CN106095939A - The acquisition methods of account authority and device - Google Patents
The acquisition methods of account authority and device Download PDFInfo
- Publication number
- CN106095939A CN106095939A CN201610414403.XA CN201610414403A CN106095939A CN 106095939 A CN106095939 A CN 106095939A CN 201610414403 A CN201610414403 A CN 201610414403A CN 106095939 A CN106095939 A CN 106095939A
- Authority
- CN
- China
- Prior art keywords
- text
- word
- classification
- target account
- attribute
- 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.)
- Granted
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
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2457—Query processing with adaptation to user needs
-
- 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
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Marketing (AREA)
- Economics (AREA)
- Development Economics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses acquisition methods and the device of a kind of account authority.Wherein, the method includes: receiving inquiry request, wherein, described inquiry request performs the authority of scheduled operation for the capability identification of requesting query target account, described capability identification for indicating described target account；Obtain the browing record of described target account, wherein, the text attribute of the text that described browing record has been read for indicating described target account；Obtain the capability identification of the described target account corresponding with the described text attribute of described browing record instruction；Send described capability identification.The present invention at least solves to utilize incomplete data to evaluate user right, the technical problem of the accuracy of impact evaluation authority.
Description
Technical field
The present invention relates to field of information processing, in particular to acquisition methods and the device of a kind of account authority.
Background technology
Authority evaluation platform can gather and record the behavioural information of user, sets up the power of user based on these behavioural informations
Limit report, this authority evaluation platform can utilize the authority report of user externally to provide authority information service, as by the power of user
Limit information is supplied to third party, inquires about for third party, and such as, above-mentioned authority can perform the authority of resource transfers operation,
Then authority evaluation platform can be reference platform, reference platform the report of the authority of user is supplied to bank, financial regulator,
And the third party such as judicial department, for the authority of the transferring resource of these mechanisms inquiry user.
At present, during the authority of evaluation user, it is based primarily upon virtual resource exchange record and the void of user's history
Intend the resource volume of holding (volume residual as in bank card) determine, as bank card flowing water, electricity business's platform on consumption record,
Creditor-debtor entry etc.).Only rely on virtual resource exchange record with the virtual resource volume of holding to evaluate the authority of user, determine that user weighs
The data of limit are incomplete, utilize incomplete data to evaluate user right, the accuracy of impact evaluation authority.
Summary of the invention
Embodiments provide acquisition methods and the device of a kind of account authority, incomplete at least to solve to utilize
User right is evaluated by data, the technical problem of the accuracy of impact evaluation authority.
An aspect according to embodiments of the present invention, it is provided that the acquisition methods of a kind of account authority, the method includes: connect
Receiving inquiry request, wherein, described inquiry request is for the capability identification of requesting query target account, and described capability identification is used for referring to
Show that described target account performs the authority of scheduled operation；Obtain the browing record of described target account, wherein, described browing record
For indicating the text attribute of text that described target account read；Obtain the described text with the instruction of described browing record to belong to
The capability identification of the described target account that property is corresponding；Send described capability identification.
An aspect according to embodiments of the present invention, it is provided that the acquisition device of a kind of account authority, this device includes: account
The acquisition device of family authority, it is characterised in that including: first receives unit, is used for receiving inquiry request, wherein, described inquiry
Request is for the capability identification of requesting query target account, and described capability identification is used for indicating described target account to perform predetermined behaviour
The authority made；First acquiring unit, for obtaining the browing record of described target account, wherein, described browing record is used for referring to
Show the text attribute of the text that described target account read；Second acquisition unit, indicates with described browing record for obtaining
The capability identification of described target account corresponding to described text attribute；First transmitting element, is used for sending described capability identification.
In embodiments of the present invention, after receiving inquiry request, obtain the browing record of target account, for this mesh
The attribute of the text that the account of the browing record instruction of mark account has read, determines the interest tendency in user knowledge field, and
Obtain the capability identification of the target account corresponding with the text attribute of browing record instruction, and capability identification is sent to request
Side.By above-described embodiment, when evaluating target account performs the authority of scheduled operation, add the browing record of target account
Making, the authority using more comprehensively data that user performs scheduled operation is evaluated, and improves the accuracy defined the competence, and solves
Prior art of having determined utilize incomplete data user right is evaluated, the problem of the accuracy of impact evaluation authority.
Accompanying drawing explanation
Accompanying drawing described herein is used for providing a further understanding of the present invention, constitutes the part of the application, this
Bright schematic description and description is used for explaining the present invention, is not intended that inappropriate limitation of the present invention.In the accompanying drawings:
Fig. 1 is the schematic diagram of the hardware environment of the acquisition methods of a kind of account authority according to embodiments of the present invention；
Fig. 2 is the flow chart of the acquisition methods of a kind of optional account authority according to embodiments of the present invention；
Fig. 3 is the flow chart of the acquisition methods of another kind of optional account authority according to embodiments of the present invention；
Fig. 4 is the flow chart of the acquisition methods of according to embodiments of the present invention another optional account authority；
Fig. 5 is another flow chart of the acquisition methods of a kind of optional account authority according to embodiments of the present invention；
Fig. 6 is the schematic diagram of the acquisition methods of a kind of optional account authority according to embodiments of the present invention；
Fig. 7 is the structured flowchart of terminal according to embodiments of the present invention.
Detailed description of the invention
In order to make those skilled in the art be more fully understood that the present invention program, below in conjunction with in the embodiment of the present invention
Accompanying drawing, is clearly and completely described the technical scheme in the embodiment of the present invention, it is clear that described embodiment is only
The embodiment of a present invention part rather than whole embodiments.Based on the embodiment in the present invention, ordinary skill people
The every other embodiment that member is obtained under not making creative work premise, all should belong to the model of present invention protection
Enclose.
It should be noted that term " first " in description and claims of this specification and above-mentioned accompanying drawing, "
Two " it is etc. for distinguishing similar object, without being used for describing specific order or precedence.Should be appreciated that so use
Data can exchange in the appropriate case, in order to embodiments of the invention described herein can with except here diagram or
Order beyond those described is implemented.Additionally, term " includes " and " having " and their any deformation, it is intended that cover
Cover non-exclusive comprising, such as, contain series of steps or the process of unit, method, system, product or equipment are not necessarily limited to
Those steps clearly listed or unit, but can include the most clearly listing or for these processes, method, product
Or intrinsic other step of equipment or unit.
Embodiment 1
According to embodiments of the present invention, it is provided that the acquisition methods of a kind of account authority, in the present embodiment, above-mentioned account power
The acquisition methods of limit can apply in hardware environment as shown in Figure 1, and wherein, hardware environment includes network environment.
Alternatively, in the present embodiment, the acquisition methods of above-mentioned account authority can apply to server as shown in Figure 1
101 and the hardware environment that constituted of terminal 103 in.As it is shown in figure 1, server 101 is attached with terminal 103 by network,
Above-mentioned network includes but not limited to: wide area network, Metropolitan Area Network (MAN) or LAN.
Wherein, above-mentioned terminal 103 can be personal computer, it is also possible to for mobile terminal, this mobile terminal can be
Smart mobile phone, panel computer lamp, this terminal can be provided with authority evaluation application, and as reference is applied, user can be by being somebody's turn to do
After the evaluation application of terminal logon rights, actively record user profile；Or after the evaluation application of user's logon rights, gather and use
The behavioural information at family, to evaluate the authority of this user；Alternatively, after the evaluation application of user's logon rights, can be by the
Tripartite applies the behavioural information gathering user, performs the authority of scheduled operation evaluating this user.
Above-mentioned scheduled operation can be the authority of acquisition request virtual resource, such as debt-credit etc..
As in figure 2 it is shown, the acquisition methods of the account authority in the embodiment of the present application, can be achieved by the steps of:
Step S202: receiving inquiry request, wherein, inquiry request is for the capability identification of requesting query target account, power
Limit mark performs the authority of scheduled operation for indicating target account；
Step S204: obtaining the browing record of target account, wherein, browing record is for indicating target account to read
The text attribute of text；
Step S206: obtain the capability identification of the target account corresponding with the text attribute of browing record instruction；
Step S208: sending permission identifies.
In the above-described embodiments, after receiving inquiry request, obtain the browing record of target account, for this target
The attribute of the text that the account of the browing record instruction of account has read, determines the interest tendency in user knowledge field, and obtains
Take the capability identification of the target account corresponding with the text attribute of browing record instruction, and capability identification is sent to requesting party.
