CN108428200A - A kind of the electric business field patent infringement decision-making system and determination method of case-based reasioning - Google Patents
A kind of the electric business field patent infringement decision-making system and determination method of case-based reasioning Download PDFInfo
- Publication number
- CN108428200A CN108428200A CN201810217918.XA CN201810217918A CN108428200A CN 108428200 A CN108428200 A CN 108428200A CN 201810217918 A CN201810217918 A CN 201810217918A CN 108428200 A CN108428200 A CN 108428200A
- Authority
- CN
- China
- Prior art keywords
- case
- similarity
- attribute
- information
- infringement
- 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
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/18—Legal services; Handling legal documents
- G06Q50/184—Intellectual property management
-
- 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
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
Abstract
The present invention discloses a kind of the electric business field patent infringement decision-making system and infringement determination method of case-based reasioning, the system building includes that user logs in and user right control module, management module of deciding a case and intelligent decision module on case judgement cloud computing center；The infringement determination method includes that structure patent judges case library, and each patent in library is labeled, the new judgement case of input and model parameter information, new judgement case is put into patent judgement case library and is retrieved, then use is for statistical analysis to retrieval result based on weight integral model, provides the instruction of new judgement case automatically.The system and method for the present invention alleviate the work load for the personnel of deciding a case, main the case where manually judging the patent infringement of electric business field at present is changed, working efficiency is improved, increases the foundation decided a case, so that flow of deciding a case is more intelligent, the transparency of electric business field patent infringement judgement is promoted.
Description
Technical field
The present invention relates to electric business field patent infringement judgements, and in particular to a kind of electric business field patent of case-based reasioning
Decision-making system of encroaching right and determination method.
Background technology
Universal with computer and internet, e-commerce has obtained significant progress, more and more people in China
Get used to through third party transaction gondola sales and the free choice of goods.However, network is while so that people is enjoyed convenient to the full,
A large amount of intellectual property protection problem is brought, all kinds of IP disputes of the outstanding behaviours in e-commerce transaction are increasingly
It is more.In recent years, in the intellectual property case that judicial department and administrative department accept and investigate and prosecute, it is related to network trading increasingly
It is more.
Zhejiang is the big province of E-commerce market, and the whole province's e-commerce total turnover in 2013 breaks through 1.6 trillion yuans, increases on year-on-year basis
It is about 30%, accounts for the whole nation 1/6th；Existing all kinds of on-line shops of the whole province more than 130 ten thousand realize 3821.25 hundred million yuan of e-tailing, together
Than increasing by 88.48%, it is equivalent to the 25% of the whole province's total retail sales of consumer goods, total amount accounts for the whole nation 1/5th.Wherein, network
Retail sales surpasses hundred million yuan of families more than 100 of enterprise, has emerged the well-known net quotient both at home and abroad such as Alibaba, Taobao, Yiwu purchase, electronics quotient
Business number of the enterprise and scale occupy the leading place in the whole country.What is particularly worth mentioning is that Taobao's on November 11st, 2013 make it is " double
11 " net purchase sections, odd-numbered day turnover reach 350.19 hundred million yuan, strike a bargain 1.7 hundred million.
But the tort of intellectual property rights of e-commerce field personation phenomenon also emerges one after another, to third party transaction platform operation
For person, the IP dispute complaint being connected to every year rises year by year, is counted according to being investigated to group of Alibaba, Alibaba
It is 1,050,000 that the tort of intellectual property rights of processing in 2010, which complains commodity, and 2011 are 910,000, and 2012 are 940,000；It is washing in a pan
Precious online, it is 8,700,000 that the tort of intellectual property rights of processing in 2010, which complains commodity, and 2011 are 5,700,000, and 2012 are 680
Ten thousand.By 2013, the IP dispute of Taobao year processing was complained just up to 8,610,000, and 1,410,000 person-times of member is punished.
Wherein, trade mark accounts for 52%, and copyright accounts for 40%, and patent accounts for 6%, other account for 2%.Alibaba's Chinese and English transaction platform
(1688.com) handles tort of intellectual property rights for 2013 and complains more than 1,110,000 to rise, and wherein trade mark accounts for 60%, and patent accounts for 30%, copyright
Account for 10%.By the analysis of complaint amount, we are not difficult to find out, although two e-commerce platform quotient are in terms of intellectual property protection
It steps up its investment, the trend being growing on and on is presented in intellectual property complaint amount.In terms of the content that intellectual property is complained, Alibaba is flat
IP dispute on platform is concentrated mainly in Patent ＆ Trademark；And the IP dispute on Taobao's platform is concentrated mainly on
On trade mark and copyright.
Obviously, if the intellectual property illegal activities being directed to not in time in E-commerce market transaction carry out Valid Regulation, gesture
It must serious infringement patentee interests, damage consumers' rights and interests, the sound development for threatening transaction platform and e-commerce industry.
