CN108346007A - A kind of mobile phone labeling detection data analysis method based on FP-Growth algorithms - Google Patents
A kind of mobile phone labeling detection data analysis method based on FP-Growth algorithms Download PDFInfo
- Publication number
- CN108346007A CN108346007A CN201810174321.1A CN201810174321A CN108346007A CN 108346007 A CN108346007 A CN 108346007A CN 201810174321 A CN201810174321 A CN 201810174321A CN 108346007 A CN108346007 A CN 108346007A
- Authority
- CN
- China
- Prior art keywords
- unqualified
- frequent
- subdata
- mobile phone
- predicate set
- 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
Classifications
-
- 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
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06395—Quality analysis or management
-
- 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—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/04—Manufacturing
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Abstract
A kind of mobile phone based on FP Growth algorithms of the present invention labels detection data analysis method：S1 pre-processes mobile phone labeling detecting system testing result database；S2 identifies the cubical frequent predicate set of each subdata to each subdata cube application FP Growth algorithms statistics；The cubical frequent predicate set of multiple subdatas is merged into the frequent predicate set of data cube by S3；S4 after obtaining the frequent predicate set of data cube, judges that the relationship of unqualified type and unqualified factor is strong and weak by calculating.The present invention can handle mass historical data, as a result comprehensively, really, reliably；Support is obtained by the polymerizing value of data cube, simplifies calculating；FP Growth algorithms are transformed, are allowed to only generate the strong incidence relation between unqualified type and unqualified factor, without generating the strong incidence relation between unqualified factor and unqualified factor, accelerate the speed of service of system.
Description
Technical field
The present invention relates to a kind of, and the mobile phone based on FP-Growth algorithms labels detection data analysis method, belongs at data
Manage technical field.
Background technology
With the development of mobile Internet, mobile phone essential consumer goods in having become for people's lives.Global mobile phone needs
Explosive increase is asked to push flourishing for whole mobile phone industry.In order to meet the needs of people are to mobile phone, mobile-phone manufacturers
It needs to accelerate mobile phone research and development speed and speed of production.In the production line of mobile phone, mobile-phone manufacturers are needed multiple labels
It invests on mobile phone.But due to labelling the carelessness of personnel, labeling personnel often will appear the underproof situation of labeling.In order to carry
High mobile phone product export qualification rate needs to be detected mobile phone labeling.
Data correlation is a kind of important knowledge being found present in database.If two or more variables take
There are certain regularity between value, it is known as being associated with.With increasing sharply for mobile phone output, mobile phone labels detection data and presents
Explosive growth, traditional data statistical analysis method have been no longer desirable for the correlation point of magnanimity mobile phone labeling detection data
Analysis.And increasing with detection data quantity, the correlativity between data are also more apparent, so, there is an urgent need for one kind to be suitable for
The data analysing method research mobile phone of magnanimity detection data labels underproof key factor.
Invention content
The main object of the present invention is to be to provide a kind of mobile phone labeling detection data analysis based on FP-Growth algorithms
Method, it is intended to which the detection data that department is labelled to mobile-phone manufacturers mobile phones carries out data mining and analysis, to be produced for mobile phone
The adjustment that quotient labels assembly line provides data support.
To achieve the goals above, a kind of mobile phone based on FP-Growth algorithms provided by the invention labels detection data
Analysis method includes the following steps：
S1 pre-processes mobile phone labeling detecting system testing result database；It is specific as follows：
S11 rejects the qualified correlated results of mobile phone labeling detection, retains mobile phone labeling and detect underproof correlated results；
S12 is extracted and is labelled underproof type and relevant labeling personnel age, gender, station number in testing result
And the information such as labeling time；
S13 handles the data of extraction, specifically includes whole data record deletion to missing values and exceptional value.
S14, by treated, data pass through OLAP data modeling tool Workbench one five dimension data cube of generation
Body.The data cube is formed by labelling personnel's age, labeling personnel gender, time, station and the unqualified dimension of type five of labeling.
