CN104346442B - A kind of Rules extraction method of Process-Oriented object data - Google Patents
A kind of Rules extraction method of Process-Oriented object data Download PDFInfo
- Publication number
- CN104346442B CN104346442B CN201410541881.8A CN201410541881A CN104346442B CN 104346442 B CN104346442 B CN 104346442B CN 201410541881 A CN201410541881 A CN 201410541881A CN 104346442 B CN104346442 B CN 104346442B
- Authority
- CN
- China
- Prior art keywords
- cluster
- object data
- chain
- data
- clusters
- 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
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention relates to a kind of Rules extraction method of Process-Oriented object data, comprise the following steps：Step S1：Determine the optimal quantity that clusters of flow object data；Step S2：Flow object data is clustered using K means algorithms, while the reasonability of the optimal quantity that clusters in verification step S1, goes to step S3 if the optimal reasonable quantity that clusters in step S1, otherwise go to step S1；Step S3：Association rule algorithm excavates the correlation rule between the cluster of different measuring points between being tieed up using Apriori；Step S4：Determine the most strong association chain of flow object data；Step S5：According to the state value of all measuring points on most strong association chain, obtain recording the state chain of each measuring point state value, relevant industries are instructed according to state chain；The efficiency that data rule is extracted is improved, and from the ability of flow object extracting data knowledge.
Description
Technical field
The invention belongs to Data Analysis Services technical field, it is related to a kind of Rules extraction method of data, it is especially a kind of
The Rules extraction method of Process-Oriented object data；The efficiency that data rule is extracted is improved, and is carried from flow object data
Take the ability of knowledge.
Background technology
With the development of big data technology, carry out Knowledge Discovery using big data and paid close attention to by people, its major reason be by
Increasingly enriched with the accumulation of data in knowledge, Industry Control flow, auxiliary control decision are optimized using these data
Demand be continuously increased；In Industry Control, single process control optimization means cause us to lack foundation in production and go
More effectively improve procedure parameter, it is difficult to further refine control strategy, optimal control parameter, therefore how from flow object
Extraction process parameter and control parameter are simultaneously arranged, flow object running status and rule are extracted from substantial amounts of historical data
Then and to be used be the key solved the problems, such as；And the process data accumulated in process industry control object, exist stronger
Procedure parameter complexity, High relevancy, the non-linear and sequential of data variation and the inconsistency of sampling, to data mining and
Rule discovery brings difficulty, operation efficiency of the existing rule discovery algorithm in terms of processing too many levels, Complicated Flow object
It is relatively low, and obtained result of calculation is difficult to auxiliary optimization function of the embodiment to process industry.Therefore existing big data is improved
Rule discovery method, is allowed to from flow object extraction process parameter and control parameter, utilizes in substantial amounts of historical data and extract
Go out flow object running status and rule and be used, ask as big data process field at this stage is in the urgent need to address
Topic.
The content of the invention
It is an object of the present invention to for defect present in above-mentioned prior art, there is provided design a kind of Process-Oriented pair
The Rules extraction method of image data, to solve above-mentioned technical problem.
To achieve the above object, the present invention provides following technical scheme：
A kind of Rules extraction method of Process-Oriented object data, comprises the following steps：
Step S1：Determine the optimal quantity that clusters of flow object data；
Step S2：Flow object data is clustered using K-means algorithms, while optimal poly- in verification step S1
The reasonability of number of clusters amount, goes to step S3 if the optimal reasonable quantity that clusters in step S1, otherwise goes to step S1；
Step S3：Association rule algorithm excavates the correlation rule between the cluster of different measuring points between being tieed up using Apriori；
Step S4：Determine the most strong association chain of flow object data；
Step S5：According to the state value of all measuring points on most strong association chain, obtain recording the state chain of each measuring point state value,
Relevant industries are instructed according to state chain.