Implement business EC is information-based and paperless trading, the carrier that he is carried are networks.However, network trading
Main body anonymity, space virtual, the non-promptness merchandising and complete a business transaction, make existing intellectual property law system in network sky
Between it is applicable receive great impact, from dynamic then million grades of the case load of the international station of Alibaba's Chinese and English and Taobao and
The reasons such as the complexity of case, far beyond the ability purely manually handled.Whoever started the trouble should end it, so e-commerce
The protection of domain knowledge property right infringement also needs to the advanced information technologies such as the calculating of introducing big data, artificial intelligence and Knowledge Discovery and adds
To solve, patent of the present invention is exactly to be proposed under this background.
Patent 02135226.7 " a kind of IC card for requesting and approving building engineerings manages system and its application process " carries
A kind of IC card for requesting and approving building engineerings has been supplied to manage system and its application process, the method achieve reports to build hall work
The digital information hardware and software platform of Cheng Jianshe data realizes information sharing and the business linkage of different business properties window, makes construction
The paper carrier data that unit need to be submitted is greatly reduced, help to improve the office automation of government department, with no paper degree and
Report builds " sunlight " of examination and approval work, is the quantum jump of the specific business processing working procedure of government administration section, is to build item
One important embodiment of the mesh reform of the system of administrative approval and an innovation of urban construction administration mechanism.The invention is suitable for engineering
The report of construction project builds, examines, management work.Patent 201010288981.6 " intelligent Examination and approval system and method " provides
A kind of intelligent Examination and approval system and method that can be used in urban construction case examination ＆ approval field, the system include：Examine knowledge
Library, is used for stored knowledge point, and each knowledge point includes in examination ＆ approval concern information and at least partly described knowledge point and operation system
Business datum it is associated；Operating procedure administrative unit, for creating business approval operating procedure, partial service review operation step
Rapid includes the knowledge point in examination ＆ approval knowledge base；Approval process administrative unit, for realizing approval process in each examination ＆ approval executor
Between automatic flow, each examination ＆ approval executor executes business approval operating procedure successively.The invention also provides a kind of corresponding
Intelligent Examination and approval method, according to preset operation by the predetermined registration operation step before being loaded into operation system and in loading system
Step creates approval process, increases the flexibility of operation system.Both methods mainly handles electricity by the way of workflow
Sub- government affairs case examination and approval work, and electronization and informationization that case is examined are realized based on " examination ＆ approval rule base ".But these
Method and system is storage and inquiry and the electronic approval function for realizing examination ＆ approval data, these is not made full use of to go through
History examines data, does not more carry out data analysis or excacation to these data, therefore there is no fundamentally improve to examine
Core efficiency.
Invention content
The purpose of the present invention is overcoming deficiency in the prior art, a kind of electric business field patent of case-based reasioning is provided
Decision-making system of encroaching right and determination method, it is specific as follows：
A kind of electric business field patent infringement decision-making system of case-based reasioning, which is characterized in that the system building is on record
Part judges on cloud computing center that the center uses the Hadoop cluster platforms based on (SuSE) Linux OS, the Platform deployment
Tomcat servers and Hadoop clusters, HBase clusters, Zookeeper clusters and SolrCloud clusters；The system packet
Include following module：
(1) user logs in and user right control module：User of the module for all kinds of roles logs in, and to different
Role sets necessary permission；
(2) it decides a case management module：The module is used for storage, inquiry, retrieval and the maintenance of each item data, each item number
According to including each patent case classification, controlled patent information, controlled patent characteristic information, information of deciding a case, case weight and
With number；
(3) intelligent decision module：For the module for completing reasoning by cases and interpretation of result, which includes reasoning by cases list
Member, Similar case search results management unit, similar cases statistical analysis unit and distributed full-text search unit.
A kind of electric business field patent infringement determination method of case-based reasioning, which is characterized in that this method is based on right
It is required that system described in 1 realizes that the method includes the following steps：
(1) case library is judged using the case representation method structure patent based on object-oriented, and marks each patent
Case classification, controlled patent information, controlled patent characteristic information and information of deciding a case；
(2) input newly judges case and model parameter information, and marks the case classification information of new judgement case, is controlled specially
Sharp information, controlled patent characteristic information and CBR model parameter information, the CBR model parameter information include approximate case
Number K, case attribute weight vector W and similarity retrieval threshold value a；
(3) operation is submitted to Hadoop cluster platforms and carries out KNN MapReduce Case Retrievals；
(4) when the similar cases retrieved cannot meet the number K and similarity retrieval threshold value a of approximate case simultaneously,
A or W is then adjusted, is then retrieved again, until the similar cases retrieved while the number K and similarity that meet approximate case
Retrieval threshold a；
(5) it is for statistical analysis to retrieval result to be based on " weight integral model "：Using different type attributes similarity meter
Calculation method calculates each case in case library and the similarity between new input case, then further according to similarity and K values
Match satisfactory similar cases；According to its similarity size different weights is set, similarity is bigger, then the case
Weight is bigger；Then summation is weighted according to the judgement result of each case, judgement result, which is belonged to of a sort case, puts
The cumulative of weight is carried out together, that maximum a kind of result of deciding a case of final weight is anticipated as the reference of this Case Retrieval
See；The case judgement result classification includes " infringement is set up ", " data is uneven undetermined ", and " infringement is invalid " " hands in grade to wait for
It is fixed " four classes；
(6) when the case that the retrieval in step (4) returns is still undesirable, then by inputting keyword in case
Full-text index is carried out in library, which includes each field of case and the material that prosecutor is submitted；
(7) retrieval result is evaluated and is corrected：When some case in retrieval result solve it is current to be determined
The patent problem of case then increases the weight of the case retrieved.