Wherein, it labels unqualified type dimension there is leakage label, inclined patch, fold, defective four members.
S15, the member in being tieed up according to unqualified type to five obtained dimension data cubes carry out OLAP sectionings, obtain
To the four-dimensional subdata cube corresponding to each unqualified type dimension member.
S2 identifies the cubical frequency of each subdata to each subdata cube application FP-Growth algorithms statistics
Numerous predicate collection；
The statistics identification step of the cubical frequent predicate set of subdata based on FP-Growth algorithms is as follows：
S21 sets the minimum support of FP-Growth algorithms；
S22 scans subdata cube for the first time, calculates the support of each dimension member, and support with the minimum of setting
Degree compares to obtain 1- frequent predicate sets；
S23, second of scanning subdata cube, creates FP-tree.
The root node for creating FP-tree first, is labeled as " null ", then reads in the record on subdata cube, and
One branch is created to each record, when increasing branch for a record, the node counts on overlay path increase corresponding
Count values, nonoverlapping part, which creates new node and creates link, is directed toward prefix；Until all records are mapped to FP-
On the paths tree；
S24 excavates the cubical frequent predicate set of subdata from FP-tree.
S241 constructs the conditional pattern base of 1- frequent predicate sets, assigns conditional pattern base as transaction set structural environment FP-
tree；
S242 finds condition frequent predicate set according to condition FP-tree, then merges with suffix pattern, frequently called
Word set.
S243, the iterative step S241 and step S242 on condition FP-tree, until setting comprising a predicate, from
And it unites
Meter identifies the cubical frequent predicate set of subdata.
The cubical frequent predicate set of multiple subdatas that above-mentioned steps obtain is merged into the frequent of data cube by S3
Predicate collection；
S31 traverses the cubical frequent predicate set list of each subdata, if the cubical frequent predicate set row of subdata
There are identical frequent predicate sets for table, then are added the support of the frequent predicate set, obtain the frequency for only including unqualified factor
Numerous predicate collection；
S32, by the cubical frequent predicate set of each subdata plus the new frequent meaning of corresponding unqualified type composition
Word set includes the frequent predicate set of unqualified factor and unqualified type.
S4 after obtaining the frequent predicate set of data cube, judges unqualified type b and unqualified factor a by calculating
Relationship it is strong and weak.
S41, setting min confidence min_conf；
S42 is calculated by the following formula the confidence level conf of unqualified type b and unqualified factor a：
Wherein, num (a) is the support of the frequent predicate set of only unqualified factor a compositions；Num (a | b) it is by not conforming to
The support of the frequent predicate set of lattice factor a and unqualified type b compositions；
S43, judges whether confidence level conf is more than the min confidence min_conf of setting, if so, thinking that this does not conform to
It is unqualified because being known as strong incidence relation described in lattice type b and a.
S44, traverses all frequent predicate sets, and result of calculation exports after being arranged from high to low according to confidence level.
Beneficial effects of the present invention：
(1) mobile phone labeling data detection method of the present invention can handle mass historical data, and result is comprehensively, very
It is real, reliable；
(2) support is obtained by the polymerizing value of data cube, enormously simplifies calculating.
(3) FP-Growth algorithms are transformed, are allowed to generate strong between unqualified type and unqualified factor
Incidence relation greatly accelerates the operation of system without generating the strong incidence relation between unqualified factor and unqualified factor
Speed.
Description of the drawings
Fig. 1 is the overall flow schematic diagram of data analysing method of the present invention.
Fig. 2 is the flow diagram of data prediction in the method for the present invention.
Fig. 3 is the flow diagram of frequent predicate set.
Specific implementation mode
In order to specifically describe the present invention, below in conjunction with the accompanying drawings and specific implementation mode to technical scheme of the present invention carry out
It is described in detail.
The related data of the mobile phone labeling testing result of the present embodiment is stored in MYSQL database.It is wrapped altogether in the database
Detection information is labelled containing 5,000,000 mobile phones.There are 6 attributes per data.Wherein, table 1 provide database each attribute and its
Codomain.
Table 2 gives the partial data content in database.
Table 1
Table 2
As shown in Figure 1, first, the data that mobile phone labels in detecting system testing result database are carried out pretreatment behaviour
Make；Secondly, the cubical frequent predicate of subdata is identified to each subdata cube running body FP-Growth algorithms statistics
Collection；Then, the cubical frequent predicate set of subdata is merged into the frequent predicate set of data cube；Finally, according to setting
Min confidence, excavate the strong incidence relation between unqualified type and unqualified factor.It is as follows：
As described in step S1, mobile phone labeling detecting system testing result database is pre-processed.As shown in Fig. 2, first
The qualified correlated results of mobile phone labeling detection is first rejected, retains mobile phone labeling and detects underproof correlated results.Then extraction inspection
It surveys in result and labels the information such as unqualified type and relevant labeling personnel age, gender, station number and labeling time.Its
In, 48662 mobile phone labeling are extracted altogether detects unqualified result.
The mobile phone labeling of extraction is detected whole data record with missing values in unqualified result to delete, by database
In have "+-：”“" as meaningless character whole data record delete.And the descriptive information in information is converted
For numerical information.Finally, by treated, data pass through OLAP data modeling tool Workbench one five dimension data of generation
Cube.Convert data cube to corresponding data cube body surface, as shown in table 3 below.
Table 3
Wherein, unqualified subtype member " leakage is labeled ", " patch partially ", " fold ", " defective " use A1, A2, A3, A4 respectively
It indicates.Time member " 8：00-10:00”、“10:00-12:00”、“13:00-15:00”、“15:00-17:00 " respectively use B1,
B2, B3, B4 are indicated.Station number member 1,2,3 is indicated with C1, C2, C3 respectively.Age member " 20-30 ", " 30-40 ", " 40-
50 " are indicated with D1, D2, D3 respectively.Gender member " man " " female " is indicated with E1, E2 respectively.
Member in being tieed up according to unqualified type to five obtained dimension data cubes carries out OLAP sectionings, obtains every
Four-dimensional subdata cube corresponding to a unqualified type dimension member.
Subdata cube body surface in the case of A1 is shown in Table 4.Subdata cube body surface in the case of A2 is shown in Table 5.In the case of A3
Subdata cube body surface be shown in Table 6.Subdata cube body surface in the case of A4 is shown in Table 7.
Table 4
Table 5
Table 6
Table 7
As described in step S2, by each subdata cube running body FP-Growth algorithms.If support threshold min_sup
=1000；
Input：Subdata cube, support threshold min_sup
Output：The cubical frequent predicate set of subdata
(1) FP-tree is constructed according to the following steps
(a) subdata cube is scanned for the first time, calculates the support of each predicate, by successively decreasing, sequence sorts, with
Min_sup relatively obtains frequent 1- item collections, abandons nonmatching grids.
(b) second of scanning subdata cube, creates FP-tree.The root node for creating FP-tree first, is labeled as
" null " reads in the record on subdata cube, and creates a branch to each record successively, increases when for a record
When branch, the node counts on overlay path increase corresponding count values, and nonoverlapping part creates new node and create chain
Connect direction prefix.Until all records are mapped on the paths FP-tree.
(2) excavation of FP-tree is realized by calling process Fp-growth (FP-tree, null).
// the process is realized as follows
procedure Fp-growth(tree,α)
1) if tree include single path P then
2) each combination of nodes (being denoted as β) of for path Ps
3) pattern β ∪ α, the minimum support of support support=β interior joints are generated；
4)else for each aiTree head
5) pattern β=a is generatedi∪ β, support support=ai。support
6) conditional pattern base of structural model β, and construct the condition FP-Tree of β
7)ifthen
8) Fp-growth (tree, β) is called；}
The cubical frequent predicate set of each subdata such as the (son in the case of A1 of table 8 are obtained after operation FP-Growth algorithms
The frequent predicate set that data cube generates), table 9 (frequent predicate set that the subdata cube in the case of A2 generates), table 10
(frequent predicate set that the subdata cube in the case of A3 generates), (frequency that the subdata cube in the case of A4 generates of table 11
Numerous predicate collection) shown in.
Table 8
Table 9
Table 10
Table 11
As described in step S3, the cubical frequent predicate set of multiple subdatas that above-mentioned steps obtain is merged into data and is stood
The frequent predicate set of cube.As shown in figure 3, the cubical frequent predicate set list of each subdata is traversed successively, if different frequent
There are identical frequent predicate sets for the list of predicate collection, then are added the support of the frequent predicate set, obtain only comprising unqualified
The frequent predicate set and its support of factor；The cubical frequent predicate set of each subdata is added into corresponding unqualified type
New frequent predicate set is formed, that is, includes the frequent predicate set and its support of unqualified factor and unqualified type.Five dimensions
The frequent predicate set list of data cube is shown in Table 12.
Table 12
As described in step S4, the min confidence min_conf=0.6 of setting, according to formula
Calculate the confidence level of current incidence relation.If being more than setting most by the confidence level conf that above formula is calculated
Small confidence level min_conf, it can be said that this bright unqualified type has stronger relationship with the relation factor.
Strong incidence relation excavation is carried out according to above step, the results are shown in Table 13.
Table 13
Example the above is only the implementation of the present invention is not intended to limit the scope of the invention.It is every according to this hair
Bright announcement, any improvement made to embodiment and modification all should be within protection scope of the present invention.
Claims (5)
1. a kind of mobile phone based on FP-Growth algorithms labels detection data analysis method, it is characterised in that：This method include with
Lower step：
S1 pre-processes mobile phone labeling detecting system testing result database；It is specific as follows：
S11 rejects the qualified correlated results of mobile phone labeling detection, retains mobile phone labeling and detect underproof correlated results；
S12, extract labelled in testing result underproof type and relevant labeling personnel age, gender, station number and
Label the information such as time；
S13 handles the data of extraction, specifically includes whole data record deletion to missing values and exceptional value；
S14, by treated, data pass through OLAP data modeling tool Workbench one five dimension data cube of generation；
S15, the member in being tieed up according to unqualified type to five obtained dimension data cubes carry out OLAP sectionings, obtain every
Four-dimensional subdata cube corresponding to a unqualified type dimension member；
S2 identifies the cubical frequent meaning of each subdata to each subdata cube application FP-Growth algorithms statistics
Word set；
The statistics identification step of the cubical frequent predicate set of subdata based on FP-Growth algorithms is as follows：
S21 sets the minimum support of FP-Growth algorithms；
S22 scans subdata cube for the first time, calculates the support of each dimension member, and with the minimum support of setting ratio
Relatively obtain 1- frequent predicate sets；
S23, second of scanning subdata cube, creates FP-tree；
S24 excavates the cubical frequent predicate set of subdata from FP-tree；
The cubical frequent predicate set of multiple subdatas that above-mentioned steps obtain is merged into the frequent predicate of data cube by S3
Collection；
S31 traverses the cubical frequent predicate set list of each subdata, if the cubical frequent predicate set list of subdata is deposited
In identical frequent predicate set, then the support of the frequent predicate set is added, obtains the frequent meaning for only including unqualified factor
Word set；
The cubical frequent predicate set of each subdata is formed new frequent predicate set by S32 plus corresponding unqualified type,
It include the frequent predicate set of unqualified factor and unqualified type；
After obtaining the frequent predicate set of data cube, the pass of unqualified type b and unqualified factor a are judged by calculating by S4
System is strong and weak.
2. a kind of mobile phone based on FP-Growth algorithms according to claim 1 labels detection data analysis method, special
Sign is：Data cube is not conformed to by labelling personnel's age, labeling personnel gender, time, station and labeling in the step S14
The dimension composition of lattice type five；Wherein, it labels unqualified type dimension there is leakage label, inclined patch, fold, defective four members.
3. a kind of mobile phone based on FP-Growth algorithms according to claim 1 labels detection data analysis method, special
Sign is：FP-tree is created in the step S23, and the specific method is as follows：The root node for creating FP-tree first, is labeled as
" null " then reads in the record on subdata cube, and creates a branch to each record, increases when for a record
When branch, the node counts on overlay path increase corresponding count values, and nonoverlapping part creates new node and create chain
Connect direction prefix；Until all records are mapped on the paths FP-tree.
4. a kind of mobile phone based on FP-Growth algorithms according to claim 1 labels detection data analysis method, special
Sign is：The step S24 excavates the cubical frequent predicate set of subdata from FP-tree, is as follows：
S241 constructs the conditional pattern base of 1- frequent predicate sets, assigns conditional pattern base as transaction set structural environment FP-
tree；
S242 finds condition frequent predicate set according to condition FP-tree, then merges with suffix pattern, obtain frequent predicate set；
S243, the iterative step S241 and step S242 on condition FP-tree, until setting comprising a predicate, to unite
Meter identifies the cubical frequent predicate set of subdata.
5. a kind of mobile phone based on FP-Growth algorithms according to claim 1 labels detection data analysis method, special
Sign is：The step S4 is as follows：
S41, setting min confidence min_conf；
S42 is calculated by the following formula the confidence level conf of unqualified type b and unqualified factor a：
Wherein, num (a) is the support of the frequent predicate set of only unqualified factor a compositions；Num (a | b) be by it is unqualified because
The support of the frequent predicate set of plain a and unqualified type b compositions；
S43, judges whether confidence level conf is more than the min confidence min_conf of setting, if so, thinking the unqualified class
It is unqualified because being known as strong incidence relation described in type b and a；
S44, traverses all frequent predicate sets, and result of calculation exports after being arranged from high to low according to confidence level.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810174321.1A CN108346007A (en) | 2018-03-02 | 2018-03-02 | A kind of mobile phone labeling detection data analysis method based on FP-Growth algorithms |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810174321.1A CN108346007A (en) | 2018-03-02 | 2018-03-02 | A kind of mobile phone labeling detection data analysis method based on FP-Growth algorithms |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108346007A true CN108346007A (en) | 2018-07-31 |
Family
ID=62958949
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810174321.1A Pending CN108346007A (en) | 2018-03-02 | 2018-03-02 | A kind of mobile phone labeling detection data analysis method based on FP-Growth algorithms |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108346007A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109299197A (en) * | 2018-08-07 | 2019-02-01 | 威富通科技有限公司 | A kind of partition method of data cube, system and terminal device |
| CN110516704A (en) * | 2019-07-19 | 2019-11-29 | 中国科学院地理科学与资源研究所 | A kind of MLKNN multi-tag classification method based on correlation rule |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2007124616A1 (en) * | 2006-04-27 | 2007-11-08 | Intel Corporation | System based method for content-based partitioning and mining |
| 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 |
| CN101739818A (en) * | 2009-11-19 | 2010-06-16 | 北京世纪高通科技有限公司 | Method and device for data mining of road traffic accident based on association rule |
| CN105159952A (en) * | 2015-08-14 | 2015-12-16 | 安徽新华博信息技术股份有限公司 | Data processing method based on frequent item set mining |
| CN106056222A (en) * | 2016-05-27 | 2016-10-26 | 大连楼兰科技股份有限公司 | FP-Tree sequence pattern mining-based fault diagnosis and evaluation method |
| CN107578149A (en) * | 2017-08-16 | 2018-01-12 | 国网浙江省电力公司经济技术研究院 | A kind of power grid enterprises' critical data analysis method |
-
2018
- 2018-03-02 CN CN201810174321.1A patent/CN108346007A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2007124616A1 (en) * | 2006-04-27 | 2007-11-08 | Intel Corporation | System based method for content-based partitioning and mining |
| 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 |
| CN101739818A (en) * | 2009-11-19 | 2010-06-16 | 北京世纪高通科技有限公司 | Method and device for data mining of road traffic accident based on association rule |
| CN105159952A (en) * | 2015-08-14 | 2015-12-16 | 安徽新华博信息技术股份有限公司 | Data processing method based on frequent item set mining |
| CN106056222A (en) * | 2016-05-27 | 2016-10-26 | 大连楼兰科技股份有限公司 | FP-Tree sequence pattern mining-based fault diagnosis and evaluation method |
| CN107578149A (en) * | 2017-08-16 | 2018-01-12 | 国网浙江省电力公司经济技术研究院 | A kind of power grid enterprises' critical data analysis method |
Non-Patent Citations (2)
| Title |
|---|
| 王娟: "基于FP-Growth算法改进的多层次关联规则挖掘算法", 《电脑知识与技术》 * |
| 阎星娥 等: "OLAP中基于FP-增长的关联规则挖掘", 《计算机科学》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109299197A (en) * | 2018-08-07 | 2019-02-01 | 威富通科技有限公司 | A kind of partition method of data cube, system and terminal device |
| CN110516704A (en) * | 2019-07-19 | 2019-11-29 | 中国科学院地理科学与资源研究所 | A kind of MLKNN multi-tag classification method based on correlation rule |
| CN110516704B (en) * | 2019-07-19 | 2022-07-15 | 中国科学院地理科学与资源研究所 | MLKNN multi-label classification method based on association rule |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110245496B (en) | Source code vulnerability detection method and detector and training method and system thereof | |
| CN107609052A (en) | A kind of generation method and device of the domain knowledge collection of illustrative plates based on semantic triangle | |
| CN107688748A (en) | Fragility Code Clones detection method and its device based on leak fingerprint | |
| CN109214191A (en) | A method of utilizing deep learning forecasting software security breaches | |
| CN109492106B (en) | Automatic classification method for defect reasons by combining text codes | |
| CN111459799A (en) | Software defect detection model establishing and detecting method and system based on Github | |
| CN103902591A (en) | Decision tree classifier establishing method and device | |
| CN105677640A (en) | Domain concept extraction method for open texts | |
| CN108959395A (en) | A kind of level towards multi-source heterogeneous big data about subtracts combined cleaning method | |
| CN110287329A (en) | A kind of electric business classification attribute excavation method based on commodity text classification | |
| CN108346007A (en) | A kind of mobile phone labeling detection data analysis method based on FP-Growth algorithms | |
| CN107291877A (en) | A kind of Mining Frequent Itemsets based on Apriori algorithm | |
| CN109325062A (en) | A kind of data dependence method for digging and system based on distributed computing | |
| CN105844398A (en) | PLM (product life-cycle management) database-based mining algorithm for DPIPP (distributed parameterized intelligent product platform) product families | |
| CN103793589B (en) | High-speed train fault handling method | |
| CN111814528B (en) | Connectivity analysis noctilucent image city grade classification method | |
| CN109858025A (en) | A kind of segmenting method and system of Address Standardization corpus | |
| CN113487211A (en) | Nuclear power equipment quality tracing method and system, computer equipment and medium | |
| CN115438199A (en) | Knowledge platform system based on smart city scene data middling platform technology | |
| CN114841789A (en) | Block chain-based auditing and auditing pricing fault data online editing method and system | |
| CN107590225A (en) | A kind of Visualized management system based on distributed data digging algorithm | |
| CN106411704A (en) | Distributed junk short message recognition method | |
| CN105224603A (en) | Corpus acquisition methods and device | |
| CN114116957A (en) | Electromagnetic information leakage intelligent analysis method based on knowledge graph | |
| CN114706558A (en) | K-TRUSS-based complex software key module identification method |
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 | ||
| AD01 | Patent right deemed abandoned |
Effective date of abandoning: 20220429 |
|
| AD01 | Patent right deemed abandoned |