By above-described embodiment, when evaluating target account performs the authority of scheduled operation, the browing record adding target account is made,
The authority using more comprehensively data that user performs scheduled operation is evaluated, and improves the accuracy defined the competence, and solves
Prior art utilize the incomplete data to evaluate user right, the problem of the accuracy of impact evaluation authority.
Wherein, the acquisition methods of the account authority in above-described embodiment can be applied on server or terminal, the application
Embodiment, all as a example by server, describes the scheme of the application in detail, and specifically, after server receives inquiry request, acquisition is stepped on
The browing record of record target account in the authority evaluation application of first terminal.
Being provided with authority evaluation application on this first terminal, the evaluation application of this authority can be self-existent application, as
The client of this authority evaluation application is installed on first terminal；This authority evaluation platform can also be deposited for relying on third-party application
Application, the html page that can be provided by browser application such as this authority evaluation platform is entered, and this authority evaluates platform
Can also be embedded in third-party application, as authority evaluation platform can be entered by the authority evaluation public number that wechat is applied.
User can be after logon rights evaluation application, the capability identification of requesting query oneself.With authority evaluation application
As a example by independent application, user starts authority evaluation client, uses account and this client of password login, logs in client
Afterwards, on the page of this client, record has the button of search access right, after user clicks on the button of this search access right, generates
Inquiry request.
Third party's account can generate inquiry request after receiving the request of target account execution scheduled operation, with
Inquiry target account performs the authority of this scheduled operation.As a example by authority evaluation application is for independent application, third party's account opens
Dynamic authority evaluation client, uses account and this client of password login, after logging in client, receives the use of target account
Perform the request that performs of scheduled operation in request, third party's account responds this and performs request, generates inquiry request, and by this inquiry
Request sends to server.
Wherein, inquiry request at least carries the information of scheduled operation and the information of target account.
After server receives inquiry request, the data base from capability identification is applied can obtain target account
Browing record, in order to obtain more complete data, server can also obtain the reading of target account from third-party application
Record, this third-party application can be set up relevant with authority evaluation application, such as authority evaluation application and third-party application
Bind.
Server with the target account in the evaluation application of this authority of real time record and can evaluate association with this authority
Account behavior on third-party application, to obtain complete data, this third-party application can be media file application, e.g., money
News application, reading application and document application, wherein, information application can provide substantial amounts of photo information, in reading application
Electronic book documentary, document application can be provided to can be used to read or editor's document (such as word document and PDF document etc.).
After server obtains browing record, obtain the target account corresponding with the text attribute of browing record instruction
Capability identification, to obtain the authority of target account execution scheduled operation.
Wherein, capability identification can by use at least one following in the way of show: word description, numerical value, grade, rank,
Probit and weighted value, this is not limited by the application.
When the mode that capability identification uses word to describe represents, identify word, to determine that target account performs predetermined behaviour
The authority height made.Such as, the capability identification obtained is " target account haves no right to check document A ", then be appreciated that target account
Authority ratio relatively low；If the capability identification obtained is " target account can download document A ", then learn the authority ratio of target account
Higher.
When the mode of capability identification service rating represents, from the Permission Levels table pre-set, read target account
Authority.Such as, asking to perform to check operation to document A, in the Permission Levels table as shown in table 1, grade 1 does not have document A
Performing the authority of any operation, grade 2 has the authority that document A performs to check operation, and grade 3 has to be looked into document A execution
See the authority with down operation.If the capability identification obtained is 3, then this capability identification instruction target account has and holds document A
The authority that row is checked and downloaded.
Table 1
| Grade | Authority |
| 1 | Have no right |
| 2 | Check |
| 3 | Check, download |
When capability identification uses level to represent otherwise, it is also possible to read target from the level-right table pre-set
The authority of account.When the mode that capability identification uses numerical value represents, this numerical value can be read from the table of grading pre-set
The authority that the numerical range at place points to.When the mode that capability identification uses probit and weighted value represents, define the competence mark
The mode of the authority knowing instruction is similar with the implementation in above-described embodiment, does not repeats them here.
Obtaining after capability identification, this capability identification is sent to requesting party, requesting party receive capability identification it
After, in terminal, this capability identification can be resolved in the manner described above the authority that capability identification is corresponding, and by capability identification and
The authority parsed shows on the screen of terminal.
Being applied by above-described embodiment on the credit evaluation of target account, the interest that can analyze user knowledge field is inclined
To and change, user's bank card service condition that conventional credit evaluation is focused on and loan profile, these data be one important
Supplement, accuracy and crowd's coverage of credit scoring can be improved.
According to the abovementioned embodiments of the present invention, the power of the target account corresponding with the text attribute of browing record instruction is obtained
Limit mark may include that the text attribute of the multiple texts read in preset duration according to target account, determines target account
The attribute summary information at family；Attribute summary information is carried out mapping process, obtains the capability identification of target account.
Owing to the behavior of user is ceaselessly changing as time goes by, target account is carried out authority every time and comment
Regularly, user's authority within this time period can be evaluated according to the behavior in scheduled duration of this target account, can be no longer
The behavioural information that reference distance current time is remote, so that the evaluation to user right is more accurate.
Specifically, text attribute represents with multiple property parameters, has read in preset duration according to target account
The text attribute of multiple texts, determines that the attribute summary information of target account includes: be retrieved as each property parameters and be pre-configured with
Weight coefficient；According to weight coefficient, configure each property parameters of each text, obtain the property parameters value of each text；
Collect the property parameters value of each property parameters of multiple text, obtain attribute summary information.
When the property parameters of the classification that property parameters is text, the parameter value of property parameters can be that text belongs to such
Other probit, or the number of times that word based on the category occurs in the text determines, and as occurred ten times, then parameter value is 10.
When the property parameters of the Feature Words that property parameters is text, the parameter value of property parameters can be characterized the spy of word
Levy vector, it is also possible to for this feature word, probit in the text occurs, it is also possible to for occurring at text based on this feature word
In number of times determine, as occur ten times, then parameter value is 10.
Different attribute parameter is different on the impact of the mark that finally defines the competence, and in this embodiment, utilizes property parameters
Weight coefficient, configuration attribute parameter, obtain the property parameters value of each property parameters, configuration here can utilize product real
Existing, as using the parameter value of property parameters and the product of weight coefficient as the property parameters value of each property parameters.In this enforcement
In example, the property parameters that weight coefficient is the biggest is the biggest on the impact of capability identification.
Such as, computer other weight could be arranged to 0.9, and the weight of gambling classification could be arranged to 0.01, so
The browing record of computer other positive energy text accounting when evaluating capability identification is heavier, the capability identification that obtains accurate
Rate is the highest.
It is set to pre-it should be noted that the embodiment of the present application differs with the weight coefficient sum of the different parameters of dimension
Definite value 1, the weight coefficient in the application is for representing that different attribute parameter affects size to the accuracy rate of capability identification.
In above-described embodiment, attribute summary information includes the information of at least one dimension, wherein, enters attribute summary information
Row mapping processes, and the capability identification obtaining target account includes: obtain the regression model pre-set, wherein, in regression model
Record has the mapping relations of attribute summary information and capability identification；Using defeated as regression model of the information of at least one dimension
Enter, utilize the mapping relations of record in regression model, obtain the capability identification of target account.
Wherein, the text attribute in above-described embodiment can be identical with the dimension of attribute summary information, the most at least includes literary composition
This classification and/or the Feature Words the two dimension of text.In the above-described embodiments, can by the summarized results of text categories and
The summarized results of the Feature Words of text is in the lump as input, it is also possible to only by summarized results or the Feature Words of text of file class
Summarized results as input.
Below as a example by performing virtual resource operation in detail, the embodiment of the present application is described in detail.
In an optional embodiment, server receives inquiry request, obtains target account and has read the class of text
Other attribute summary information (e.g., article label reading is distributed), obtains the capability identification of target account.
Specifically, when user reads an article, it is believed that this article reflects the interest in user knowledge field and inclines
To, α can be usedmWmExpress the property parameters of classification, wherein WmIt is the label (i.e. classification) of this article, αmIt is corresponding WmWeight
Coefficient.Label distribution v under i-th time interval, obtained by the m piece article that user is readiCan be: vi=H (α1W1,…,αiWm).Accumulative m time interval, can obtain user's (the most above-mentioned preset duration) in the range of a period of time
Label distribution G (x)=F (vn,vn-1,…,v1), by the regression model built, G (x) is mapped, available user credit
Scoring (the most above-mentioned capability identification).
Above-mentioned H and F represents certain functional relationship, and as H represents intersection, weighted calculation, accumulation calculating etc., F represents conjunction
Collection, weighted calculation and accumulation calculating etc..
In an optional embodiment, server receives inquiry request, obtains target account and has read the spy of text
Levy the attribute summary information (e.g., article Feature Words is distributed) of word, obtain the capability identification of target account.
Specifically, when user reads an article, β can be usednDnExpress the property parameters of the Feature Words of text, its
Middle DnRepresent the distribution of the Feature Words of article, βnIt is Feature Words DnWeight coefficient.Under i-th time interval, user is read
N piece article obtained by Feature Words distribution can be:
vj=H (β1D1,…βnDn).Accumulative n time interval, can obtain user x feature in the range of a period of time
Word distribution P (x)=F (vn,vn-1,…,v1), Feature Words is distributed the P (x) input layer as neutral net, what input built returns
Return in model, available user credit scoring (i.e. capability identification).
Wherein, above-mentioned H and F represents certain functional relationship, as H represents intersection, weighted calculation, accumulation calculating etc., F table
Show intersection, weighted calculation and accumulation calculating etc..
In another optional embodiment, server receives inquiry request, collects target account and has read text
The attribute of the classification that the attribute summary information (e.g., article Feature Words distribution) of Feature Words and target account have read text collects letter
Breath (e.g., article label reading is distributed), obtains the capability identification of target account.
Specifically, the story label of user is distributed G (x) and Feature Words distribution P (x) uses the mode spliced, obtain new
Distribution H (x), build regression model, available user credit scoring (such as capability identification).
In above-mentioned optional embodiment, capability identification can be represented with scoring.
According to the abovementioned embodiments of the present invention, when text attribute includes the classification of text, obtain the reading of target account
Record may include that and obtains the dictionary previously generated, and in dictionary, record has the N number of classification obtained in advance；To text with obtain in advance
The N number of classification taken carries out matching treatment, obtains at least one classification of text；The probability that text belongs to each classification is recorded as
The property parameters of corresponding classification, wherein, browing record includes the property parameters of at least one classification.
In above-described embodiment, text is carried out matching treatment with the N number of classification obtained in advance, obtains text at least one
Classification may include that multiple first words obtaining the i-th classification preserved in dictionary, wherein, 1≤i≤N；Add up the first word
There is word frequency in the text and duplicate removal word number in language, and wherein, word frequency, for representing the total degree occurring the first word in text, is gone
Weight word number is for stating the number occurring the first word in text；If word frequency exceedes first threshold and duplicate removal number of times more than second
Threshold value, it is determined that i-th classification is the classification of text.
Further, before obtaining the dictionary previously generated, method can also include: extracts second comprised in sample
Word；The term vector utilizing the second word carries out cluster operation to the second word, to be divided in M classification by the second word；
The identification information of each classification is set；Preserve the second word that M classification, the identification information of each classification and each classification comprise
Language, generates dictionary.
In order to ensure the accuracy rate of dictionary, the dictionary that can generate with immediate updating, specifically, after generating dictionary,
In the case of Sample Refreshment being detected, obtain the dictionary of more new samples；The dictionary of the dictionary generated and more new samples is closed
And, the dictionary after being updated.
Below in conjunction with table 2, Fig. 3 and Fig. 4, the mode updating dictionary of above-described embodiment of the application is described in detail.
As it is shown on figure 3, this embodiment can be achieved by the steps of:
Step S301: obtain sample A.
Step S302: sample A is carried out pretreatment and obtains the word that A comprises.
Specifically, after obtaining the text A of a large amount of articles, text is carried out participle, stop words process, part of speech filtration etc.
Text is correlated with pretreatment, obtains the word that A comprises.
Step S303: calculate the term vector of word.
Step S304: utilize term vector that word carries out cluster and obtain N class word.
Specifically, by using clustering algorithm, the word that text A comprises is divided into N class.
Step S305: be identified N class word, generates dictionary.
It is alternatively possible to provide the meaning tag of each class word, label and a class word are relations one to one.Such as table 2
The corresponding relation illustrated, class ID after cluster in table 2 and label belong to such identification information.
Table 2
Fig. 4 shows the renewal iterative process of dictionary, and this process can realize with following steps as shown in Figure 4:
Step S401: obtain sample B.
Step S402: sample B is carried out pretreatment and obtains the word that B comprises.
Specifically, after obtaining the text B of a large amount of articles, text is carried out participle, stop words process, part of speech filtration etc.
Text is correlated with pretreatment, obtains the word that B comprises.
Step S403: calculate the term vector of word.
Step S404: utilize term vector that word carries out cluster and obtain N class word, generate dictionary.
Specifically, by using clustering algorithm, the word that text B comprises is divided into M class.
From above-mentioned steps, it is thus achieved that the text of the most a collection of a large amount of articles, carry out same Text Pretreatment, calculate word
Vector and clustering algorithm, obtain new N class dictionary B.
Step S405: the similarity coefficient between calculating dictionary A and dictionary B is all kinds of.
Step S406: compare the size of similarity coefficient and threshold value beta.
Wherein, when the similarity coefficient of two class words is not less than threshold value beta, then perform step S407: merge dictionary A and dictionary B
Being a class, share a label, merging here refers to retain a dictionary and deletes a dictionary, can delete word sum
Or that dictionary that categorical measure is few.
When the similarity coefficient of two class words is less than threshold value beta, then perform step S408: collect dictionary B and dictionary.
Step S409: the dictionary after collecting is carried out duplicate removal.
It should be noted that new dictionary C after collecting does not has the classification information of identification information can increase mark letter
The configuration of breath.
Step S410: calculate the label distribution of article.
Specifically, the method using statistics, for the i-th label in dictionary C, the word frequency to this label of paper statistics
As long as (word in i-th label just note one number, the sum finally obtained occur) and duplicate removal word number are (in i-th label
Word quantity in the text occurs, word does not carry out repeat count, e.g., word A occurs ten times, and note is once), set threshold value respectively
CiAnd Di, respectively by word frequency and duplicate removal word number and above-mentioned threshold value CiAnd DiMake comparisons, retain word frequency and duplicate removal word number exceedes threshold value
Label, finally give this article text label distribution Wj。
In another optional embodiment, when text attribute includes the Feature Words of text, obtain readding of target account
Read record may include that and obtains at least one Feature Words that text is comprised；Obtain the property parameters of each Feature Words, wherein,
Browing record includes the property parameters of Feature Words.
The property parameters of different characteristic word is different on the impact of the mark that finally defines the competence, and in this embodiment, utilizes spy
The property parameters indicative character word levying word can be quantitative Feature Words is analyzed.
Specifically, at least one Feature Words that acquisition text is comprised may include that and text is carried out word segmentation processing, obtains
Multiple key words；Multiple key words are carried out stop words process and filter operation, obtains target keyword；Utilize target keyword
Term vector, from target keyword, filter out at least one Feature Words of text.
Alternatively, the property parameters obtaining each Feature Words may include that adding up each Feature Words occurs in the text
Word frequency；Word frequency number based on each Feature Words and weight coefficient generate the characteristic vector of each Feature Words；By characteristic vector record
It is characterized the property parameters of word.
The effect of weight coefficient here is identical with the weight coefficient effect in above-described embodiment, based on each Feature Words
Word frequency number and weight coefficient generate the characteristic vector of each Feature Words, it is possible to use product realizes, as by the parameter value of Feature Words
With the product of weight coefficient as the property parameters value of each property parameters.In this embodiment, the attribute that weight coefficient is the biggest
Parameter is the biggest on the impact of capability identification.
Such as, computer other weight could be arranged to 0.9, and the weight of gambling classification could be arranged to 0.01, so
The browing record of computer other positive energy text accounting when evaluating capability identification is heavier, the capability identification that obtains accurate
Rate is the highest.
Below in conjunction with Fig. 5, the embodiment of the present application is explained in detail, as it is shown in figure 5, this embodiment can be by following step
Rapid realization:
Step S501: obtain text.
Step S502: text is carried out pretreatment, obtains multiple key word.
Specifically, text carrying out participle, stop words processes, the text such as part of speech filtration is correlated with pretreatment, obtains multiple pass
Keyword.
Step S503: determine the target keyword in key word.
By the term vector distribution N number of word of screening as article Feature Words, then can express all literary compositions by the dictionary with word number as N
The feature of chapter.
Step S504: calculate the distribution of article Feature Words.
One article can represent with a length of N vector V, VijRepresent i-th word characteristic vector in jth article (as
Weight): Vij=F (Wi,Cij), wherein, CijFor i-th word at the word frequency number of jth piece article, WiWeight coefficient for i-th word.
Therein, F represents predetermined functional relationship, such as weighted calculation, accumulation calculating, or the calculating that is multiplied.
Prior art is determined by artificial dictionary the text attribute of text, so needs the word of each label of manual maintenance
Allusion quotation, efficiency is low.And use the term vector cluster mode of the application, can quickly obtain the word belonging to a class in a large number together, and permissible
Dynamically update, reduce the difficulty of maintenance directory.
The above embodiment of the present invention, uses term vector cluster, is polymerized by the word under similar semantic, in conjunction with less artificial mark
Note, obtains the word that each label is comprised rapidly, and the method can iteration update, in order to identifies new story label rapidly.Logical
Cross the data such as the method added up, the word frequency under the statistics each label of article, require to obtain the standard of this article with relatively low calculated performance
True label distribution.
Although it is understood that this distribution of content by obtaining the label distribution of article, divided by the label of statistics author
Cloth, then it will be seen that the distribution of content that this author creates.It is distributed by the label of the reading articles of counting user, then it will be seen that
The article label reading distribution of this user, and then analyze user's reflected interest difference under different interest tags distributions, OK
For difference, Professional Differences, therefrom obtain credit difference, obtain credit scoring.
Story label distribution and Feature Words distribution that the present invention produces when using Internet user's reading articles obtain user
Marking mode, user base number based on bank card service condition than traditional bank reference is bigger, can supplement widely user；
The present invention can collect the Behavioral change of user in time, and contrast bank card is swiped the card and recorded and lend-borrow action record, and network data is more
Comprehensively with in time, and then credit scoring more accurately can be obtained on the basis of bank reference.
By above-described embodiment, after obtaining the capability identification of target account, according to the target account of capability identification instruction
Family performs the authority of scheduled operation, provides service for target account.
In table 1, if the grade of the capability identification obtained is 2, it is determined that authority checks authority, can be that target account carries
The document of a reading mode of confession.
The a kind of of the above embodiment of the present invention offer is obtained by story label distribution during analysis user's reading articles
The method of the capability identification of user, carries out participle, stop words process, part of speech filtration, word carries out term vector calculating text, right
Term vector clusters, and by being manually labeled cluster result, determines the label of such word, and combine iterative algorithm can
Dynamically to update the label of this word, the label obtaining this section of text in the means such as word frequency of different labels by statistics text divides
Cloth, by article term vector distributed acquisition article Feature Words, obtains one section of literary composition by means such as the word frequency of statistics text feature word
This Feature Words distribution.By the distribution of article label reading or Feature Words distribution calculating user, user credit is marked.
Thus can easy credit scoring cover customer volume and improve credit scoring accuracy.
It should be noted that for aforesaid each method embodiment, in order to be briefly described, therefore it is all expressed as a series of
Combination of actions, but those skilled in the art should know, the present invention is not limited by described sequence of movement because
According to the present invention, some step can use other orders or carry out simultaneously.Secondly, those skilled in the art also should know
Knowing, embodiment described in this description belongs to preferred embodiment, involved action and the module not necessarily present invention
Necessary.
Through the above description of the embodiments, those skilled in the art is it can be understood that arrive according to above-mentioned enforcement
The method of example can add the mode of required general hardware platform by software and realize, naturally it is also possible to by hardware, but a lot
In the case of the former is more preferably embodiment.Based on such understanding, technical scheme is the most in other words to existing
The part that technology contributes can embody with the form of software product, and this computer software product is stored in a storage
In medium (such as ROM/RAM, magnetic disc, CD), including some instructions with so that a station terminal equipment (can be mobile phone, calculate
Machine, server, or the network equipment etc.) perform the method described in each embodiment of the present invention.
According to embodiments of the present invention, the account authority of a kind of acquisition methods for implementing above-mentioned account authority is additionally provided
Acquisition device, as shown in Figure 6, this device may include that
First receives unit 61, is used for receiving inquiry request, and wherein, inquiry request is for the power of requesting query target account
Limit mark, capability identification performs the authority of scheduled operation for indicating target account；
First acquiring unit 62, for obtaining the browing record of target account, wherein, browing record is used for indicating target account
The text attribute of the text that family has been read；
Second acquisition unit 63, for obtaining the authority mark of the target account corresponding with the text attribute of browing record instruction
Know；
First transmitting element 64, identifies for sending permission.
In the above-described embodiments, after receiving inquiry request, obtain the browing record of target account, for this target
The attribute of the text that the account of the browing record instruction of account has read, determines the interest tendency in user knowledge field, and obtains
Take the capability identification of the target account corresponding with the text attribute of browing record instruction, and capability identification is sent to requesting party.
By above-described embodiment, when evaluating target account performs the authority of scheduled operation, the browing record adding target account is made,
The authority using more comprehensively data that user performs scheduled operation is evaluated, and improves the accuracy defined the competence, and solves
Prior art utilize the incomplete data to evaluate user right, the problem of the accuracy of impact evaluation authority.
Wherein, the acquisition methods of the account authority in above-described embodiment can be applied on server or terminal, the application
Embodiment, all as a example by server, describes the scheme of the application in detail, and specifically, after server receives inquiry request, acquisition is stepped on
The browing record of record target account in the authority evaluation application of first terminal.
Being provided with authority evaluation application on this first terminal, the evaluation application of this authority can be self-existent application, as
The client of this authority evaluation application is installed on first terminal；This authority evaluation platform can also be deposited for relying on third-party application
Application, the html page that can be provided by browser application such as this authority evaluation platform is entered, and this authority evaluates platform
Can also be embedded in third-party application, as authority evaluation platform can be entered by the authority evaluation public number that wechat is applied.
User can be after logon rights evaluation application, the capability identification of requesting query oneself.With authority evaluation application
As a example by independent application, user starts authority evaluation client, uses account and this client of password login, logs in client
Afterwards, on the page of this client, record has the button of search access right, after user clicks on the button of this search access right, generates
Inquiry request.
Third party's account can generate inquiry request after receiving the request of target account execution scheduled operation, with
Inquiry target account performs the authority of this scheduled operation.As a example by authority evaluation application is for independent application, third party's account opens
Dynamic authority evaluation client, uses account and this client of password login, after logging in client, receives the use of target account
Perform the request that performs of scheduled operation in request, third party's account responds this and performs request, generates inquiry request, and by this inquiry
Request sends to server.
Wherein, inquiry request at least carries the information of scheduled operation and the information of target account.
After server receives inquiry request, the data base from capability identification is applied can obtain target account
Browing record, in order to obtain more complete data, server can also obtain the reading of target account from third-party application
Record, this third-party application can be set up relevant with authority evaluation application, such as authority evaluation application and third-party application
Bind.
Server with the target account in the evaluation application of this authority of real time record and can evaluate association with this authority
Account behavior on third-party application, to obtain complete data, this third-party application can be media file application, e.g., money
News application, reading application and document application, wherein, information application can provide substantial amounts of photo information, in reading application
Electronic book documentary, document application can be provided to can be used to read or editor's document (such as word document and PDF document etc.).
After server obtains browing record, obtain the target account corresponding with the text attribute of browing record instruction
Capability identification, to obtain the authority of target account execution scheduled operation.
Wherein, capability identification can by use at least one following in the way of show: word description, numerical value, grade, rank,
Probit and weighted value, this is not limited by the application.
When the mode that capability identification uses word to describe represents, identify word, to determine that target account performs predetermined behaviour
The authority height made.Such as, the capability identification obtained is " target account haves no right to check document A ", then be appreciated that target account
Authority ratio relatively low；If the capability identification obtained is " target account can download document A ", then learn the authority ratio of target account
Higher.
When the mode of capability identification service rating represents, from the Permission Levels table pre-set, read target account
Authority.Such as, asking to perform to check operation to document A, in the Permission Levels table as shown in table 1, grade 1 does not have document A
Performing the authority of any operation, grade 2 has the authority that document A performs to check operation, and grade 3 has to be looked into document A execution
See the authority with down operation.If the capability identification obtained is 3, then this capability identification instruction target account has and holds document A
The authority that row is checked and downloaded.
When capability identification uses level to represent otherwise, it is also possible to read target from the level-right table pre-set
The authority of account.When the mode that capability identification uses numerical value represents, this numerical value can be read from the table of grading pre-set
The authority that the numerical range at place points to.When the mode that capability identification uses probit and weighted value represents, define the competence mark
The mode of the authority knowing instruction is similar with the implementation in above-described embodiment, does not repeats them here.
Obtaining after capability identification, this capability identification is sent to requesting party, requesting party receive capability identification it
After, in terminal, this capability identification can be resolved in the manner described above the authority that capability identification is corresponding, and by capability identification and
The authority parsed shows on the screen of terminal.
Being applied by above-described embodiment on the credit evaluation of target account, the interest that can analyze user knowledge field is inclined
To and change, user's bank card service condition that conventional credit evaluation is focused on and loan profile, these data be one important
Supplement, accuracy and crowd's coverage of credit scoring can be improved.
According to the abovementioned embodiments of the present invention, second acquisition unit may include that first determines module, for according to target
The text attribute of multiple texts that account has been read in preset duration, determines the attribute summary information of target account；Map mould
Block, for attribute summary information is carried out mapping process, obtains the capability identification of target account.
Alternatively, text attribute represents with multiple property parameters, and first determines that module includes: first obtains submodule,
For being retrieved as the weight coefficient that each property parameters is pre-configured with；Configuration submodule, for according to weight coefficient, configures each
Each property parameters of text, obtains the property parameters value of each text；Collect submodule, for collecting each of multiple text
The property parameters value of property parameters, obtains attribute summary information.
Specifically, attribute summary information includes the information of at least one dimension, and wherein, mapping block includes: second obtains
Submodule, for obtaining the regression model pre-set, wherein, in regression model, record has attribute summary information and capability identification
Mapping relations；First processes submodule, is used for the information of at least one dimension as the input of regression model, utilizes and return
In model, the mapping relations of record, obtain the capability identification of target account.
According to the abovementioned embodiments of the present invention, when text attribute includes the classification of text, the first acquiring unit includes: the
One acquisition module, for obtaining the dictionary previously generated, in dictionary, record has the N number of classification obtained in advance；Matching module, is used for
Text is carried out matching treatment with the N number of classification obtained in advance, obtains at least one classification of text；First logging modle, uses
Be recorded as the property parameters of corresponding classification in the probability that text belongs to each classification, wherein, browing record includes at least one
The property parameters of classification.
Specifically, matching module includes: the 3rd obtains submodule, many for obtaining the i-th classification preserved in dictionary
Individual first word, wherein, 1≤i≤N；, there is word frequency in the text for statistics the first word and goes in first statistics submodule
Weight word number, wherein, word frequency is for representing that the total degree occurring the first word in text, duplicate removal word number are used for stating in text appearance
The number of the first word；First determines submodule, if exceeding first threshold and duplicate removal number of times exceedes Second Threshold for word frequency,
Then determine the classification that i-th classification is text.
In the above-described embodiments, device can also include: extraction unit, is used for before obtaining the dictionary previously generated,
Extract the second word comprised in sample；Cluster cell, for utilizing the term vector of the second word to cluster the second word
Operation, to be divided to the second word in M classification；Unit is set, for arranging the identification information of each classification；Preserve single
Unit, for preserving the second word that M classification, the identification information of each classification and each classification comprise, generates dictionary.
Specifically, when text attribute includes the Feature Words of text, the first acquiring unit includes: the second acquisition module, uses
In obtaining at least one Feature Words that text is comprised；3rd acquisition module, for obtaining the property parameters of each Feature Words, its
In, browing record includes the property parameters of Feature Words.
In the above embodiment of the present invention, the second acquisition module includes: participle submodule, for text is carried out participle
Process, obtain multiple key word；Second processes submodule, for multiple key words are carried out stop words process and filter operation,
Obtain target keyword；Filter submodule, for utilizing the term vector of target keyword, from target keyword, filter out text
At least one Feature Words.
Further, the second acquisition module includes: the second statistics submodule, is used for adding up each Feature Words and occurs in text
In word frequency；Generate submodule, generate the feature of each Feature Words for word frequency number based on each Feature Words and weight coefficient
Vector；Second record sub module, for being recorded as the property parameters of Feature Words by characteristic vector.
The using method that modules provided in the present embodiment step corresponding with embodiment of the method is provided is identical, should
Can also be identical by scene.It is noted, of course, that the scheme that above-mentioned module relates to can be not limited in above-described embodiment
Content and scene, and above-mentioned module may operate in terminal or mobile terminal, can be realized by software or hardware.
According to embodiments of the present invention, additionally provide the terminal of a kind of acquisition methods for implementing above-mentioned account authority, as
Shown in Fig. 7, this terminal includes:
As it is shown in fig. 7, this terminal includes: one or more (only illustrating one in figure) processor 701, memorizer 703, with
And transmitting device (user interface as in above-described embodiment), it is also possible to include input/output unit as shown in Figure 7.
Wherein, memorizer 703 can be used for storing software program and module, such as the account authority in the embodiment of the present invention
Programmed instruction/module that acquisition methods is corresponding with device, the software journey that processor 701 is stored in memorizer 703 by operation
Sequence and module, thus perform the application of various function and data process, i.e. realize the acquisition methods of above-mentioned account authority.Deposit
Reservoir 703 can include high speed random access memory, it is also possible to includes nonvolatile memory, as one or more magnetic storage fills
Put, flash memory or other non-volatile solid state memories.In some instances, memorizer 703 can farther include relative to place
The memorizer that reason device 701 is remotely located, these remote memories can be connected to terminal by network.The example bag of above-mentioned network
Include but be not limited to the Internet, intranet, LAN, mobile radio communication and combinations thereof.
Above-mentioned transmitting device is for receiving via a network or sending data, it is also possible to for processor and storage
Data transmission between device.Above-mentioned network instantiation can include cable network and wireless network.In an example, transmission
Device includes a network adapter (Network Interface Controller, NIC), and they can be by netting twine and other nets
Network equipment is connected with router thus can carry out communication with the Internet or LAN.In an example, transmitting device is radio frequency
(Radio Frequency, RF) module, it is for wirelessly carrying out communication with the Internet.
Wherein, specifically, memorizer 703 is used for storing application program.
Processor 701 can call the application program of memorizer 703 storage by transmitting device, to perform following step:
Receiving inquiry request, wherein, inquiry request is for the capability identification of requesting query target account, and capability identification is used for indicating target
Account performs the authority of scheduled operation；Obtaining the browing record of target account, wherein, browing record is used for indicating target account
The text attribute of the text read；Obtain the capability identification of the target account corresponding with the text attribute of browing record instruction；Send out
Send capability identification.
Processor 701 can call the application program of memorizer 703 storage by transmitting device, to perform following step:
According to the text attribute of multiple texts that target account has been read in preset duration, determine that the attribute of target account collects letter
Breath；Attribute summary information is carried out mapping process, obtains the capability identification of target account.
Processor 701 can call the application program of memorizer 703 storage by transmitting device, to perform following step:
Text attribute represents with multiple property parameters, according to the text genus of multiple texts that target account has been read in preset duration
Property, determine that the attribute summary information of target account includes: be retrieved as the weight coefficient that each property parameters is pre-configured with；According to power
Weight coefficient, configures each property parameters of each text, obtains the property parameters value of each text；Collect each of multiple text
The property parameters value of property parameters, obtains attribute summary information.
Processor 701 can call the application program of memorizer 703 storage by transmitting device, to perform following step:
Attribute summary information includes the information of at least one dimension, wherein, attribute summary information is carried out mapping process, obtains target account
The capability identification at family includes: obtaining the regression model pre-set, wherein, in regression model, record has attribute summary information and power
The mapping relations of limit mark；Using the information of at least one dimension as the input of regression model, utilize record in regression model
Mapping relations, obtain the capability identification of target account.
Processor 701 can call the application program of memorizer 703 storage by transmitting device, to perform following step:
When text attribute includes the classification of text, the browing record obtaining target account includes: obtain the dictionary previously generated, dictionary
Middle record has the N number of classification obtained in advance；Text is carried out matching treatment with the N number of classification obtained in advance, obtains text extremely
A few classification；The probability that text belongs to each classification is recorded as the property parameters of corresponding classification, and wherein, browing record includes
The property parameters of at least one classification.
Processor 701 can call the application program of memorizer 703 storage by transmitting device, to perform following step:
Obtain multiple first words of the i-th classification preserved in dictionary, wherein, 1≤i≤N；Add up the first word to occur in the text
Word frequency and duplicate removal word number, wherein, word frequency is for representing that the total degree occurring the first word in text, duplicate removal word number are used for stating
Text occurs the number of the first word；If word frequency exceedes first threshold and duplicate removal number of times exceedes Second Threshold, it is determined that i-th
Individual classification is the classification of text.
Processor 701 can call the application program of memorizer 703 storage by transmitting device, to perform following step:
Before obtaining the dictionary previously generated, extract the second word comprised in sample；Utilize the term vector of the second word to second
Word carries out cluster operation, to be divided in M classification by the second word；The identification information of each classification is set；Preserve M class
Not, the second word of comprising of the identification information of each classification and each classification, generate dictionary.
Processor 701 can call the application program of memorizer 703 storage by transmitting device, to perform following step:
After generating dictionary, in the case of Sample Refreshment being detected, obtain the dictionary of more new samples；To the dictionary generated and renewal
The dictionary of sample merges, the dictionary after being updated.
Processor 701 can call the application program of memorizer 703 storage by transmitting device, to perform following step:
When text attribute includes the Feature Words of text, the browing record obtaining target account includes: obtains text and is comprised at least
One Feature Words；Obtaining the property parameters of each Feature Words, wherein, browing record includes the property parameters of Feature Words.
Processor 701 can call the application program of memorizer 703 storage by transmitting device, to perform following step:
Text is carried out word segmentation processing, obtains multiple key word；Multiple key words are carried out stop words process and filter operation, obtains mesh
Mark key word；Utilize the term vector of target keyword, from target keyword, filter out at least one Feature Words of text.
Processor 701 can call the application program of memorizer 703 storage by transmitting device, to perform following step:
Add up each Feature Words and word frequency in the text occurs；Word frequency number based on each Feature Words and weight coefficient generate each feature
The characteristic vector of word；Characteristic vector is recorded as the property parameters of Feature Words.
Need it is further noted that the depositor deposited in the internal memory and system processor that region is system.
It will appreciated by the skilled person that the structure shown in Fig. 7 is only signal, terminal can be smart mobile phone
(such as Android phone, iOS mobile phone etc.), panel computer, palm PC and mobile internet device (Mobile Internet
Devices, MID), the terminal unit such as PAD.Fig. 7 its structure of above-mentioned electronic installation is not caused restriction.Such as, terminal is also
Can include the assembly (such as network interface, display device etc.) more or more less than shown in Fig. 7, or have with shown in Fig. 7
Different configurations.
One of ordinary skill in the art will appreciate that all or part of step in the various methods of above-described embodiment is can
Completing carrying out the device-dependent hardware of command terminal by program, this program can be stored in a computer-readable recording medium
In, storage medium may include that flash disk, read only memory (Read-Only Memory, ROM), random access device
(RandomAccess Memory, RAM), disk or CD etc..
Embodiments of the invention additionally provide a kind of storage medium.Alternatively, in the present embodiment, above-mentioned storage medium can
For the program code performing above-mentioned information-pushing method.
Alternatively, in the present embodiment, multiple during above-mentioned storage medium may be located at the network shown in above-described embodiment
On at least one network equipment in the network equipment.
Embodiments of the invention additionally provide a kind of storage medium.Alternatively, in the present embodiment, above-mentioned storage medium can
Program code for the acquisition methods performing above-mentioned account authority.
Alternatively, in the present embodiment, multiple during above-mentioned storage medium may be located at the network shown in above-described embodiment
On at least one network equipment in the network equipment.
Alternatively, in the present embodiment, storage medium is arranged to storage for the program code performing following steps:
Receiving inquiry request, wherein, inquiry request is for the capability identification of requesting query target account, and capability identification is used for
Instruction target account performs the authority of scheduled operation；Obtaining the browing record of target account, wherein, browing record is used for indicating mesh
The text attribute of the text that mark account has been read；Obtain the authority of the target account corresponding with the text attribute of browing record instruction
Mark；Sending permission identifies.
Alternatively, in the present embodiment, storage medium is arranged to storage for the program code performing following steps: root
According to the text attribute of multiple texts that target account has been read in preset duration, determine the attribute summary information of target account；
Attribute summary information is carried out mapping process, obtains the capability identification of target account.
Alternatively, in the present embodiment, storage medium is arranged to storage for the program code performing following steps: literary composition
This attribute represents with multiple property parameters, according to the text genus of multiple texts that target account has been read in preset duration
Property, determine that the attribute summary information of target account includes: be retrieved as the weight coefficient that each property parameters is pre-configured with；According to power
Weight coefficient, configures each property parameters of each text, obtains the property parameters value of each text；Collect each of multiple text
The property parameters value of property parameters, obtains attribute summary information.
Alternatively, in the present embodiment, storage medium is arranged to storage for the program code performing following steps: belong to
Property summary information includes the information of at least one dimension, wherein, attribute summary information carries out mapping process, obtains target account
Capability identification include: obtaining the regression model that pre-sets, wherein, in regression model, record has attribute summary information and authority
The mapping relations of mark；Using the information of at least one dimension as the input of regression model, utilize what regression model recorded to reflect
Penetrate relation, obtain the capability identification of target account.
Alternatively, in the present embodiment, storage medium is arranged to storage for the program code performing following steps: when
When text attribute includes the classification of text, the browing record obtaining target account includes: obtain the dictionary previously generated, in dictionary
Record has the N number of classification obtained in advance；Text is carried out matching treatment with the N number of classification obtained in advance, obtains text at least
One classification；Text is belonged to the probability of each classification and is recorded as the property parameters of corresponding classification, wherein, browing record include to
The property parameters of a few classification.
Alternatively, in the present embodiment, storage medium is arranged to storage for the program code performing following steps: obtain
Take multiple first words of the i-th classification preserved in dictionary, wherein, 1≤i≤N；Add up the first word to occur in the text
Word frequency and duplicate removal word number, wherein, word frequency is for representing that the total degree occurring the first word in text, duplicate removal word number are used for stating literary composition
The number of the first word occurs in Ben；If word frequency exceedes first threshold and duplicate removal number of times exceedes Second Threshold, it is determined that i-th
Classification is the classification of text.
Alternatively, in the present embodiment, storage medium is arranged to storage for the program code performing following steps:
Before obtaining the dictionary previously generated, extract the second word comprised in sample；Utilize the term vector of the second word to the second word
Language carries out cluster operation, to be divided in M classification by the second word；The identification information of each classification is set；Preserve M class
Not, the second word of comprising of the identification information of each classification and each classification, generate dictionary.
Alternatively, in the present embodiment, storage medium is arranged to storage for the program code performing following steps:
After generating dictionary, in the case of Sample Refreshment being detected, obtain the dictionary of more new samples；To the dictionary generated and renewal sample
This dictionary merges, the dictionary after being updated.
Alternatively, in the present embodiment, storage medium is arranged to storage for the program code performing following steps: when
When text attribute includes the Feature Words of text, obtain target account browing record include: obtain text comprised at least one
Individual Feature Words；Obtaining the property parameters of each Feature Words, wherein, browing record includes the property parameters of Feature Words.
Alternatively, in the present embodiment, storage medium is arranged to storage for the program code performing following steps: right
Text carries out word segmentation processing, obtains multiple key word；Multiple key words are carried out stop words process and filter operation, obtains target
Key word；Utilize the term vector of target keyword, from target keyword, filter out at least one Feature Words of text.
Alternatively, in the present embodiment, storage medium is arranged to storage for the program code performing following steps: system
Count each Feature Words and word frequency in the text occurs；Word frequency number based on each Feature Words and weight coefficient generate each Feature Words
Characteristic vector；Characteristic vector is recorded as the property parameters of Feature Words.
Alternatively, the concrete example in the present embodiment is referred to the example described in above-described embodiment, the present embodiment
Do not repeat them here.
The invention described above embodiment sequence number, just to describing, does not represent the quality of embodiment.
If the integrated unit in above-described embodiment realizes and as independent product using the form of SFU software functional unit
When selling or use, can be stored in the storage medium that above computer can read.Based on such understanding, the skill of the present invention
Part that prior art is contributed by art scheme the most in other words or this technical scheme completely or partially can be with soft
The form of part product embodies, and this computer software product is stored in storage medium, including some instructions with so that one
Platform or multiple stage computer equipment (can be for personal computer, server or the network equipment etc.) perform each embodiment institute of the present invention
State all or part of step of method.
In the above embodiment of the present invention, the description to each embodiment all emphasizes particularly on different fields, and does not has in certain embodiment
The part described in detail, may refer to the associated description of other embodiments.
In several embodiments provided herein, it should be understood that disclosed client, can be by other side
Formula realizes.Wherein, device embodiment described above is only schematically, the division of the most described unit, and the most only one
Kind of logic function divides, actual can have when realizing other dividing mode, the most multiple unit or assembly can in conjunction with or
It is desirably integrated into another system, or some features can be ignored, or do not perform.Another point, shown or discussed mutual it
Between coupling direct-coupling or communication connection can be the INDIRECT COUPLING by some interfaces, unit or module or communication link
Connect, can be being electrical or other form.
The described unit illustrated as separating component can be or may not be physically separate, shows as unit
The parts shown can be or may not be physical location, i.e. may be located at a place, or can also be distributed to multiple
On NE.Some or all of unit therein can be selected according to the actual needs to realize the mesh of the present embodiment scheme
's.
It addition, each functional unit in each embodiment of the present invention can be integrated in a processing unit, it is also possible to
It is that unit is individually physically present, it is also possible to two or more unit are integrated in a unit.Above-mentioned integrated list
Unit both can realize to use the form of hardware, it would however also be possible to employ the form of SFU software functional unit realizes.
The above is only the preferred embodiment of the present invention, it is noted that for the ordinary skill people of the art
For Yuan, under the premise without departing from the principles of the invention, it is also possible to make some improvements and modifications, these improvements and modifications also should
It is considered as protection scope of the present invention.
Claims (21)
1. the acquisition methods of an account authority, it is characterised in that including:
Receiving inquiry request, wherein, described inquiry request is for the capability identification of requesting query target account, described capability identification
For indicating described target account to perform the authority of scheduled operation；
Obtaining the browing record of described target account, wherein, described browing record is for indicating described target account to read
The text attribute of text；
Obtain the capability identification of the described target account corresponding with the described text attribute of described browing record instruction；
Send described capability identification.
Method the most according to claim 1, it is characterised in that obtain and the described text attribute of described browing record instruction
The capability identification of corresponding described target account includes:
According to the text attribute of the multiple described text that described target account has been read in preset duration, determine described target account
The attribute summary information at family；
Described attribute summary information is carried out mapping process, obtains the capability identification of described target account.
Method the most according to claim 2, it is characterised in that described text attribute represents with multiple property parameters, root
According to the text attribute of the multiple described text that described target account has been read in preset duration, determine the genus of described target account
Property summary information includes:
It is retrieved as the weight coefficient that each property parameters is pre-configured with；
According to described weight coefficient, configure each property parameters of each described text, obtain the attribute ginseng of each described text
Numerical value；
Collect the property parameters value of each property parameters of multiple described text, obtain described attribute summary information.
The most according to the method in claim 2 or 3, it is characterised in that described attribute summary information includes at least one dimension
Information, wherein, described attribute summary information is carried out mapping process, the capability identification obtaining described target account includes:
Obtaining the regression model pre-set, wherein, in described regression model, record has described attribute summary information and authority mark
The mapping relations known；
Using the information of at least one dimension described as the input of described regression model, utilize what described regression model recorded to reflect
Penetrate relation, obtain the capability identification of described target account.
Method the most according to claim 1, it is characterised in that when described text attribute includes the classification of described text,
The browing record obtaining described target account includes:
Obtaining the dictionary previously generated, in described dictionary, record has the N number of classification obtained in advance；
Described text is carried out matching treatment with the N number of classification obtained in advance, obtains at least one classification of described text；
The probability that described text belongs to each described classification is recorded as the property parameters of corresponding described classification, wherein, described in read
Read record includes the property parameters of at least one classification described.
A method the most according to claim 5, it is characterised in that described text and the N number of classification obtained in advance are carried out
Joining process, at least one classification obtaining described text includes:
Obtain multiple first words of the i-th classification preserved in described dictionary, wherein, 1≤i≤N；
Adding up described first word and occur in the word frequency in described text and duplicate removal word number, wherein, described word frequency is used for representing institute
Stating the total degree occurring described first word in text, described duplicate removal word number is used for stating in described text and described first word occurs
The number of language；
If described word frequency exceedes first threshold and described duplicate removal number of times exceedes Second Threshold, it is determined that described i-th classification is institute
State the classification of text.
Method the most according to claim 5, it is characterised in that before obtaining the dictionary previously generated, described method is also
Including:
Extract the second word comprised in sample；
The term vector utilizing described second word carries out cluster operation to described second word, to be divided to by described second word
In M classification；
The identification information of each described classification is set；
Preserve the second word that described M classification, the identification information of each described classification and each described classification comprise, generate institute
Predicate storehouse.
Method the most according to claim 7, it is characterised in that after generating described dictionary, described method also includes:
In the case of Sample Refreshment being detected, obtain the dictionary of more new samples；
The dictionary of the dictionary generated and described more new samples is merged, the dictionary after being updated.
Method the most according to claim 1, it is characterised in that when described text attribute includes the Feature Words of described text
Time, the browing record obtaining described target account includes:
Obtain at least one Feature Words that described text is comprised；
Obtaining the property parameters of each described Feature Words, wherein, described browing record includes the property parameters of described Feature Words.
Method the most according to claim 9, it is characterised in that obtain at least one Feature Words that described text is comprised
Including:
Described text is carried out word segmentation processing, obtains multiple key word；
The plurality of key word is carried out stop words process and filter operation, obtains target keyword；
Utilize the term vector of described target keyword, from described target keyword, filter out at least one feature of described text
Word.
11. methods according to claim 9, it is characterised in that the property parameters obtaining each described Feature Words includes:
Add up each described Feature Words and occur in the word frequency in described text；
Word frequency number based on each described Feature Words and weight coefficient generate the characteristic vector of each described Feature Words；
Described characteristic vector is recorded as the property parameters of described Feature Words.
The acquisition device of 12. 1 kinds of account authorities, it is characterised in that including:
First receives unit, is used for receiving inquiry request, and wherein, described inquiry request is for the authority of requesting query target account
Mark, described capability identification performs the authority of scheduled operation for indicating described target account；
First acquiring unit, for obtaining the browing record of described target account, wherein, described browing record is used for indicating described
The text attribute of the text that target account has been read；
Second acquisition unit, for obtaining the described target account corresponding with the described text attribute of described browing record instruction
Capability identification；
First transmitting element, is used for sending described capability identification.
13. devices according to claim 12, it is characterised in that described second acquisition unit includes:
First determines module, and the text of the multiple described text for having read in preset duration according to described target account belongs to
Property, determine the attribute summary information of described target account；
Mapping block, for described attribute summary information is carried out mapping process, obtains the capability identification of described target account.
14. devices according to claim 13, it is characterised in that described text attribute represents with multiple property parameters,
Described first determines that module includes:
First obtains submodule, for being retrieved as the weight coefficient that each property parameters is pre-configured with；
Configuration submodule, for according to described weight coefficient, configures each property parameters of each described text, obtains each institute
State the property parameters value of text；
Collect submodule, for collecting the property parameters value of each property parameters of multiple described text, obtain described attribute and converge
Total information.
15. according to the device described in claim 13 or 14, it is characterised in that described attribute summary information includes that at least one is tieed up
The information of degree, wherein, described mapping block includes:
Second obtains submodule, and for obtaining the regression model pre-set, wherein, in described regression model, record has described genus
Property summary information and capability identification mapping relations；
First processes submodule, is used for the information of at least one dimension described as the input of described regression model, utilizes institute
State the mapping relations of record in regression model, obtain the capability identification of described target account.
16. devices according to claim 12, it is characterised in that when described text attribute includes the classification of described text
Time, described first acquiring unit includes:
First acquisition module, for obtaining the dictionary previously generated, in described dictionary, record has the N number of classification obtained in advance；
Matching module, for described text is carried out matching treatment with the N number of classification obtained in advance, obtains described text at least
One classification；
First logging modle, is recorded as the attribute of corresponding described classification for described text belongs to the probability of each described classification
Parameter, wherein, described browing record includes the property parameters of at least one classification described.
17. devices according to claim 16, it is characterised in that described matching module includes:
3rd obtains submodule, for obtaining multiple first words of the i-th classification preserved in described dictionary, wherein, 1≤i
≤N；
First statistics submodule, occurs in the word frequency in described text and duplicate removal word number for adding up described first word, wherein,
Described word frequency is for representing that the total degree occurring described first word in described text, described duplicate removal word number are used for stating described literary composition
The number of described first word occurs in Ben；
First determines submodule, if exceeding first threshold and described duplicate removal number of times exceedes Second Threshold, the most really for described word frequency
Fixed described i-th classification is the classification of described text.
18. devices according to claim 16, it is characterised in that described device also includes:
Extraction unit, for before obtaining the dictionary previously generated, extracts the second word comprised in sample；
Cluster cell, for utilizing the term vector of described second word described second word to be carried out cluster operation, with by described
Second word is divided in M classification；
Unit is set, for arranging the identification information of each described classification；
Storage unit, for preserve that described M classification, the identification information of each described classification and each described classification comprise the
Two words, generate described dictionary.
19. devices according to claim 12, it is characterised in that when described text attribute includes the Feature Words of described text
Time, described first acquiring unit includes:
Second acquisition module, for obtaining at least one Feature Words that described text is comprised；
3rd acquisition module, for obtaining the property parameters of each described Feature Words, wherein, described browing record includes described spy
Levy the property parameters of word.
20. devices according to claim 19, it is characterised in that described second acquisition module includes:
Participle submodule, for described text is carried out word segmentation processing, obtains multiple key word；
Second processes submodule, for the plurality of key word is carried out stop words process and filter operation, obtains target critical
Word；
Filter submodule, for utilizing the term vector of described target keyword, from described target keyword, filter out described literary composition
This at least one Feature Words.
21. devices according to claim 19, it is characterised in that described second acquisition module includes:
Second statistics submodule, occurs in the word frequency in described text for adding up each described Feature Words；
Generate submodule, generate the spy of each described Feature Words for word frequency number based on each described Feature Words and weight coefficient
Levy vector；
Second record sub module, for being recorded as the property parameters of described Feature Words by described characteristic vector.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610414403.XA CN106095939B (en) | 2016-06-12 | 2016-06-12 | The acquisition methods and device of account authority |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610414403.XA CN106095939B (en) | 2016-06-12 | 2016-06-12 | The acquisition methods and device of account authority |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106095939A true CN106095939A (en) | 2016-11-09 |
| CN106095939B CN106095939B (en) | 2018-04-20 |
Family
ID=57846586
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610414403.XA Active CN106095939B (en) | 2016-06-12 | 2016-06-12 | The acquisition methods and device of account authority |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106095939B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106789989A (en) * | 2016-12-09 | 2017-05-31 | 武汉斗鱼网络科技有限公司 | A kind of user right verification method and device |
| CN108629610A (en) * | 2017-03-23 | 2018-10-09 | 腾讯科技（深圳）有限公司 | The determination method and apparatus of promotion message light exposure |
| CN110244917A (en) * | 2019-04-30 | 2019-09-17 | 小濠(深圳)科技有限公司 | Portable printer authority information indicating means, device, terminal and storage medium |
| CN110263248A (en) * | 2019-05-21 | 2019-09-20 | 平安科技（深圳）有限公司 | A kind of information-pushing method, device, storage medium and server |
| CN110363206A (en) * | 2018-03-26 | 2019-10-22 | 阿里巴巴集团控股有限公司 | Cluster, data processing and the data identification method of data object |
| CN112883390A (en) * | 2021-02-18 | 2021-06-01 | 腾讯科技（深圳）有限公司 | Authority control method and device and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101493913A (en) * | 2008-01-23 | 2009-07-29 | 阿里巴巴集团控股有限公司 | Method and system for assessing user credit in internet |
| CN102163310A (en) * | 2010-02-22 | 2011-08-24 | 深圳市腾讯计算机系统有限公司 | Information pushing method and device based on credit rating of user |
| CN105138897A (en) * | 2015-08-24 | 2015-12-09 | 百度在线网络技术（北京）有限公司 | Method and device for determining user right |
| CN105631732A (en) * | 2015-12-31 | 2016-06-01 | 北京京东尚科信息技术有限公司 | Method and device for determining user authority |
-
2016
- 2016-06-12 CN CN201610414403.XA patent/CN106095939B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101493913A (en) * | 2008-01-23 | 2009-07-29 | 阿里巴巴集团控股有限公司 | Method and system for assessing user credit in internet |
| CN102163310A (en) * | 2010-02-22 | 2011-08-24 | 深圳市腾讯计算机系统有限公司 | Information pushing method and device based on credit rating of user |
| CN105138897A (en) * | 2015-08-24 | 2015-12-09 | 百度在线网络技术（北京）有限公司 | Method and device for determining user right |
| CN105631732A (en) * | 2015-12-31 | 2016-06-01 | 北京京东尚科信息技术有限公司 | Method and device for determining user authority |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106789989A (en) * | 2016-12-09 | 2017-05-31 | 武汉斗鱼网络科技有限公司 | A kind of user right verification method and device |
| CN106789989B (en) * | 2016-12-09 | 2019-08-02 | 武汉斗鱼网络科技有限公司 | A kind of user right verification method and device |
| CN108629610A (en) * | 2017-03-23 | 2018-10-09 | 腾讯科技（深圳）有限公司 | The determination method and apparatus of promotion message light exposure |
| CN108629610B (en) * | 2017-03-23 | 2022-02-08 | 腾讯科技（深圳）有限公司 | Method and device for determining popularization information exposure |
| CN110363206A (en) * | 2018-03-26 | 2019-10-22 | 阿里巴巴集团控股有限公司 | Cluster, data processing and the data identification method of data object |
| CN110363206B (en) * | 2018-03-26 | 2023-06-27 | 阿里巴巴集团控股有限公司 | Clustering of data objects, data processing and data identification method |
| CN110244917A (en) * | 2019-04-30 | 2019-09-17 | 小濠(深圳)科技有限公司 | Portable printer authority information indicating means, device, terminal and storage medium |
| CN110244917B (en) * | 2019-04-30 | 2022-11-11 | 厦门汉印电子技术有限公司 | Portable printer permission information indication method, device, terminal and storage medium |
| CN110263248A (en) * | 2019-05-21 | 2019-09-20 | 平安科技（深圳）有限公司 | A kind of information-pushing method, device, storage medium and server |
| CN110263248B (en) * | 2019-05-21 | 2023-11-28 | 平安科技（深圳）有限公司 | Information pushing method, device, storage medium and server |
| CN112883390A (en) * | 2021-02-18 | 2021-06-01 | 腾讯科技（深圳）有限公司 | Authority control method and device and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106095939B (en) | 2018-04-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106095939B (en) | The acquisition methods and device of account authority | |
| CN108363821A (en) | A kind of information-pushing method, device, terminal device and storage medium | |
| CN106815297A (en) | A kind of academic resources recommendation service system and method | |
| CN109872162B (en) | Wind control classification and identification method and system for processing user complaint information | |
| CN109767318A (en) | Loan product recommended method, device, equipment and storage medium | |
| CN107818344A (en) | The method and system that user behavior is classified and predicted | |
| CN110458324B (en) | Method and device for calculating risk probability and computer equipment | |
| CN103064987A (en) | Bogus transaction information identification method | |
| CN106503025A (en) | Method and system is recommended in a kind of application | |
| CN107545038B (en) | Text classification method and equipment | |
| CN112632405B (en) | Recommendation method, recommendation device, recommendation equipment and storage medium | |
| CN104346425A (en) | Method and system of hierarchical internet public sentiment indication system | |
| CN104834739B (en) | Internet information storage system | |
| CN111159763A (en) | System and method for analyzing portrait of law-related personnel group | |
| CN104809252A (en) | Internet data extraction system | |
| CN106294882A (en) | Data digging method and device | |
| WO2021027569A1 (en) | Law enforcement analysis method and apparatus, electronic device, and storage medium | |
| CN113420018A (en) | User behavior data analysis method, device, equipment and storage medium | |
| CN109960719A (en) | A kind of document handling method and relevant apparatus | |
| CN114819967A (en) | Data processing method and device, electronic equipment and computer readable storage medium | |
| CN105786810B (en) | The method for building up and device of classification mapping relations | |
| CN112487284A (en) | Bank customer portrait generation method, equipment, storage medium and device | |
| CN109697626A (en) | Marketing system, method, electric terminal and medium based on JeeSite quick development platform | |
| CN104731937A (en) | User behavior data processing method and device | |
| CN111784360B (en) | Anti-fraud prediction method and system based on network link backtracking |
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 |