Preferably, the step (1) is specially：
It is decided a case data according to history, by each four-tuple that case representation is C=(E, P, L, R) of deciding a case, wherein E=
(e1,e2,...,em) description vectors of case classification information, including two class of patent infringement disputes and Holiday culture are represented, it is described
Patent infringement disputes include patent infringement disputes, Granting of patent right does not pay appropriate expense using invention after application for a patent for invention
With four kinds of situations such as dispute, right to patent and the dispute of the patent right of attribution, inventor's designer's qualification dispute；The personation is special
Profit include on product or product packaging affixing patent indication, be labelled with patent indication in sale, marked on product description
Three kinds of situations such as patent；P=(p1,p2,...,pn) represent be controlled patent information description vectors, including patent type, it is case-involving specially
Four kinds of profit number, the affiliated industry of product and obligee；L=(l1,l2,...,lp) description vectors for being controlled patent characteristic information are represented,
Including technical field, the field character pair, Patent right requirement content；R=(r1,r2,...,rq) represent retouching for information of deciding a case
State vector, the information of deciding a case includes basic merit, result of deciding a case and is decided a case the time.
Preferably, the Case Retrieval in the step (3) uses the KNN case retrieval algorithms based on MapReduce.
Preferably, the different type attributes similarity computational methods in the step (5) are：
S1：Calculate each case in case library and the similarity between the new input single attribute of case, specific calculating side
Formula is as follows：
(1) similarity calculation of continuous type numerical attribute, calculation formula are as follows：
S indicates that original bill example attribute, t indicate that the identical attribute of target case in case library, max (c, t) indicate c and t institute's generations
The maximum value of the codomain of the attribute of table, min (s, t) indicate the minimum value of the codomain for the attribute that s and t is represented；
(2) the orderly calculating of the similarity of attribute：Orderly attribute is reduced to order enumeration type first, and according to semanteme
Strong and weak order is arranged, it is assumed that attribute is divided into n grade, then the calculating formula of similarity between grade i and grade j is such as
Under：
Wherein ord (i) is order of the attribute value i in codomain set；cardiIt is the series of attribute point；
(3) similarity calculation of character type attribute：The similarity calculation of character type attribute uses of MMSEG Chinese word segmentations
With algorithm：Specific calculating formula of similarity is as follows：
Wherein, Stringtoken () function is the participle array obtained using the matching algorithm of MMSEG Chinese word segmentations,
Same () function calculates the number of synonymous word after two character string participles, and maxlen () calculates of longest character string participle
Number；
(4) similarity calculation of fuzzy interval attribute：
The first step：Membership function is constructed according to fuzzy interval.
Second step：Two fuzzy intervals and its overlapping interval corresponding area, the weight of area are calculated separately according to membership function
Folded similarity of the rate as fuzzy interval, the calculation formula of similarity are as follows：
Wherein, SiIndicate that one of fuzzy interval i attributes pass through the calculated corresponding area of membership function, TiShow another
By the calculated corresponding area of membership function, S indicates to pass through the calculated corresponding surface of membership function a fuzzy interval i attributes
Product；
S2：Calculate each case in case library and the similarity between the case of new input case.Calculation formula is as follows：
Wherein, wkThe weights of k-th of feature in case characteristic vector are represented,aSkIndicate k-th of spy of case S
The weights of sign, aTkThe value of k-th of feature of case T, sim (a are indicated respectivelySk,aTk) be case k-th of feature of S and T phase
Like degree.
The beneficial effects of the invention are as follows：The present invention proposes a kind of electric business field patent infringement judgement of case-based reasioning
System and determination method judge cloud computing center by case, and distributed inspection is carried out using the MapReduce frames of Hadoop
Rope establishes the distributed Case Retrieval model of case-based reasioning technology.The present invention innovatively proposes " weight integral model "
It is for statistical analysis to the similar cases retrieved, and then obtain the guidance beneficial to newly judging case.Meanwhile this cloud computing
The deployment of system also mitigates the work load for the personnel of deciding a case so that they only need terminal that can be networked can be with
When realize that intelligence is decided a case everywhere, change the case where manually judge the patent infringement of electric business field main at present, raising work
Make efficiency, increase the foundation decided a case so that flow of deciding a case is more intelligent, promotes the transparent of electric business field patent infringement judgement
Degree.
Description of the drawings
Fig. 1 is the Hadoop collection for the electric business field patent infringement determination method for realizing case-based reasioning proposed by the present invention
Gang fight composition；
Fig. 2 is the net of the electric business field patent infringement judgement cloud computing system for the case-based reasioning technology that the present invention realizes
Network topological structure figure；
Fig. 3 is the MapReduce distributed structure/architecture figures for the Case Retrieval that the present invention describes；
Fig. 4 is the electric business field patent infringement determination method flow chart of case-based reasioning technology proposed by the present invention.
Specific implementation mode
Below according to attached drawing and the preferred embodiment detailed description present invention, the objects and effects of the present invention will become brighter
In vain, below in conjunction with drawings and examples, the present invention will be described in further detail.It should be appreciated that described herein specific
Embodiment is only used to explain the present invention, is not intended to limit the present invention.
A kind of electric business field patent infringement decision-making system of case-based reasioning, the system building judge cloud computing in case
On center, which uses Hadoop cluster platforms based on (SuSE) Linux OS, Platform deployment Tomcat servers with
And Hadoop clusters, HBase clusters, Zookeeper clusters and SolrCloud clusters；The system includes following module：
(1) user logs in and user right control module：User of the module for all kinds of roles logs in, and to different
Role sets necessary permission；
(2) it decides a case management module：The module is used for storage, inquiry, retrieval and the maintenance of each item data, each item number
According to including each patent case classification, controlled patent information, controlled patent characteristic information, information of deciding a case, case weight and
With number；
(3) intelligent decision module：For the module for completing reasoning by cases and interpretation of result, which includes reasoning by cases list
Member, Similar case search results management unit, similar cases statistical analysis unit and distributed full-text search unit.
In case judgement cloud computing in the electric business field patent infringement decision-making system of the case-based reasioning of the present invention
Heart platform has used 4 PC machine, and model is Dell Precision WorkStation T3400, monokaryon CPU, 4G memory,
500G hard disks.Wherein one installation Window7 steerable system installs Linux CentOS6.4 as exploitation host, excess-three platform
Operating system is as working cluster.By three PC Hadoop aggregated structures formed as shown in Figure 1, specific network topological diagram such as
Shown in 2.
A kind of electric business field patent infringement determination method of case-based reasioning, as shown in figure 4, this method is wanted based on right
The system described in 1 is asked to realize, the method includes the following steps：
(1) case library is judged using the case representation method construct patent based on object-oriented, and marks each patent
Case classification, controlled patent information, controlled patent characteristic information and information of deciding a case；
Use Hive and Pig that the data of these multi-data sources have been carried out with pretreatment and the ETL operations of data first, most
Throughout one's life at the case of deciding a case of above structure, and it is stored in the form of a table in HBase databases.
It is decided a case data according to history, by each four-tuple that case representation is C=(E, P, L, R) of deciding a case, wherein E=
(e1,e2,...,em) description vectors of case classification information, including two class of patent infringement disputes and Holiday culture are represented, it is described
Patent infringement disputes include patent infringement disputes, Granting of patent right does not pay appropriate expense using invention after application for a patent for invention
With four kinds of situations such as dispute, right to patent and the dispute of the patent right of attribution and inventor's designer's qualification dispute；The personation
Patent include on product or product packaging affixing patent indication, be labelled with patent indication in sale and in product description subscript
Note three kinds of situations such as patent；P=(p1,p2,...,pn) description vectors for being controlled patent information are represented, including it is patent type, case-involving
Four kinds of the affiliated industry of the patent No., product and obligee；L=(l1,l2,...,lp) represent be controlled the description of patent characteristic information to
Amount, including technical field, the field character pair and Patent right requirement content (and not corresponded in following table)；R=
(r1,r2,...,rq) representing the description vectors of information of deciding a case, the information of deciding a case includes basic merit, result of deciding a case and is decided a case
Time (and not corresponded in following table), as shown in table 1.
1 patent of table judges the judgement vector and its Judging index of each patent in case library
(2) input newly judges case and model parameter information, and marks the case classification information of new judgement case, is controlled specially
Sharp information, controlled patent characteristic information and CBR model parameter information, the CBR model parameter information include approximate case
Number K, case attribute weight vector W and similarity retrieval threshold value a；
(3) by operation be submitted to Hadoop cluster platforms carry out KNN MapReduce Case Retrievals, using based on
The KNN case retrieval algorithms of MapReduce.
KNN case retrieval algorithms based on MapReduce are realized in MapReduce distributed computing platforms, are closed
Key is the design of map functions, reduce functions and jobCreate functions.Map functions, which are mainly responsible for, searches HBase sublists
Local k similarity meet similarity requirement case.Reduce functions are responsible for summarizing the output result of map functions and be generated
The global K final cases for meeting similarity requirement.JobCreate () function is used for completing user about job run
It custom-configures and is submitted in cluster and run.Mapreduce operation associated class figures are as shown in Figure 3.
The judgement case of the present invention is stored in HBase, therefore the InputFormat of Mapreduce operations is set as
TableInputFormat.Since the case being retrieved will be also stored in the interim table of HBase, therefore Mapreduce
The OutputFormat of operation is set as TableOutputFormat.It is basis when Hadoop is using HBase table as inputting
The Region data of HBase table divide Split, i.e. each Region corresponds to a Split, thus also corresponds to one
Mapper.It is TableInputFormat by the way that InputFormat is arranged, Mapper divides each Region according to rowKey
At<key,value>Right, key corresponds to each rowKey of the sublist, and value corresponds to data that the row is included (in class figure
For Result).SearchKNNCaseMapper is inherited from TableMapper<Text,DoubleWritable>, thus may be used
Directly to handle the data in HBase table.SearchKNNCaseReducer is inherited from TableReducer<Text,
DoubleWritable>, the output result of reduce functions can be thus written in HBase table.
SearchKNNCaseDriver is responsible for configuring distributed operation cluster environment, generates Mapreduce operations and be submitted in cluster
It executes.SearchKNNCaseUtils classes provide some tool functions, for example calculate the similarity etc. between two cases.
The major function of Mapper is to find out K local case for meeting similarity threshold and according to the big float of similarity
Sequence is then transferred in Reducer and handles.
The major function of Reducer is to summarize the output of each Mapper as a result, and being carried out according to the size of similarity value
It is exported after sequence.The output of all Mapper is stored in a HashMap container by the Reducer in this system, utilizes profit
HashMap containers are ranked up with TreeMap, K case before exporting.After realizing map functions and reduce functions,
Also need to the operation information of setting Mapreduce operations.JobCreate () function in figure is exactly to be used for being arranged one
User is returned to after Mapreduce operations.Main setting information include the JobTracker host ips of job run, operation name
Realization class name, the realization class name of Reducer, InputFormat formats, the OutputFormat lattice of title, JAR class names, Mapper
Position of input data and output data of formula and operation etc..This system uses HBase table outputting and inputting as operation
Position, therefore set InputFormat to TableInputFormat, it sets OutputFormat to
TableOutputFormat so that HBase combinations Mapreduce carries out distributed data processing.Configure operation, so that it may with
Operation is submitted in cluster and goes to run, key code is as follows：
(4) when the similar cases retrieved cannot meet the number K and similarity retrieval threshold value a of approximate case simultaneously,
A or W is then adjusted, is then retrieved again, until the similar cases retrieved while the number K and similarity that meet approximate case
Retrieval threshold a；
(5) different type attributes similarity computational methods is used to calculate each case in case library and new input case
Between similarity, then match satisfactory similar cases further according to similarity and K values；It is set according to its similarity size
Different weights is set, similarity is bigger, then the weight of the case is bigger；Then it is weighted according to the judgement result of each case
Summation will judge that result belongs to of a sort case and puts together and carries out the cumulative of weight, and final weight is maximum, and that is a kind of
Advisory opinion of the result of deciding a case as this Case Retrieval；The case judgement result classification includes " infringement is set up ", " data
It is uneven undetermined ", " infringement is invalid ", " it is undetermined to hand in grade " four class；
Wherein, different type attributes similarity computational methods are：
S1：Calculate each case in case library and the similarity between the new input single attribute of case, specific calculating side
Formula is as follows：
(1) similarity calculation of continuous type numerical attribute, calculation formula are as follows：
S indicates that original bill example attribute, t indicate that the identical attribute of target case in case library, max (c, t) indicate c and t institute's generations
The maximum value of the codomain of the attribute of table, min (s, t) indicate the minimum value of the codomain for the attribute that s and t is represented；
(2) the orderly calculating of the similarity of attribute：Orderly attribute is reduced to order enumeration type first, and according to semanteme
Strong and weak order is arranged, it is assumed that attribute is divided into n grade, then the calculating formula of similarity between grade i and grade j is such as
Under：
Wherein ord (i) is order of the attribute value i in codomain set；cardiIt is the series of attribute point；
(3) similarity calculation of character type attribute：The similarity calculation of character type attribute is calculated using the matching based on participle
Method：Specific calculating formula of similarity is as follows：
Wherein, Stringtoken () function is the participle array obtained using the matching algorithm of MMSEG Chinese word segmentations,
Same () function calculates the number of synonymous word after two character string participles, and maxlen () calculates of longest character string participle
Number；
(4) similarity calculation of fuzzy interval attribute：
The first step：Membership function is constructed according to fuzzy interval.
Second step：Two fuzzy intervals and its overlapping interval corresponding area, the weight of area are calculated separately according to membership function
Folded similarity of the rate as fuzzy interval, the calculation formula of similarity are as follows：
Wherein, SiIndicate that one of fuzzy interval i attributes pass through the calculated corresponding area of membership function, TiShow another
By the calculated corresponding area of membership function, S indicates to pass through the calculated corresponding surface of membership function a fuzzy interval i attributes
Product；
S2：Calculate each case in case library and the similarity between the case of new input case.Calculation formula is as follows：
Wherein, wkThe weights of k-th of feature in case characteristic vector are represented,aSkIndicate k-th of spy of case S
The weights of sign, aTkThe value of k-th of feature of case T, sim (a are indicated respectivelySk,aTk) be case k-th of feature of S and T phase
Like degree.
(6) when in step (4) retrieval return case it is undesirable, then by input keyword in case library into
Row full-text index, the index include each field of case and the material that prosecutor is submitted；
(7) retrieval result is evaluated and is corrected：When some case in retrieval result solve it is current to be determined
The patent problem of case then increases the weight of the case retrieved；Similar cases are returned when no in step (4),
For example, the personnel of deciding a case can input keyword " U.S. face beautiful clothing ", keyword is submitted to case and judges cloud meter by system
In the SolrCloud clusters of calculation system, then starts Solr distribution full-text search tasks, finally retrieval result is returned in time
To the personnel of deciding a case, the case of return may be that history judges that land use situation investigation field contains " U.S. face " or " beautiful clothing " in case
Case, it is also possible to include the case of these keywords in the proprietary material of submission, be contained in some fields of these cases
The interested information of the personnel that decide a case, therefore some enlightenments of the personnel of deciding a case and reference can be given.Personnel decide a case according to full-text search
The detailed judgement information of case out obtains the guidance for helping to solve current problem.
It will appreciated by the skilled person that the foregoing is merely the preferred embodiment of invention, it is not used to limit
System invention, although invention is described in detail with reference to previous examples, for those skilled in the art, still
It can modify to the technical solution of aforementioned each case history or equivalent replacement of some of the technical features.It is all
Within the spirit and principle of invention, modification, equivalent replacement for being made etc. should be included within the protection domain of invention.
Claims (5)
1. a kind of electric business field patent infringement decision-making system of case-based reasioning, which is characterized in that the system building is in case
Judge on cloud computing center, which uses the Hadoop cluster platforms based on (SuSE) Linux OS, the Platform deployment
Tomcat servers and Hadoop clusters, HBase clusters, Zookeeper clusters and SolrCloud clusters；The system packet
Include following module：
(1) user logs in and user right control module：User of the module for all kinds of roles logs in, and to different roles
Set necessary permission.
(2) it decides a case management module：The module is used for storage, inquiry, retrieval and the maintenance of each item data, every data packet
Include case classification, controlled patent information, controlled patent characteristic information, information of deciding a case, case weight and the matching time of each patent
Number.
(3) intelligent decision module：The module for completing reasoning by cases and interpretation of result, the module include reasoning by cases unit,
Similar case search results management unit, similar cases statistical analysis unit and distributed full-text search unit.
2. a kind of electric business field patent infringement determination method of case-based reasioning, which is characterized in that this method is wanted based on right
The system described in 1 is asked to realize, the method includes the following steps：
(1) case library is judged using the case representation method structure patent based on object-oriented, and marks the case of each patent
Classification, controlled patent information, controlled patent characteristic information and information of deciding a case；
(2) the new judgement case of input and model parameter information, and mark the case classification information of new judgement case, controlled patent letter
Breath, controlled patent characteristic information and CBR model parameter information, the CBR model parameter information include of approximate case
Number K, case attribute weight vector W and similarity retrieval threshold value a；
(3) operation is submitted to Hadoop cluster platforms and carries out KNN MapReduce Case Retrievals；
(4) it when the similar cases retrieved cannot meet the number K and similarity retrieval threshold value a of approximate case simultaneously, then adjusts
Whole a or W, is then retrieved again, the similar cases until retrieving while the number K and similarity retrieval for meeting approximate case
Threshold value a；
(5) it is for statistical analysis to retrieval result to be based on " weight integral model "：Using different type attributes similarity calculating side
Method calculates each case in case library and the similarity between new input case, is then matched further according to similarity and K values
Go out satisfactory similar cases；Different weights is set according to its similarity size, similarity is bigger, then the weight of the case
It is bigger；Then summation is weighted according to the judgement result of each case, judgement result, which is belonged to of a sort case, is placed on one
It rises and carries out the cumulative of weight, that the maximum a kind of advisory opinion of result as this Case Retrieval of deciding a case of final weight；Institute
The case judgement result classification stated includes " infringement is set up ", " data is uneven undetermined ", " infringement is invalid ", " it is undetermined to hand in grade " four
Class；
(6) when the case that the retrieval in step (4) returns is still undesirable, then by inputting keyword in case library
Full-text index is carried out, which includes each field of case and the material that prosecutor is submitted；
(7) retrieval result is evaluated and is corrected：When some case in retrieval result solves current case to be determined
Patent problem, then the weight of the case retrieved is increased.
3. the electric business field patent infringement determination method of case-based reasioning according to claim 2, which is characterized in that institute
The step of stating (1) is specially：
It is decided a case data according to history, by each four-tuple that case representation is C=(E, P, L, R) of deciding a case, wherein E=(e1,
e2,...,em) represent the description vectors of case classification information, including two class of patent infringement disputes and Holiday culture, the patent
Infringement disputes include patent infringement disputes, Granting of patent right is not paid appropriate expense and entangled using invention after application for a patent for invention
Confusingly, four kinds of situations such as right to patent and the dispute of the patent right of attribution, inventor's designer's qualification dispute；The Holiday culture packet
It includes on product or product packaging affixing patent indication, be labelled with patent indication, the affixing patent on product description in sale
Deng three kinds of situations；P=(p1,p2,...,pn) represent the description vectors for being controlled patent information, including patent type, case-involving patent
Number, four kinds of the affiliated industry of product and obligee；L=(l1,l2,...,lp) represent the description vectors for being controlled patent characteristic information, packet
Include technical field, the field character pair, Patent right requirement content；R=(r1,r2,...,rq) represent the description of information of deciding a case
Vector, the information of deciding a case includes basic merit, result of deciding a case and is decided a case the time.
4. the electric business field patent infringement determination method of case-based reasioning according to claim 2, which is characterized in that institute
Case Retrieval in the step of stating (3) uses the KNN case retrieval algorithms based on MapReduce.
5. the electric business field patent infringement determination method of case-based reasioning according to claim 2 or 4, feature exist
In the different type attributes similarity computational methods in the step (5) are：
S1：Each case in case library and the similarity between the new input single attribute of case are calculated, specific calculation is such as
Under：
(1) similarity calculation of continuous type numerical attribute, calculation formula are as follows：
S indicates that original bill example attribute, t indicate that the identical attribute of target case in case library, max (c, t) indicate representated by c and t
The maximum value of the codomain of attribute, min (s, t) indicate the minimum value of the codomain for the attribute that s and t is represented；
(2) the orderly calculating of the similarity of attribute：Orderly attribute is reduced to order enumeration type first, and according to semantic strong and weak
Order arranged, it is assumed that attribute is divided into n grade, then the calculating formula of similarity between grade i and grade j is as follows：
Wherein ord (i) is order of the attribute value i in codomain set；cardiIt is the series of attribute point；
(3) similarity calculation of character type attribute：The similarity calculation of character type attribute is calculated using the matching of MMSEG Chinese word segmentations
Method：Specific calculating formula of similarity is as follows：
Wherein, Stringtoken () function is the participle array obtained using the matching algorithm of MMSEG Chinese word segmentations, same ()
Function calculates the number of synonymous word after two character string participles, and maxlen () calculates the number of longest character string participle；
(4) similarity calculation of fuzzy interval attribute：
The first step：Membership function is constructed according to fuzzy interval.
Second step：Two fuzzy intervals and its overlapping interval corresponding area, the Duplication of area are calculated separately according to membership function
As the similarity of fuzzy interval, the calculation formula of similarity is as follows：
Wherein, SiIndicate that one of fuzzy interval i attributes pass through the calculated corresponding area of membership function, TiShow another mould
Section i attributes are pasted by the calculated corresponding area of membership function, S indicates to pass through the calculated corresponding area of membership function；
S2：Calculate each case in case library and the similarity between the case of new input case.Calculation formula is as follows：
Wherein, wkThe weights of k-th of feature in case characteristic vector are represented,aSkIndicate k-th of feature of case S
Weights, aTkThe value of k-th of feature of case T, sim (a are indicated respectivelySk,aTk) be case k-th of feature of S and T similarity.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810217918.XA CN108428200A (en) | 2018-03-16 | 2018-03-16 | A kind of the electric business field patent infringement decision-making system and determination method of case-based reasioning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810217918.XA CN108428200A (en) | 2018-03-16 | 2018-03-16 | A kind of the electric business field patent infringement decision-making system and determination method of case-based reasioning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108428200A true CN108428200A (en) | 2018-08-21 |
Family
ID=63158293
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810217918.XA Pending CN108428200A (en) | 2018-03-16 | 2018-03-16 | A kind of the electric business field patent infringement decision-making system and determination method of case-based reasioning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108428200A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109448793A (en) * | 2018-10-15 | 2019-03-08 | 智慧芽信息科技(苏州)有限公司 | The interest field identification of gene order, retrieval and infringement determination method, system |
| CN109493259A (en) * | 2018-10-18 | 2019-03-19 | 上海右上角文化传媒有限公司 | A kind of method and apparatus for by calculating equipment processing case of encroachment of right |
| CN109582964A (en) * | 2018-11-29 | 2019-04-05 | 天津工业大学 | Intelligent legal advice auxiliary system based on marriage law judicial decision document big data |
| CN109800416A (en) * | 2018-12-14 | 2019-05-24 | 天津大学 | A kind of power equipment title recognition methods |
| CN112561456A (en) * | 2019-09-26 | 2021-03-26 | 北京国双科技有限公司 | Examination and approval auxiliary method and device, storage medium and equipment |
| CN113609256A (en) * | 2021-08-05 | 2021-11-05 | 郑州银丰电子科技有限公司 | Smart court management system and method based on big data |
| CN114048170A (en) * | 2021-10-20 | 2022-02-15 | 北京鲸鲮信息系统技术有限公司 | Method, apparatus, device and medium for searching files across containers |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104573106A (en) * | 2015-01-30 | 2015-04-29 | 浙江大学城市学院 | Intelligent urban construction examining and approving method based on case-based reasoning technology |
| CN104851025A (en) * | 2015-05-09 | 2015-08-19 | 湘南学院 | Case-reasoning-based personalized recommendation method for E-commerce website commodity |
-
2018
- 2018-03-16 CN CN201810217918.XA patent/CN108428200A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104573106A (en) * | 2015-01-30 | 2015-04-29 | 浙江大学城市学院 | Intelligent urban construction examining and approving method based on case-based reasoning technology |
| CN104851025A (en) * | 2015-05-09 | 2015-08-19 | 湘南学院 | Case-reasoning-based personalized recommendation method for E-commerce website commodity |
Non-Patent Citations (1)
| Title |
|---|
| 王二朋: "大数据技术和案例推理在城市建设审批中的研究与应用", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109448793A (en) * | 2018-10-15 | 2019-03-08 | 智慧芽信息科技(苏州)有限公司 | The interest field identification of gene order, retrieval and infringement determination method, system |
| CN109448793B (en) * | 2018-10-15 | 2021-04-20 | 智慧芽信息科技(苏州)有限公司 | Method and system for labeling, searching and information labeling of right range of gene sequence |
| CN109493259A (en) * | 2018-10-18 | 2019-03-19 | 上海右上角文化传媒有限公司 | A kind of method and apparatus for by calculating equipment processing case of encroachment of right |
| CN109493259B (en) * | 2018-10-18 | 2024-03-05 | 上海右云信息技术有限公司 | Method and device for processing infringement cases through computing device |
| CN109582964A (en) * | 2018-11-29 | 2019-04-05 | 天津工业大学 | Intelligent legal advice auxiliary system based on marriage law judicial decision document big data |
| CN109800416A (en) * | 2018-12-14 | 2019-05-24 | 天津大学 | A kind of power equipment title recognition methods |
| CN112561456A (en) * | 2019-09-26 | 2021-03-26 | 北京国双科技有限公司 | Examination and approval auxiliary method and device, storage medium and equipment |
| CN113609256A (en) * | 2021-08-05 | 2021-11-05 | 郑州银丰电子科技有限公司 | Smart court management system and method based on big data |
| CN113609256B (en) * | 2021-08-05 | 2022-03-15 | 郑州银丰电子科技有限公司 | Smart court management system and method based on big data |
| CN114048170A (en) * | 2021-10-20 | 2022-02-15 | 北京鲸鲮信息系统技术有限公司 | Method, apparatus, device and medium for searching files across containers |
| CN114048170B (en) * | 2021-10-20 | 2024-04-02 | 北京字节跳动网络技术有限公司 | Method, apparatus, device and medium for searching files across containers |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108428200A (en) | A kind of the electric business field patent infringement decision-making system and determination method of case-based reasioning | |
| CA2910866C (en) | Digital communications interface and graphical user interface | |
| Rostamzadeh et al. | Prioritizing effective 7Ms to improve production systems performance using fuzzy AHP and fuzzy TOPSIS (case study) | |
| Cables et al. | The LTOPSIS: An alternative to TOPSIS decision-making approach for linguistic variables | |
| CN104573106B (en) | A kind of intelligent measures and procedures for the examination and approval of the urban construction of case-based reasioning technology | |
| Amiri | Project selection for oil-fields development by using the AHP and fuzzy TOPSIS methods | |
| US8712955B2 (en) | Optimizing federated and ETL'd databases with considerations of specialized data structures within an environment having multidimensional constraint | |
| CN106776822B (en) | Conglomerate's report data extracting method and system | |
| CN106844407B (en) | Tag network generation method and system based on data set correlation | |
| CN107016068A (en) | Knowledge mapping construction method and device | |
| CN109840730B (en) | Method and device for data prediction | |
| CN106156135A (en) | The method and device of inquiry data | |
| AU2011210742A1 (en) | Method and system for conducting legal research using clustering analytics | |
| KR102121901B1 (en) | System for online public fund investment management assessment service | |
| Purnomo et al. | E-money Academic: Lesson from Literature Visualizing Scientometric Positioning (1968-2019) | |
| Malik et al. | A new approach to solve fully intuitionistic fuzzy linear programming problem with unrestricted decision variables | |
| Verma et al. | Data mining: next generation challenges and futureDirections | |
| Cao et al. | A knowledge discovery model for third-party payment networks based on rough set theory | |
| Liu et al. | Application of master data classification model in enterprises | |
| Ruzgas | Big data mining and knowledge discovery | |
| Ben Khalifa et al. | Evidential group spammers detection | |
| Fan et al. | Spatially enabled customer segmentation using a data classification method with uncertain predicates | |
| Zhang et al. | Research on Cross-border E-commerce platform selection in China small & medium-sized enterprises | |
| Bal et al. | Creating competitive advantage by using data mining technique as an innovative method for decision making process in business | |
| Mahalik | Selection of a plant site: A multi criteria decision making using AHP and GRA |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20180821 |