Preferably, the step S1 comprises the following steps：
Step S101：Take period TM, any point X in flow object data after the sequence that clocks adjustmentjVariable quantity be Di,
Wherein,
Step S102：Obtain TMThe maximum measuring point of interior variable quantity, remembers DM=max { Di, i=1,2 ..., n }, obtain TMIt is interior to become
The maximum measuring point X of change amountM；
Step S103：To the measuring point X obtained in step S102MCarry out optimum k value calculating；
Step S104：By the measuring point X obtained in step S102MClustered, and result evaluated with evaluation function,
Determine optimum k value.
Preferably, the step S2 comprises the following steps：
Step S201：Flow object data is clustered using K-means algorithms, clustering formula is：Wherein, p is CiIn object, miFor cluster CiAverage；
Step S202：Using the optimal quantity that clusters in the silhouette coefficient verification step S1 based on condensation degree and separating degree
Reasonability；
Step S203：The reasonability of the optimal quantity that clusters, goes to step S3, otherwise if rationally in judgment step S202
Go to step S1.
Preferably, the step S3 is comprised the following steps that：
If any two measuring point XiAnd XjBetween it is any two cluster between correlation rule be XiCluster kia→XjCluster kjb, its
In, i, j ∈ { 1,2...n } and i ≠ j, by XiCluster kia→XjCluster kjbIa → jb is designated as, the support of the correlation rule is S
(ia → jb),
Wherein, | ia, ib | it is X in data setiCluster kiaAnd XjCluster kjbSimultaneous affairs number, | T | represent institute
Some affairs sums, the regular interest-degree is I (ia → jb),
Wherein, C (ia → jb) is the regular confidence level, and S (jb) is regular consequent XjCluster kjbSupport, as I (ia
→ jb)=1 when, represent XiCluster kiaAnd XjCluster kjbIt is separate；As I (ia → jb) ＞ 1, X is representediCluster kiaAnd XjIt is poly-
Class kjbIt is positively related, I (ia → jb) is bigger, illustrates XiCluster kiaTo XjCluster kjbFacilitation is bigger；As I (ia → jb) ＜
When 1, X is representediCluster kiaAnd XjCluster kjbIt is negatively correlated, I (ia → jb) is smaller, illustrates XiCluster kiaTo XjCluster kjbSuppress
Effect is bigger.
Preferably, the step S4 comprises the following steps：
Step S401：Choose any point and be used as first node；
Step S402：By the first node of preceding paragraph all measuring points rule in find the degree of association it is maximum and it is consequent not repeat chain
The rule of upper existing measuring point, and next node is used as using the rule is consequent；
Step S403：The next node determined in judgment step S402 whether be the chain last node, if
It is to go to step S404, if it is not, then using the next node as new first node, going to step S402；
Step S404：Circulation is terminated, obtains most associating chain by force.
Preferably, the step S4 also comprises the following steps：
Using all nodes as first node, the most strong association chain by first node of all nodes is obtained.
Preferably, the step S4 also comprises the following steps：
Step S45：Judge whether all most strong association chains include all measuring points, if it is, step S5 is gone to, if not
It is then to go to step S46；
Step S46：Relevance tree is constructed, each most strong chain that associates is as a branch of the relevance tree.
Preferably, the step S202 specifically includes following steps：
Step S2021：I-th of data point is taken, the data point every other data point into cluster where it is calculated and is averaged
Apart from ai；
Step S2022：The average distance of i-th of data point all data points into other clusters is calculated, and remembers all average
Minimum value is b in distancei；
Step S2023：The silhouette coefficient of i-th of data point is calculated, is designated as
Step S2024：Silhouette coefficient of the clusters number for K cluster set is calculated, k is designated asM=mink{maxnsk, wherein,
skSpan be [- 1,1], work as sijDuring ＞ 0, cluster result is represented preferably, and sijValue got over closer to 1 cluster result
It is good；Work as sijWhen≤0, represent that cluster result is poor, and sijValue it is poorer closer to -1 cluster result.
The beneficial effects of the present invention are divided node data by the state of measuring point, it is possible to achieve between measuring point
Correlation rule excavation, obtain the incidence relation between any point different conditions；Meanwhile, data complexity is reduced, is carried
High data operation speed；In addition, design principle of the present invention is reliable, with application prospect widely.
As can be seen here, the present invention compared with prior art, improves with prominent substantive distinguishing features and significantly, and it is implemented
Beneficial effect be also obvious.
Embodiment
Below by specific embodiment, the present invention will be described in detail, and following examples are explanation of the invention, and
The invention is not limited in implementation below.
A kind of Rules extraction method for Process-Oriented object data that the present invention is provided, comprises the following steps：
Step S1：Determine the optimal quantity that clusters of flow object data；
Step S2：Flow object data is clustered using K-means algorithms, while optimal poly- in verification step S1
The reasonability of number of clusters amount, goes to step S3 if the optimal reasonable quantity that clusters in step S1, otherwise goes to step S1；
Step S3：Association rule algorithm excavates the correlation rule between the cluster of different measuring points between being tieed up using Apriori；
Step S4：Determine the most strong association chain of flow object data；
Step S5：According to the state value of all measuring points on most strong association chain, obtain recording the state chain of each measuring point state value,
Relevant industries are instructed according to state chain.
In the present embodiment, the step S1 comprises the following steps：
Step S101：Take period TM, any point X in flow object data after the sequence that clocks adjustmentjVariable quantity be Di,
Wherein,
Step S102：Obtain TMThe maximum measuring point of interior variable quantity, remembers DM=max { Di, i=1,2 ..., n }, obtain TMIt is interior to become
The maximum measuring point X of change amountM；
Step S103：To the measuring point X obtained in step S102MCarry out optimum k value calculating；
Step S104：By the measuring point X obtained in step S102MClustered, and result evaluated with evaluation function,
Determine optimum k value.
In the present embodiment, the step S2 comprises the following steps：
Step S201：Flow object data is clustered using K-means algorithms, clustering formula is：Wherein, p is CiIn object, miFor cluster CiAverage；
Step S202：Using the optimal quantity that clusters in the silhouette coefficient verification step S1 based on condensation degree and separating degree
Reasonability；
Step S203：The reasonability of the optimal quantity that clusters, goes to step S3, otherwise if rationally in judgment step S202
Go to step S1.
In the present embodiment, the step S3's comprises the following steps that：
If any two measuring point XiAnd XjBetween it is any two cluster between correlation rule be XiCluster kia→XjCluster kjb, its
In, i, j ∈ { 1,2...n } and i ≠ j, by XiCluster kia→ Xj clusters kjbIa → jb is designated as, the support of the correlation rule is S
(ia → jb),
Wherein, | ia, jb | it is X in data setiCluster kiaAnd XjCluster kjbSimultaneous affairs number, | T | represent institute
Some affairs sums, the regular interest-degree is I (ia → jb),
Wherein, C (ia → jb) is the regular confidence level, and S (jb) is regular consequent XjCluster kjbSupport, as I (ia
→ jb)=1 when, represent XiCluster kiaAnd XjCluster kjbIt is separate；As I (ia → jb) ＞ 1, X is representediCluster kiaAnd XjIt is poly-
Class kjbIt is positively related, I (ia → jb) is bigger, illustrates XiCluster kiaTo XjCluster kjbFacilitation is bigger；As I (ia → jb) ＜
When 1, X is representediCluster kiaAnd XjCluster kjbIt is negatively correlated, I (ia → jb) is smaller, illustrates XiCluster kiaTo XjCluster kjbSuppress
Effect is bigger.
In the present embodiment, the step S4 comprises the following steps：
Step S401：Choose any point and be used as first node；
Step S402：By the first node of preceding paragraph all measuring points rule in find the degree of association it is maximum and it is consequent not repeat chain
The rule of upper existing measuring point, and next node is used as using the rule is consequent；
Step S403：The next node determined in judgment step S402 whether be the chain last node, if
It is to go to step S404, if it is not, then using the next node as new first node, going to step S402；
Step S404：Circulation is terminated, obtains most associating chain by force.
In the present embodiment, the step S4 also comprises the following steps：
Using all nodes as first node, the most strong association chain by first node of all nodes is obtained.
In the present embodiment, the step S4 also comprises the following steps：
Step S45：Judge whether all most strong association chains include all measuring points, if it is, step S5 is gone to, if not
It is then to go to step S46；
Step S46：Relevance tree is constructed, each most strong chain that associates is as a branch of the relevance tree.
In the present embodiment, the step S202 specifically includes following steps：
Step S2021：I-th of data point is taken, the data point every other data point into cluster where it is calculated and is averaged
Apart from ai；
Step S2022：The average distance of i-th of data point all data points into other clusters is calculated, and remembers all average
Minimum value is b in distancei；
Step S2023：The silhouette coefficient of i-th of data point is calculated, is designated as
Step S2024：Silhouette coefficient of the clusters number for K cluster set is calculated, k is designated asM=mink{maxnsk, wherein,
skSpan be [- 1,1], work as sijDuring ＞ 0, cluster result is represented preferably, and sijValue got over closer to 1 cluster result
It is good；Work as sijWhen≤0, represent that cluster result is poor, and sijValue it is poorer closer to -1 cluster result.
Disclosed above is only the preferred embodiment of the present invention, but the present invention is not limited to this, any this area
What technical staff can think does not have a creative change, and some improvement made without departing from the principles of the present invention and
Retouching, should all be within the scope of the present invention.
Claims (7)
1. a kind of Rules extraction method of Process-Oriented object data, comprises the following steps：
Step S1：Determine the optimal quantity that clusters of flow object data；
Step S2：Flow object data is clustered using K-means algorithms, while the optimal number that clusters in verification step S1
The reasonability of amount, goes to step S3 if the optimal reasonable quantity that clusters in step S1, otherwise goes to step S1；
Step S3：Association rule algorithm excavates the correlation rule between the cluster of different measuring points between being tieed up using Apriori；
Step S4：Determine the most strong association chain of flow object data；
Step S5：According to the state value of all measuring points on most strong association chain, obtain recording the state chain of each measuring point state value, according to
State chain is instructed relevant industries；
The step S1 comprises the following steps：
Step S101：Take period TM, any point X in flow object data after the sequence that clocks adjustmentjVariable quantity be Di, wherein,
Step S102：Obtain TMThe maximum measuring point of interior variable quantity, remembers DM=max { Di, i=1,2 ..., n }, obtain TMInterior variable quantity
Maximum measuring point XM；
Step S103：To the measuring point X obtained in step S102MCarry out optimum k value calculating；
Step S104：By the measuring point X obtained in step S102MClustered, and result is evaluated with evaluation function, it is determined that
Optimum k value.
2. the Rules extraction method of Process-Oriented object data according to claim 1, it is characterised in that：The step S2
Comprise the following steps：
Step S201：Flow object data is clustered using K-means algorithms, clustering formula is：Wherein, p is CiIn object, miFor cluster CiAverage；
Step S202：Using optimal the reasonable of quantity that cluster in the silhouette coefficient verification step S1 based on condensation degree and separating degree
Property；
Step S203：The reasonability of the optimal quantity that clusters, goes to step S3 if rationally, otherwise goes in judgment step S202
Step S1.
3. the Rules extraction method of Process-Oriented object data according to claim 2, it is characterised in that：The step S3
Comprise the following steps that：
If any two measuring point XiAnd XjBetween it is any two cluster between correlation rule be XiCluster kia→XjCluster kjb, wherein, i,
J ∈ { 1,2...n } and i ≠ j, by XiCluster kia→XjCluster kjbBe designated as ia → jb, the support of the correlation rule for S (ia →
Jb),
Wherein, | ia, jb | it is X in data setiCluster kiaAnd XjCluster kjbSimultaneous affairs number, | T | represent all
Affairs sum, the regular interest-degree is I (ia → jb),
Wherein, C (ia → jb) is the regular confidence level, and S (jb) is regular consequent XjCluster kjbSupport.
4. the Rules extraction method of Process-Oriented object data according to claim 3, it is characterised in that：The step S4
Comprise the following steps：
Step S401：Choose any point and be used as first node；
Step S402：By the first node of preceding paragraph all measuring points rule in find the degree of association it is maximum and it is consequent repeat chain on
There is the rule of measuring point, and be used as next node using the rule is consequent；
Step S403：The next node determined in judgment step S402 whether be the chain last node, if it is
Step S404 is gone to, if it is not, then using the next node as new first node, going to step S402；
Step S404：Circulation is terminated, obtains most associating chain by force.
5. the Rules extraction method of Process-Oriented object data according to claim 4, it is characterised in that：The step S4
Also comprise the following steps：
Using all nodes as first node, the most strong association chain by first node of all nodes is obtained.
6. the Rules extraction method of Process-Oriented object data according to claim 5, it is characterised in that：The step S4
Also comprise the following steps：
Step S45：Judge whether all most strong association chains include all measuring points, if it is, step S5 is gone to, if it is not,
Then go to step S46；
Step S46：Relevance tree is constructed, each most strong chain that associates is as a branch of the relevance tree.
7. the Rules extraction method of Process-Oriented object data according to claim 6, it is characterised in that：The step
S202 specifically includes following steps：
Step S2021：I-th of data point is taken, the average distance of the data point every other data point into cluster where it is calculated
ai；
Step S2022：The average distance of i-th of data point all data points into other clusters is calculated, and remembers all average distances
Middle minimum value is bi；
Step S2023：The silhouette coefficient of i-th of data point is calculated, is designated as
Step S2024：Silhouette coefficient of the clusters number for K cluster set is calculated, k is designated asM=mink{maxΩ sk, wherein, sk's
Span is [- 1,1].
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410541881.8A CN104346442B (en) | 2014-10-14 | 2014-10-14 | A kind of Rules extraction method of Process-Oriented object data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410541881.8A CN104346442B (en) | 2014-10-14 | 2014-10-14 | A kind of Rules extraction method of Process-Oriented object data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104346442A CN104346442A (en) | 2015-02-11 |
| CN104346442B true CN104346442B (en) | 2017-10-20 |
Family
ID=52502033
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410541881.8A Expired - Fee Related CN104346442B (en) | 2014-10-14 | 2014-10-14 | A kind of Rules extraction method of Process-Oriented object data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104346442B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108647808A (en) * | 2018-04-11 | 2018-10-12 | 济南大学 | A kind of manufacturing parameter Optimization Prediction method, apparatus, equipment and storage medium |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104732067A (en) * | 2015-02-26 | 2015-06-24 | 济南大学 | Industrial process modeling forecasting method oriented at flow object |
| CN107241226B (en) * | 2017-06-29 | 2020-10-16 | 北京工业大学 | Fuzzy test method based on industrial control private protocol |
| CN107315822B (en) * | 2017-07-04 | 2020-05-12 | 国网浙江省电力公司杭州供电公司 | Knowledge point association mining method |
| CN114896228B (en) * | 2022-04-27 | 2024-04-05 | 西北工业大学 | Industrial data stream cleaning model and method based on filtering rule multistage combination optimization |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101187804A (en) * | 2006-11-15 | 2008-05-28 | 北京华电天仁电力控制技术有限公司 | Thermal power unit operation optimization rule extraction method based on data excavation |
| CN101610265A (en) * | 2009-07-24 | 2009-12-23 | 深圳市永达电子股份有限公司 | A kind of flow process recognition methods of Business Works |
| CN101655857A (en) * | 2009-09-18 | 2010-02-24 | 西安建筑科技大学 | Method for mining data in construction regulation field based on associative regulation mining technology |
| CN102831269A (en) * | 2012-08-16 | 2012-12-19 | 内蒙古科技大学 | Method for determining technological parameters in flow industrial process |
-
2014
- 2014-10-14 CN CN201410541881.8A patent/CN104346442B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101187804A (en) * | 2006-11-15 | 2008-05-28 | 北京华电天仁电力控制技术有限公司 | Thermal power unit operation optimization rule extraction method based on data excavation |
| CN101610265A (en) * | 2009-07-24 | 2009-12-23 | 深圳市永达电子股份有限公司 | A kind of flow process recognition methods of Business Works |
| CN101655857A (en) * | 2009-09-18 | 2010-02-24 | 西安建筑科技大学 | Method for mining data in construction regulation field based on associative regulation mining technology |
| CN102831269A (en) * | 2012-08-16 | 2012-12-19 | 内蒙古科技大学 | Method for determining technological parameters in flow industrial process |
Non-Patent Citations (3)
| Title |
|---|
| 关联规则算法研究及其在教学系统中的应用;曲守宁等;《计算机系统应用》;20050405(第4期);第20页左栏第4-7行，第21页右栏第21-23行，第22页右栏第6行，第23页左栏第1-3行，表1 * |
| 基于关联规则算法的研究与改进;邓子穗;《中国优秀硕士学位论文全文数据库 信息科技辑》;20140915(第9期);第29页第5.1.1.1节 * |
| 基于轮廓系数的聚类有效性分析;朱连江等;《计算机应用》;20101231;第30卷;第139页右栏第2段，第139页右栏第2段，第140页第2.1节，第140页右栏第2-22行 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108647808A (en) * | 2018-04-11 | 2018-10-12 | 济南大学 | A kind of manufacturing parameter Optimization Prediction method, apparatus, equipment and storage medium |
| CN108647808B (en) * | 2018-04-11 | 2022-03-29 | 济南大学 | Production parameter optimization prediction method, device, equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104346442A (en) | 2015-02-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104346442B (en) | A kind of Rules extraction method of Process-Oriented object data | |
| CN101950284B (en) | Chinese word segmentation method and system | |
| CN102801709B (en) | Phishing website identification system and method | |
| CN102629305B (en) | Feature selection method facing to SNP (Single Nucleotide Polymorphism) data | |
| CN103678602B (en) | Webpage filtration method with sensitivity calculation function | |
| CN107644057A (en) | A kind of absolute uneven file classification method based on transfer learning | |
| CN108550401A (en) | A kind of illness data correlation method based on Apriori | |
| CN103034687B (en) | A kind of relating module recognition methodss based on 2 class heterogeneous networks | |
| CN106384300A (en) | Big data and fuzzy model-based building abnormal energy consumption detection method and system | |
| CN103235774A (en) | Extraction method of feature words of science and technology project application form | |
| CN106844786A (en) | A kind of public sentiment region focus based on text similarity finds method | |
| CN104199838B (en) | A kind of user model constructing method based on label disambiguation | |
| CN115099998B (en) | Independent metering partitioning method, terminal equipment and storage medium of water supply network | |
| CN103729467A (en) | Community structure discovery method in social network | |
| CN109063759A (en) | A kind of neural network structure searching method applied to the more attribute forecasts of picture | |
| CN101488126A (en) | Double-language sentence alignment method and device | |
| CN104573009A (en) | Method for expanding domain knowledge database attribute | |
| CN110472518A (en) | A kind of fingerprint image quality judgment method based on full convolutional network | |
| CN102521402B (en) | Text filtering system and method | |
| CN114625868A (en) | Electric power data text classification algorithm based on selective ensemble learning | |
| CN104580254B (en) | A kind of fishing website identifying system and method | |
| CN109213869B (en) | Hot spot technology prediction method based on multi-source data | |
| CN105354264A (en) | Locality-sensitive-hashing-based subject label fast endowing method | |
| CN104217026A (en) | Chinese microblog tendency retrieving method based on graph model | |
| CN106354886A (en) | Method for screening nearest neighbor by using potential neighbor relation graph in recommendation 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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20171020 Termination date: 20181014 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |