before and after cut-point.
If 3.4-3) score value of cut-point is more than a certain threshold value, and it is local maximum, that is, score value is higher than former and later two points
The score value of cutpoint, this cut-point is exactly the cut-off of semantic section, i.e. step 3.4) described in the position that changes of semanteme.
The analysis identification step 2 of the position that described semanteme changes) between sentence the calculating of semantic relevancy include following
Step：
3.4-2-1) sentence is cut into the set of word；
3.4-2-2) below equation is used to calculate semantic relevancy between sentence
Wherein R (wi, wj) represent word wiWith word wjWord between semantic relevancy.
Described step 3.5) in each sentence, the importance in the semantic section that is located carries out tolerance and uses below equation meter
Calculate：
V(S1)=sum (w ∈ S1)*TFIDF(w)
Wherein, when calculating TFIDF (w), each paragraph is considered as independent file, several sections that entire article is comprised
Fall to being considered as file set.
The present invention can filter out in webpage and the unrelated word of theme, link etc., identifies the literary composition included in webpage
Zhang Zhengwen, accuracy rate is higher, and has higher robustness.Autoabstract flow process employs automatic based on text retrieval conference TREC
Digest technology, the summary coverage rate of generation is high and makes a summary more smooth.
The present invention can be directed to Web document, and the compression ratio requirement specified by user is it is only necessary to input the URL of webpage to be made a summary
So that it may within the time of several seconds, formation can cover the original text meaning for address, more accurate, smooth summary, helps user quickly accurate
True finds information in the Internet.
Brief description
Fig. 1 is flow chart of the present invention；
Fig. 2 is Web-page preprocessing flow chart of the present invention；
Fig. 3 is autoabstract flow chart of the present invention.
Specific embodiment
The invention discloses a kind of Web document abstraction generating method of Search Engine-Oriented, a Web can be automatically analyzed
Webpage, and the text snippet of reaction of formation Web page subject.
The present invention comprise a Web page text combining visual signature and text feature extract and one be based on by literary composition
This structural analysis carries out the automatic text summarization of sub-topicses division.
The present invention, using a URL as input, through Web page text extraction, two stages of autoabstract, ultimately generates literary composition
This summary.
Specific algorithm to described two stages below, makees furtherly in conjunction with as a example making a summary to a news web page
Bright：
Fig. 1 describe from URL to be made a summary to generate summary overall procedure, which includes Web-page preprocessing flow process and from
Dynamic summary flow process.
Specifically, in an embodiment, the present invention obtains in Web-page preprocessing flow process (see Fig. 2) URL input step and waits to pluck
Want the URL of news web page.Web-page preprocessing flow process is passed through to analyze visual signature, can more accurately find the textual in webpage
Point, have more high robust than additive method.Consider simultaneously text feature, text Controlling UEP, html tag feature,
Other features such as semantic feature, improve the accuracy that Web page text extracts further.
Webpage rendering step is responsible for reading the input corresponding webpage of URL, in this embodiment, using IE11 browser core
Html tag is processed, and renders this webpage.On the basis of webpage renders, Visual tree analytical procedure adopts VIPS to calculate
Method, carries out Visual tree analysis to webpage, obtains position, the area of each block.In this embodiment, this step will be to be made a summary new
Hear Web-page segmentation and become 6 blocks：One top block, a bottom block, a navigation block, an advertisement block and two
Comprise the block of text.Participle step is responsible for carrying out participle to each block.Then, text feature analytical procedure is entered to word segmentation result
Style of writing eigen analysis.Last comprehensive analysis step carries out comprehensive to the feature of each block that Visual tree analysis obtains and text feature
Close analysis, export body.
In this embodiment, P (x is calculated using following equation1, y1, x2, y2).
P(x1, y1, x2, y2)=(x2-x1)*(y2-y1)-x1*y1
Wherein x1, y1Represent the coordinate in the block upper left corner, x2, y2Represent the coordinate in the block lower right corner.Then calculate each
V (s) value of block：
V (s) value of above-mentioned 6 blocks is respectively 3.7 × 10 from big to small6, 2.3 × 106, 7.5 × 105, 5.4 × 106,
3.7×105, 1.6 × 105, 1.2 × 104.
In this embodiment, the threshold value of employing is 106, so choosing V (s) to be more than 106Block, that is, V (s) value is maximum
Two blocks.In this embodiment, two maximum blocks of V (s) value are exactly two blocks comprising text, so correct extract
Arrive body.
After extracting body, then carry out autoabstract flow process (see Fig. 3), comprise phase between Text Pretreatment, word
Relatedness computation, semantic section segmentation, these steps of summarization generation between Guan Du calculating, sentence.
One Text Pretreatment step, the basic structure in identification text, that is, identify article title, sentence completion, paragraph
Cutting.In this embodiment, body comprises 8 paragraphs, 23 sentences altogether.
The computational semantics knowledge that between word, relatedness computation step is provided based on Hownet, by calculating the former phase of justice of two words
To obtain the degree of association of two words like degree.Using formula as follows：
R(w1, w2)=max (Rele (Ci, Cj))(Ci∈w1, Cj∈w2)
Wherein R (w1, w2) illustrate semantic relevancy between two words, Rele (Ci, Cj) illustrate two former correlations of justice
Degree, takes its maximum to represent the semantic relevancy of two words.
Between sentence, degree of association step obtains the degree of association of two sentences by analyzing the degree of association between word in two sentences.
Wherein R (s1, s2) illustrate degree of association between two sentences, for the word in each sentence 1, look in sentence 2 therewith
The maximum word of degree of association, calculates the degree of association between this two words.Finally by these maximums sue for peace, obtain this two sentences it
Between degree of association.
One semantic section segmentation step, with reference to document《Ground based on the Text Structure Analysis of content relatedness computation
Study carefully》To carry out text retrieval conference TREC.First sentence after the feature of cut-point is cut-point between semantic section and some before
The degree of association very little of sentence, and larger with the degree of association of several sentences afterwards.Using below equation to 23 in this embodiment
22 cut-points between individual sentence calculate the score value of cut-point, and find the maximum point of function Q (pi)：
In this embodiment, Q (pi) comprise 2 maximum points, according to this two maximum points, this news is divided into 3
Individual semanteme section.Each semantic section contains sub-topicses of news, and in this embodiment, first semantic section is to news thing
The general introduction of part, latter two semantic section is two sides comment to this media event respectively.
One summarization generation step, requires according to user, extracts summary from the text of text formatting by a certain percentage.
In this embodiment, this summarization generation step passes through relatedness computation step between sentence, calculates in each sub-topics
The degree of association sum of sentence and article title sequence of words, so that it is determined that the value of each sub-topicses.Sentence is extracted from sub-topicses
Quantity and the degree of association of this sub-topics and article title be directly proportional.
In this embodiment, the ratio that user specifies is 0.2, that is, 5 words in extracting 23 form summary.By to 3
The value of individual sub-topicses is calculated, and determines and extracts 2,1,1 sentence respectively from 3 semantic sections.Finally, described summarization generation
5 chosen summary sentences are linked in sequence by step, are formed and make a summary and export.
Claims (5)
1. a kind of based on text retrieval conference TREC Web document summary generation method it is characterised in that：The method includes following step
Suddenly：
1) input the URL of webpage to be made a summary；
2) analyze from webpage view-based access control model to be made a summary and extract Web page text, specifically include；
2.1) using browser core, Web document is parsed and rendered；
2.2) piecemeal is carried out to webpage using vision tree algorithm, obtain position, the area of each block；
2.3) participle is carried out to each block；
2.4) to each component analysis text feature；
2.5) whether text is comprised to each block give a mark, the score value of marking is calculated using below equation：
Wherein S represents assertive sentence quantity, and N represents non-assertive sentence quantity, and P is being obtained according to block size and position calculation
Value, x1, y1Represent the coordinate in the block upper left corner, x2, y2Represent the coordinate in the block lower right corner；
2.6) text being higher than a certain threshold value by score is linked in sequence；
2.7) export Web document text；
3) autoabstract based on text retrieval conference TREC is carried out to the text extracting, specifically include：
3.1) by step 2) obtain Web page text；
3.2) participle and part-of-speech tagging are carried out to text；
3.3) carry out Text Pretreatment：Basic structure in identification text, that is, identify article title, sentence completion, paragraph cutting；
3.4) text is carried out with semantic section cutting, the position changing by text retrieval conference TREC identification semanteme, as semanteme
The mark of section cutting；
3.5) to each semantic section, using the promotion method of TFIDF, importance in the semantic section that is located for each sentence is carried out
Tolerance, then requires according to abstract word number, extracts the sentence that some sentences can represent this semantic section theme；
3.6) each sentence is linked in sequence, exports digest.
2. method according to claim 1 it is characterised in that：Step 2.4) described in text feature be number of words, font size,
Assertive sentence quantity, non-assertive sentence quantity and text fragment quantity.
3. method according to claim 1 it is characterised in that：Step 3.4) described in semanteme change position point
Analysing identification is：
3.4-1) subordinate sentence is carried out to document D, it is cut-point undetermined between the adjacent sentence of each two；
3.4-2) each cut-point undetermined is given a mark, its formula is：
Wherein, R (si, sj) represent sentence siWith sentence sjSentence between semantic relevancy；piRepresent cut-point in sentence siAnd si-1It
Between, if Q is (pi) ＞ Q (pi-1) and Q (pi) ＞ Q (pi+1), p is describediIt is the maximum point of cut-point weights, so piIt is this article
Cut-point between semantic section in this；A is an adjustable empirical parameter, the semantic analysis when identifying cut-point for the expression
Scope, that is, represent and consider each a sentence before and after cut-point；
If 3.4-3) score value of cut-point is more than a certain threshold value, and it is local maximum, that is, score value is higher than former and later two cut-points
Score value, this cut-point is exactly the cut-off of semantic section, i.e. step 3.4) described in the position that changes of semanteme.
4. method according to claim 3 it is characterised in that：Step 3.4-2) described between sentence semantic relevancy calculating
Comprise the following steps：
3.4-2-1) sentence is cut into the set of word；
3.4-2-2) below equation is used to calculate semantic relevancy between sentence
Wherein R (wi, wj) represent word wiWith word wjWord between semantic relevancy.
5. method according to claim 1 it is characterised in that：Step 3.5) described in each sentence in the semantic section that is located
In importance carry out tolerance calculated using below equation：
V(S1)=sum (w ∈ S1)*TFIDF(w)
Wherein, when calculating TFIDF (w), each paragraph is considered as independent file, several paragraphs that entire article is comprised regard
For file set.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410090200.0A CN103853834B (en) | 2014-03-12 | 2014-03-12 | Text structure analysis-based Web document abstract generation method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410090200.0A CN103853834B (en) | 2014-03-12 | 2014-03-12 | Text structure analysis-based Web document abstract generation method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103853834A CN103853834A (en) | 2014-06-11 |
| CN103853834B true CN103853834B (en) | 2017-02-08 |
Family
ID=50861489
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410090200.0A Expired - Fee Related CN103853834B (en) | 2014-03-12 | 2014-03-12 | Text structure analysis-based Web document abstract generation method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103853834B (en) |
Families Citing this family (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105677764B (en) * | 2015-12-30 | 2020-05-08 | 百度在线网络技术（北京）有限公司 | Information extraction method and device |
| CN106484768B (en) * | 2016-09-09 | 2019-12-31 | 天津海量信息技术股份有限公司 | Local feature extraction method and system for text content saliency region |
| CN106844340B (en) * | 2017-01-10 | 2020-04-07 | 北京百度网讯科技有限公司 | News abstract generating and displaying method, device and system based on artificial intelligence |
| CN108959312B (en) | 2017-05-23 | 2021-01-29 | 华为技术有限公司 | Method, device and terminal for generating multi-document abstract |
| CN107346335B (en) * | 2017-06-28 | 2020-04-14 | 浙江大学 | Webpage theme block identification method based on combination characteristics |
| CN107622046A (en) * | 2017-09-01 | 2018-01-23 | 广州慧睿思通信息科技有限公司 | A kind of algorithm according to keyword abstraction text snippet |
| CN107766325B (en) * | 2017-09-27 | 2021-05-28 | 百度在线网络技术（北京）有限公司 | Text splicing method and device |
| CN108427761B (en) * | 2018-03-21 | 2022-01-14 | 腾讯科技（深圳）有限公司 | News event processing method, terminal, server and storage medium |
| CN110889280B (en) * | 2018-09-06 | 2023-09-26 | 上海智臻智能网络科技股份有限公司 | Knowledge base construction method and device based on document splitting |
| CN110968752A (en) * | 2018-09-28 | 2020-04-07 | 珠海格力电器股份有限公司 | Data acquisition method and device, storage medium and electronic equipment |
| CN113515627B (en) * | 2021-05-19 | 2023-07-25 | 北京世纪好未来教育科技有限公司 | Document detection method, device, equipment and storage medium |
| CN114330315A (en) * | 2021-12-28 | 2022-04-12 | 浙江大华技术股份有限公司 | Method and device for processing secure text, storage medium and electronic device |
| CN114417808B (en) * | 2022-02-25 | 2023-04-07 | 北京百度网讯科技有限公司 | Article generation method and device, electronic equipment and storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1536483A (en) * | 2003-04-04 | 2004-10-13 | 陈文中 | Method for extracting and processing network information and its system |
| CN102446191A (en) * | 2010-10-13 | 2012-05-09 | 北京创新方舟科技有限公司 | Method for generating webpage content abstracts and equipment and system adopting same |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8930376B2 (en) * | 2008-02-15 | 2015-01-06 | Yahoo! Inc. | Search result abstract quality using community metadata |
-
2014
- 2014-03-12 CN CN201410090200.0A patent/CN103853834B/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1536483A (en) * | 2003-04-04 | 2004-10-13 | 陈文中 | Method for extracting and processing network information and its system |
| CN102446191A (en) * | 2010-10-13 | 2012-05-09 | 北京创新方舟科技有限公司 | Method for generating webpage content abstracts and equipment and system adopting same |
Non-Patent Citations (3)
| Title |
|---|
| "基于内容相关度计算的文本结构分析方法研究";钟茂生;《中国博士学位论文全文数据库信息科技辑》;20101015(第10期);I138-81 * |
| "基于分块的网页正文信息提取算法研究";黄文蓓 等;《计算机应用》;20070601;第6卷(第S1期);24-26 * |
| "基于潜在语义分析的多网页自动文摘研究";何媛媛;《中国优秀硕士学位论文全文数据库信息科技辑》;20080115(第01期);I138-1310 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103853834A (en) | 2014-06-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103853834B (en) | Text structure analysis-based Web document abstract generation method | |
| CN105022725B (en) | A kind of text emotion trend analysis method applied to finance Web fields | |
| TWI695277B (en) | Automatic website data collection method | |
| CN105022803B (en) | A kind of method and system for extracting Web page text content | |
| CN107590219A (en) | Webpage personage subject correlation message extracting method | |
| CN107704503A (en) | User's keyword extracting device, method and computer-readable recording medium | |
| CN104199972A (en) | Named entity relation extraction and construction method based on deep learning | |
| Piperski et al. | Big and diverse is beautiful: A large corpus of Russian to study linguistic variation | |
| CN105975453A (en) | Method and device for comment label extraction | |
| CN103294781A (en) | Method and equipment used for processing page data | |
| CN102609427A (en) | Public opinion vertical search analysis system and method | |
| CN101887443A (en) | Method and device for classifying texts | |
| CN103309862A (en) | Webpage type recognition method and system | |
| CN105843796A (en) | Microblog emotional tendency analysis method and device | |
| JP2006351002A5 (en) | ||
| CN104699797A (en) | Webpage data structured analytic method and device | |
| CN106156143A (en) | Page processor and web page processing method | |
| CN104199845A (en) | On-line comment sentiment classification method based on agent model | |
| CN111199151A (en) | Data processing method and data processing device | |
| CN104346382B (en) | Use the text analysis system and method for language inquiry | |
| CN110008473A (en) | A kind of medical text name Entity recognition mask method based on alternative manner | |
| CN107436931B (en) | Webpage text extraction method and device | |
| Siklósi | Using embedding models for lexical categorization in morphologically rich languages | |
| CN104636324B (en) | Topic source tracing method and system | |
| Park et al. | Automatic analysis of thematic structure in written English |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20170208 Termination date: 20200312 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |