CN109389501A - A kind of calculating equipment, computing system - Google Patents
A kind of calculating equipment, computing system Download PDFInfo
- Publication number
- CN109389501A CN109389501A CN201811151443.5A CN201811151443A CN109389501A CN 109389501 A CN109389501 A CN 109389501A CN 201811151443 A CN201811151443 A CN 201811151443A CN 109389501 A CN109389501 A CN 109389501A
- Authority
- CN
- China
- Prior art keywords
- account
- user
- target
- information
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/04—Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
Abstract
The present invention provides a kind of numerical procedures, comprising: receives source data, analyzes the source data, obtains one or more target accounts；Diagram data storage is converted in the memory for source data, the one or more essential informations and account associated activity information of the diagram data record user；The characteristic attribute information of target account described in diagram data is extracted；According to the characteristic attribute extracted, accounts all in diagram data are screened, obtain the account that there is same characteristic features attribute with target account；With the target account there is the characteristic attribute information of the account of same characteristic features attribute to extract in diagram data；The characteristic attribute information of the characteristic attribute information of the account with the target account with same characteristic features attribute and target account is subjected to similarity calculation one by one, obtains similarity value.
Description
Technical field
The present invention relates to big data analysis field, espespecially a kind of calculating equipment, computing system.
Background technique
In finance or the communications industry or network trading, travel system, big data analysis is usually utilized to user's
Potential real demand is excavated, and is analyzed by calculating equipment and computing system the account information data of user, is found
Specific target user or potential user group are carried out user using finance or the domestic demand of communication service behavior with realizing
It excavates, on the one hand recommends more personalized demand for services for user, it on the one hand can be with the illegal fraud row of identification division user
For.
In traditional proposed algorithm, calculates equipment and computing system is screened step by step by the information to target user
The mode of filtering is analyzed, and searching and user similar in target user's characteristic attribute, this iterative analysis process are complicated and numerous
It is trivial, and when data volume reaches to a certain degree or complexity is higher, analysis efficiency can become lower.
Summary of the invention
The object of the present invention is to provide a kind of calculating equipment, computing system, realize and obtain a target account by analysis
Or the target complex of multiple target account compositions, there is the account of same characteristic features attribute to carry out with target account to all in full figure data
Similarity calculation obtains the target complex formed with target account or target account most like account or account group.
Technical solution provided by the invention is as follows:
The present invention provides a kind of calculating equipment, comprising:
At least one processor；And
At least one processor, it includes computer program code, at least one processor and the computer journey
Sequence code is configured to execute the equipment at least to give an order using at least one described processor:
Source data is received, the source data includes the one or more essential informations and account associated activity information of user,
Wherein one or more essential informations of the user include the identity identification information of user, the relevant initial configuration of user identity
Information, the initial configuration information include the account information of user；
Wherein the one or more associations action message of the user includes the account that business contact is generated with the account of user
Family information；
The source data is analyzed, one or more target accounts are obtained；
Diagram data storage is converted in the memory for source data, the one or more of the diagram data record user
Essential information and account associated activity information；
The characteristic attribute information of target account described in diagram data is extracted；
According to the characteristic attribute extracted, accounts all in diagram data are screened, obtain having with target account
There is the account of same characteristic features attribute；
With the target account there is the characteristic attribute information of the account of same characteristic features attribute to extract in diagram data；
The characteristic attribute of the account with the target account with same characteristic features attribute is believed one by one by similarity algorithm
Breath and the characteristic attribute information of target account carry out similarity calculation, obtain similarity value；It is arranged by the similarity value
The account nearest with target account similarity degree is successively recommended user and confirmed by sequence.
Source data refers to financial system, communication system (such as Alipay, wechat, QQ interaction platform) or internet trading system
Stored data, the basic identity information including user, such as body in (such as Taobao, Alibaba) or travel system etc.
The data that part card number, account name, account number, each Account Registration time, each Account Registration place etc. have been acquired, the present invention is not
Limit the source of source data, it is any there are the data platform of big data analysis, the user information in platform include essential information and
Action message is all among the object of data of the present invention analysis；
Computer program code refers to the program code calculate being stored in the memory of equipment, including receives source number
According to, analysis source data, conversion source data be diagram data, storage convert after diagram data and diagram data is analyzed and is counted
The various program codes calculated；
Diagram data refers to that extracting one or more essential informations for including user and account associated activity information converts to be formed
Database, for example extract user's fund account, transfer amounts and the chart database for producing the formation of the attribute informations such as target account.
The essential information of user refers to some primary data information (pdi)s that user stores in original database, such as the user of user
Name, identity information, fund account remaining sum, mobile phone account remaining sum, or calculated number can directly be counted based on raw data base
According to, such as fund account balance changes number, mobile phone communication number in certain period.
The account associated activity information of user refers to user in daily economy, social communication, trip or network trading
The more new information of raw information, record of such as renaming, fund account remaining sum change record, fund account is transferred accounts record, mobile phone account
Remaining sum changes record, mobile phone call history etc..
Similarity calculation refers to that the characteristic attribute information for target user's essential information and/or associated activity calculates,
Obtain, attribute information similar user identical as target user's characteristic attribute.
Further, in above-mentioned calculating equipment, one or more essential informations in the source data of user may include:
The fund account remaining sum of the user；
The fund account history amount information of the user, the history amount information include highest history amount and minimum
History amount.
History amount information refers to, since the user's registration fund account, recorded in the account with obvious characteristic
Amount information, such as highest history amount, minimum history amount information.
Further, in above-mentioned calculating equipment, one or more of source data of user and the user account are closed
Connection action message includes at least:
One or more fund accounts with user generate the essential information of the account of business contact, which refers to and user
Account information of the account associated second to N user；
One or more fund accounts with user generate the funds transaction number of the account of business contact, which refers to use
Family with second to N user account carry out transaction total degree；
One or more fund accounts with user generate the funds transaction time of the account of business contact, the exchange hour
Refer to user with second to N user account carry out the time traded each time；
One or more fund accounts with user generate the funds transaction amount of the account of business contact, which refers to use
Family with second to N user account carry out the amount traded each time, when amount be timing, for be transferred to transaction, work as amount
When being negative, to produce transaction；
One or more fund accounts with user generate the funds transaction frequency of the account of business contact, which refers to use
The number traded in family and the preset time period of second to N user account progress.
Further, in above-mentioned calculating equipment, source data is analyzed by the computer program, obtains one
Or multiple target accounts include:
Source data is analyzed by selected one or more account essential informations, obtains one or more target moneys
Golden account, selected account essential information can be specific user identity card information；And/or
Source data is analyzed by selected one or more account associated activity information, obtains one or more targets
Fund account, it is a certain particular value that selected account associated activity information, which can be trading frequency,；
If acquisition is the group comprising multiple target fund accounts, the connection relationship between its each account is analyzed, as this
The new characteristic attribute of group.The connection relationship can be the interpersonal relationships of user between account.
Further, in above-mentioned calculating equipment, diagram data storage is converted in the memory for the source data,
The diagram data includes at least:
The funds transaction number of the fund account of all users and the account for generating business contact with it, that is, after referring to conversion
Diagram data record data be user's fund account, the target fund account of customer transaction, transaction the information such as number；
The funds transaction time of the fund account of all users and the account for generating business contact with it, that is, after referring to conversion
Diagram data record data be user's fund account, the target fund account of customer transaction, transaction the information such as time；
The funds transaction amount of the fund account of all users and the account for generating business contact with it, that is, after referring to conversion
Diagram data record data be user's fund account, the target fund account of customer transaction, transaction the information such as fund amount；
The funds transaction frequency of the fund account of all users and the account for generating business contact with it, that is, after referring to conversion
Diagram data record data be user's fund account, the target fund account of customer transaction, trade in preset time period when
Between etc. information.
Further, in above-mentioned calculating equipment, it is described by computer program to the feature of target account in diagram data
Attribute information, which extracts, includes:
Operation is executed using the computer program in memory, to the characteristic attribute information of target fund account in diagram data
It extracts,
Wherein the characteristic attribute information of the target account includes graph structure feature of the target account in chart database,
And/or
The identification card number of user and/or open an account date and/or outlet's address code of opening an account, and/or
The account fund transaction count and/or time and/or amount and/or frequency of business contact are generated with it.
Specifically, by target's feature-extraction algorithm (one of computer program in memory), to the spy of target account
Sign attribute information is calculated, and the characteristic attribute information can be the combination of above-mentioned arbitrary characteristics attribute information.
Further, described that accounts all in diagram data are sieved by computer program in above-mentioned calculating equipment
Choosing obtains the account for having same characteristic features attribute with target account:
The account that there is same characteristic features attribute with target account is filtered out according to the characteristic attribute of the target account, it is described
The characteristic attribute of target account includes the identification card number of user and/or the date and/or with opening an account outlet of opening an account in diagram data
Location code, and/or
The account fund transaction count and/or time and/or amount and/or frequency of business contact are generated with it.
Specifically, according to the characteristic attribute of above-mentioned target account, the account with this feature attribute is sieved from diagram data
It elects.Then the feature extraction algorithm for using target account, calculates the characteristic attribute information of the account filtered out, institute
Stating characteristic attribute information can be the combination of above-mentioned arbitrary characteristics attribute information.
It further, one by one will be with institute by the similarity algorithm in the computer program in above-mentioned calculating equipment
The characteristic attribute information of the characteristic attribute information and target account of stating account of the target account with same characteristic features attribute carries out phase
It is calculated like degree, obtaining similarity value includes:
There are in the target account comprising extraction or account group and with target account account or the account group of same alike result feature
It in the diagram data of characteristic attribute information, is calculated by similarity algorithm, obtains similarity degree value, the similarity algorithm packet
Include but be not limited to Euclidean distance, manhatton distance, Minkowski distance, cosine similarity, adjustment cosine similarity,
Jaccard coefficient, Pearson correlation coefficients etc. are optional.
Further, it in above-mentioned calculating equipment, is ranked up by similarity value, it will be with target account similarity degree most
Close account is successively recommended user and is confirmed further include:
According to the size of similarity value, will recommend with user account of the target account similarity value in certain interval range
Confirm to user.
Further, a kind of computing system is provided comprising:
At least one processing module；And
At least one memory module is at least stored with and receives source data, processing source data is diagram data, extracts feature category
The program codes such as property data, similarity calculation.
Further, in above-mentioned computing system, memory module is included at least:
At least one receiving submodule, to receive source data
At least one data converts submodule, for switching to the received source data of receiving submodule for diagram data；
At least one data extracting sub-module, for extracting target account from the diagram data of data transformant module converter
Characteristic attribute information；
At least one similarity calculation submodule, for target account to be had same genus with target account with what is filtered out
Property information account carry out similarity calculation；
At least one sub-module stored stores the similarity calculation result of similarity calculation submodule.
A kind of calculating equipment, the computing system provided through the invention, can bring it is following at least one the utility model has the advantages that
1, it in the present invention, by converting diagram data for source data, is mentioned based on characteristic information of the diagram data to user
It takes, according to target user's characteristic attribute information that analysis obtains, analysis obtains having similar features attribute information with target user
User automatically analyzes to realize input source data and target user and obtains other users similar with target user, greatly
The efficiency of data analysis is improved greatly.
2, in the present invention, by being converted into diagram data to financial industry user data, and based on diagram data to target account
Characteristic attribute information extract, to possess in full figure with target account same characteristic features attribute user carry out characteristic attribute star
System extracts, then calculates the similar value of the two, is obtained with current goal account user or user group activity most according to similar value sequence
Similar other users or user group, then during specific financial system instead cheats, realize automatic recommend and target suspect feature
The similar user of attribute information improves economic crime investigation efficiency.
Detailed description of the invention
Below by clearly understandable mode, preferred embodiment is described with reference to the drawings, to above-mentioned characteristic, the skill of a kind of YY
Art feature, advantage and its implementation are further described.
Fig. 1 is a kind of structural schematic diagram of the one embodiment for calculating equipment of the present invention；
Fig. 2 is a kind of structural schematic diagram for another embodiment for calculating equipment of the present invention.
Specific embodiment
In finance or the communications industry, big data analysis is usually utilized to excavate the potential real demand of user,
The account information data of user is analyzed by calculating equipment and computing system, finds specific target user or target
User group is excavated user using the domestic demand of finance or communication service behavior with realizing, is on the one hand user's recommendation
More personalized demand for services, on the one hand can be with the illegal fraud of identification division user.
In traditional proposed algorithm, calculates equipment and computing system is screened step by step by the information to target user
The mode of filtering is analyzed, and searching and user similar in target user's characteristic attribute, this iterative analysis process are complicated and numerous
It is trivial, and when data volume reaches to a certain degree or complexity is higher, analysis efficiency can become lower.
Body of the present invention provides a kind of equipment, a kind of system, the raw data base based on finance or the communications industry, by database
In source data be converted into the higher graph structure data of analysis efficiency, by analysis obtain target user after, in diagram data
Target user's information characteristics, behavioural characteristic, graph structure feature extract, and obtain its corresponding characteristic attribute information, further
Ground with target user there is the user of same characteristic features attribute also to carry out characteristic attribute information extraction in chart database, and by its
Similarity calculation is carried out with the characteristic attribute information of target user, is ranked up according to calculated result, is obtained living with target user
The most like other users of dynamic behavior.The present invention substantially increases Users'Data Analysis efficiency, has in finance or the communications industry
Application scenarios abundant.
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, Detailed description of the invention will be compareed below
A specific embodiment of the invention.It should be evident that drawings in the following description are only some embodiments of the invention, for
For those of ordinary skill in the art, without creative efforts, it can also be obtained according to these attached drawings other
Attached drawing, and obtain other embodiments.
To make simplified form, part related to the present invention is only schematically shown in each figure, they are not represented
Its practical structures as product.In addition, there is identical structure or function in some figures so that simplified form is easy to understand
Component only symbolically depicts one of those, or has only marked one of those.Herein, "one" is not only indicated
" only this ", can also indicate the situation of " more than one ".
First embodiment of the invention, as shown in Figure 1:
The present invention provides at least one processor；A kind of calculating equipment, comprising:
At least one processor；And
At least one processor, it includes computer program code, at least one processor and the computer journey
Sequence code is configured to execute the equipment at least to give an order using at least one described processor:
Source data is received, the source data includes the one or more essential informations and account associated activity information of user,
Wherein one or more essential informations of the user include the identity identification information of user, the relevant initial configuration of user identity
Information, the initial configuration information include the account information of user；
Wherein the one or more associations action message of the user includes the account that business contact is generated with the account of user
Family information；
The source data is analyzed, one or more target accounts are obtained；
Diagram data storage is converted in the memory for source data, the one or more of the diagram data record user
Essential information and account associated activity information；
The characteristic attribute information of target account described in diagram data is extracted；
According to the characteristic attribute extracted, accounts all in diagram data are screened, obtain having with target account
There is the account of same characteristic features attribute；
With the target account there is the characteristic attribute information of the account of same characteristic features attribute to extract in diagram data；
The characteristic attribute of the account with the target account with same characteristic features attribute is believed one by one by similarity algorithm
Breath and the characteristic attribute information of target account carry out similarity calculation, obtain similarity value；It is arranged by the similarity value
The account nearest with target account similarity degree is successively recommended user and confirmed by sequence.
Source data refers to financial system, communication system (such as Alipay, wechat, QQ interaction platform) or internet trading system
Stored data, the basic identity information including user, such as body in (such as Taobao, Alibaba) or travel system etc.
The data that part card number, account name, account number, each Account Registration time, each Account Registration place etc. have been acquired, the present invention is not
Limit the source of source data, it is any there are the data platform of big data analysis, the user information in platform include essential information and
Action message is all among the object of data of the present invention analysis.The data such as this are citing therein, are not limited to the present invention
Specific implementation range, persons skilled in the art are it is recognised that any can be with one of diagram data or several numbers
According to can all be included in technical idea of the invention in order to avoid user's clustering.
Computer program code refers to the program code calculate being stored in the memory of equipment, including receives source number
According to, analysis source data, conversion source data be diagram data, storage convert after diagram data and diagram data is analyzed and is counted
The various program codes calculated；
Diagram data refers to that extracting one or more essential informations for including user and account associated activity information converts to be formed
Database, for example extract user's fund account, transfer amounts and the chart database for producing the formation of the attribute informations such as target account.
Applicant it is emphasized that in financial system or communication system, travel system, internet trading system user's account
The above-mentioned essential information at family or the associated activity information of user are citing therein, are not limited to specific reality of the invention
Range is applied, persons skilled in the art are it is recognised that any can be with one of diagram data or several data, for side
Just user's clustering can be all included in technical idea of the invention.
The essential information of user refers to some primary data information (pdi)s that user stores in original database, such as the user of user
Name, identity information, fund account remaining sum, mobile phone account remaining sum, or calculated number can directly be counted based on raw data base
According to, such as fund account balance changes number, mobile phone communication number in certain period.
The account associated activity information of user refers to user's original letter in account caused by the activity such as daily economy, call
The more new information of breath, record of such as renaming, the variation of fund account remaining sum record, fund account is transferred accounts, and record, mobile phone account remaining sum become
Dynamic record, mobile phone call history etc..
Similarity calculation refers to that the characteristic attribute information for target user's essential information and/or associated activity calculates,
Obtain, attribute information similar user identical as target user's characteristic attribute.
Specifically, in the present invention, source data is obtained by the external interface in equipment, the source data can be text
Data, social data, relation data, real-time streaming data and system log etc., the information for including in source data have the user of user
The data such as name, identification card number, account name, account number, each Account Registration time, each Account Registration place.
After obtaining source data by external interface, using the corresponding program in memory, source data is analyzed, is obtained
Take one or more target users.
After obtaining source data by external interface, using the corresponding program in memory, source data is converted, it will
It is converted into graph structure data, and diagram data is stored in memory, in current diagram data, the one or more of user
Essential information and account associated activity information are all stored.For example, the account name of user, user account is associated movable specific
Content, in one embodiment, user account are mobile phone account, and activity description is call, and the target account of associated activity is another
One mobile phone account, three factors respectively constitute node, side, the terminal of diagram data.
After equipment completes the importing and conversion of source data, using the feature extraction program in memory, to mesh in diagram data
The essential information feature and associated activity information characteristics for marking user extract, when target user is the group formed to each account
When, the relationship group members is analyzed, and by this relationship characteristic, be recorded as new characteristic attribute information.
Specifically, in the first embodiment, target user's information can be the mobile phone account information of user, then,
Assuming that member identities have been identified in some telecommunication fraud clique, then the process analyzed source data, it is possible to specify
The attribute that a certain item of user has been known in source database positions it, such as according to the identification card number information searching of user
To target user.
Next, by the duration of call of the user, the target account of call, air time section, voice frequency etc. as special
Attribute is levied, the characteristic attribute information of the user is calculated, then screens all users with above-mentioned attribute from chart database
Out.Then, using feature extraction algorithm, the feature information extraction of the user filtered out or user group are come out, and carry out phase
It is calculated like degree, hereby it is achieved that the target user analyzed according to one or potential user group characteristic information, sentence a section diagram data
In which user may be similar with target user.
Second embodiment of the invention, as shown in Fig. 2, in a kind of calculating equipment, one or more in the source data of user
A essential information may include:
The fund account remaining sum of the user；
The fund account history amount information of the user, the history amount information include highest history amount and minimum
History amount.
History amount information refers to, since the user's registration fund account, recorded in the account with obvious characteristic
Amount information, such as highest history amount, minimum history amount information.
In second embodiment, in above-mentioned calculating equipment, one or more of source data of user and user's account
Family associated activity information includes at least:
One or more fund accounts with user generate the essential information of the account of business contact, which refers to and user
Account information of the account associated second to N user；
One or more fund accounts with user generate the funds transaction number of the account of business contact, which refers to use
Family with second to N user account carry out transaction total degree；
One or more fund accounts with user generate the funds transaction time of the account of business contact, the exchange hour
Refer to user with second to N user account carry out the time traded each time；
One or more fund accounts with user generate the funds transaction amount of the account of business contact, which refers to use
Family with second to N user account carry out the amount traded each time, when amount be timing, for be transferred to transaction, work as amount
When being negative, to produce transaction；
One or more fund accounts with user generate the funds transaction frequency of the account of business contact, which refers to use
The number traded in family and the preset time period of second to N user account progress.
In second embodiment, in above-mentioned calculating equipment, source data is analyzed by the computer program, is obtained
One or more target accounts include:
Source data is analyzed by selected one or more account essential informations, obtains one or more target moneys
Golden account, selected account essential information can be specific user identity card information；And/or
Source data is analyzed by selected one or more account associated activity information, obtains one or more targets
Fund account, it is a certain particular value that selected account associated activity information, which can be trading frequency,；
If acquisition is the group comprising multiple target fund accounts, the connection relationship between its each account is analyzed, as this
The new characteristic attribute of group.The connection relationship can be the interpersonal relationships of user between account.
In second embodiment, in above-mentioned calculating equipment, diagram data is converted by the source data and is stored in the storage
In device, the diagram data is included at least:
The funds transaction number of the fund account of all users and the account for generating business contact with it, that is, after referring to conversion
Diagram data record data be user's fund account, the target fund account of customer transaction, transaction the information such as number；
The funds transaction time of the fund account of all users and the account for generating business contact with it, that is, after referring to conversion
Diagram data record data be user's fund account, the target fund account of customer transaction, transaction the information such as time；
The funds transaction amount of the fund account of all users and the account for generating business contact with it, that is, after referring to conversion
Diagram data record data be user's fund account, the target fund account of customer transaction, transaction the information such as fund amount；
The funds transaction frequency of the fund account of all users and the account for generating business contact with it, that is, after referring to conversion
Diagram data record data be user's fund account, the target fund account of customer transaction, trade in preset time period when
Between etc. information.
In second embodiment, in above-mentioned calculating equipment, it is described by computer program to target account in diagram data
Characteristic attribute information, which extracts, includes:
Operation is executed using the computer program in memory, to the characteristic attribute information of target fund account in diagram data
It extracts,
Wherein the characteristic attribute information of the target account includes graph structure feature of the target account in chart database,
And/or
The identification card number of user and/or open an account date and/or outlet's address code of opening an account, and/or
The account fund transaction count and/or time and/or amount and/or frequency of business contact are generated with it.
Specifically, by target's feature-extraction algorithm (one of computer program in memory), to the spy of target account
Sign attribute information is calculated, and the characteristic attribute information can be the combination of above-mentioned arbitrary characteristics attribute information.In a reality
It applies in example, is extracted using graph structure feature of the node2vec algorithm to target；In another embodiment, target is extracted to use
The information such as time of opening an account at family, direction of transferring accounts, user identity card number, calculate in some preset time, the time of opening an account is close,
It transfers accounts and user similar in direction and recommends.
In second embodiment, in above-mentioned calculating equipment, it is described by computer program to accounts all in diagram data into
Row screening, obtains the account for having same characteristic features attribute with target account:
The account that there is same characteristic features attribute with target account is filtered out according to the characteristic attribute of the target account, it is described
The characteristic attribute of target account includes graph structure feature of the account in chart database in diagram data, and/or
The identification card number of user and/or open an account date and/or outlet's address code of opening an account, and/or
The account fund transaction count and/or time and/or amount and/or frequency of business contact are generated with it.
Specifically, according to the characteristic attribute of above-mentioned target account, the account with this feature attribute is sieved from diagram data
It elects.Then the feature extraction algorithm for using target account, calculates the characteristic attribute information of the account filtered out, institute
Stating characteristic attribute information can be the combination of above-mentioned arbitrary characteristics attribute information, for example, in one embodiment, using
Node2vec algorithm extracts the graph structure feature of target；In another embodiment, user and it is one or more other
The account of user interacts, by identity information, transfer amounts, number or the time therein for handing over easy-to-use target account and/
Or amount and/or frequency carry out feature information extraction as characteristic attribute, so that it may obtain and target account in identity information, turn
The account amount of money, number or time and/or the similar other users of amount and/or frequency.
In second embodiment, in above-mentioned calculating equipment, one by one will by the similarity algorithm in the computer program
With the target account have same characteristic features attribute account characteristic attribute information and target account characteristic attribute information into
Row similarity calculation, obtaining similarity value includes:
There are in the target account comprising extraction or account group and with target account account or the account group of same alike result feature
It in the diagram data of characteristic attribute information, is calculated by similarity algorithm, obtains similarity degree value, the similarity algorithm packet
Include but be not limited to Euclidean distance, manhatton distance, Minkowski distance, cosine similarity, adjustment cosine similarity,
Jaccard coefficient, Pearson correlation coefficients etc. are optional.
In second embodiment, in above-mentioned calculating equipment, it is ranked up by similarity value, it will journey similar to target account
It spends nearest account and successively recommends user and confirm further include:
According to the size of similarity value, will recommend with user account of the target account similarity value in certain interval range
Confirm to user.
Third embodiment of the invention provides a kind of computing system comprising:
At least one processing module；And
At least one memory module is at least stored with and receives source data, processing source data is diagram data, extracts feature category
The program codes such as property data, similarity calculation.
Further, in above-mentioned computing system, memory module is included at least:
At least one receiving submodule, to receive source data
At least one data converts submodule, for switching to the received source data of receiving submodule for diagram data；
At least one data extracting sub-module, for extracting target account from the diagram data of data transformant module converter
Characteristic attribute information；
At least one similarity calculation submodule, for target account to be had same genus with target account with what is filtered out
Property information account carry out similarity calculation；
At least one sub-module stored stores the similarity calculation result of similarity calculation submodule.
Specifically, the present embodiment is the system embodiment of above-mentioned first embodiment, and principle is similar, and effect is referring to above-mentioned the
One embodiment, therefore not to repeat here.
It should be noted that above-described embodiment can be freely combined as needed.The above is only of the invention preferred
Embodiment, it is noted that for those skilled in the art, in the premise for not departing from the principle of the invention
Under, several improvements and modifications can also be made, these modifications and embellishments should also be considered as the scope of protection of the present invention.
Claims (10)
1. a kind of calculating equipment characterized by comprising
At least one processor；And
At least one processor, it includes computer program code, at least one processor and the computer program generation
Code is configured to execute the equipment at least to give an order using at least one described processor:
Source data is received, the source data includes the one or more essential informations and account associated activity information of user, wherein
One or more essential informations of the user include the identity identification information of user, the relevant initial configuration letter of user identity
Breath, the initial configuration information includes the account information of user；
Wherein the one or more associations action message of the user includes the account letter that business contact is generated with the account of user
Breath；
The source data is analyzed, one or more target accounts are obtained；
Diagram data storage is converted in the memory for source data, and the one or more of the diagram data record user is basic
Information and account associated activity information；
The characteristic attribute information of target account described in diagram data is extracted；
According to the characteristic attribute extracted, accounts all in diagram data are screened, obtain that there is phase with target account
With the account of characteristic attribute；
With the target account there is the characteristic attribute information of the account of same characteristic features attribute to extract in diagram data；
By similarity algorithm one by one by with the target account have same characteristic features attribute account characteristic attribute information with
The characteristic attribute information of target account carries out similarity calculation, obtains similarity value；It is ranked up by the similarity value, it will
Nearest account is successively recommended user and is confirmed with target account similarity degree.
2. a kind of calculating equipment according to claim 1, which is characterized in that one or more in the source data of the user
A essential information includes:
The fund account remaining sum of the user；
The fund account history amount information of the user, the history amount information includes highest history amount and minimum history
Amount.
3. a kind of calculating equipment according to claim 2, which is characterized in that one or more in the source data of the user
A and described user account associated activity information includes:
The essential information of the account of one or more accounts generation business contacts with user,
The number of communications or funds transaction number of the account of one or more accounts generation business contacts with user,
The funds transaction time of the account of one or more accounts generation business contacts with user,
One or more accounts with user generate the account of business contacts funds transaction amount or
The funds transaction frequency of the account of one or more accounts generation business contacts with user,
Or the action message of one or more users itself.
4. a kind of calculating equipment according to claim 2, which is characterized in that it is described by the computer program to source number
According to being analyzed, obtaining one or more target accounts includes:
Source data is analyzed by selected one or more account essential informations, obtains one or more target account of the exchequer
Family；And/or
Source data is analyzed by selected one or more account associated activity information, obtains one or more target funds
Account；
If acquisition is the group comprising multiple target fund accounts, the connection relationship between its each account is analyzed, it is new as the group
Characteristic attribute.
5. a kind of calculating equipment according to claim 2, which is characterized in that convert diagram data storage for the source data
In the memory, the diagram data includes:
The funds transaction number of the fund account of all users and the account for generating business contact with it；
The funds transaction time of the fund account of all users and the account for generating business contact with it；
The funds transaction amount of the fund account of all users and the account for generating business contact with it；
The funds transaction frequency of the fund account of all users and the account for generating business contact with it.
6. a kind of calculating equipment according to claim 2, which is characterized in that the feature to target account in diagram data
Attribute information, which extracts, includes:
Operation is executed using the computer program in memory, the characteristic attribute information of target fund account in diagram data is carried out
It extracts,
Wherein the characteristic attribute information of the target account includes graph structure feature of the target account in chart database, and/or
The identification card number of user and/or open an account date and/or outlet's address code of opening an account, and/or
The account fund transaction count and/or time and/or amount and/or frequency of business contact are generated with it.
7. a kind of calculating equipment according to claim 2, which is characterized in that described to be sieved to accounts all in diagram data
Choosing obtains the account for having same characteristic features attribute with target account:
The account that there is same characteristic features attribute with target account, the figure number are filtered out according to the characteristic attribute of the target account
Characteristic attribute information according to middle target account includes graph structure feature of the target account in chart database, and/or
The identification card number of user and/or open an account date and/or outlet's address code of opening an account, and/or
The account fund transaction count and/or time and/or amount and/or frequency of business contact are generated with it.
8. a kind of calculating equipment according to claim 2, which is characterized in that it is described one by one will be with institute by similarity algorithm
The characteristic attribute information of the characteristic attribute information and target account of stating account of the target account with same characteristic features attribute carries out phase
It is calculated like degree, obtaining similarity value includes:
There are in the target account comprising extraction or account group and with target account the account or account group character of same alike result feature
In the diagram data of attribute information, calculated by similarity algorithm, obtain similarity degree value, the similarity algorithm include but
Be not limited to Euclidean distance, manhatton distance, Minkowski distance, cosine similarity, adjustment cosine similarity,
Jaccard coefficient, Pearson correlation coefficients etc. are optional.
9. a kind of calculating equipment according to claim 2, which is characterized in that it is described to be ranked up by similarity value, it will
Nearest account is successively recommended user and is confirmed with target account similarity degree further include:
According to the size of similarity value, use will be recommended with user account of the target account similarity value in certain interval range
Family confirmation.
10. a kind of computing system, characterized in that it comprises:
At least one processing module；And
At least one memory module is at least stored with and receives source data, processing source data is diagram data, extracts characteristic attribute number
According to program codes such as, similarity calculations.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811151443.5A CN109389501A (en) | 2018-09-29 | 2018-09-29 | A kind of calculating equipment, computing system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811151443.5A CN109389501A (en) | 2018-09-29 | 2018-09-29 | A kind of calculating equipment, computing system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109389501A true CN109389501A (en) | 2019-02-26 |
Family
ID=65419155
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811151443.5A Pending CN109389501A (en) | 2018-09-29 | 2018-09-29 | A kind of calculating equipment, computing system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109389501A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110675196A (en) * | 2019-09-27 | 2020-01-10 | 中国工商银行股份有限公司 | User identification method and device, electronic equipment and storage medium |
| CN110706111A (en) * | 2019-10-15 | 2020-01-17 | 支付宝(杭州)信息技术有限公司 | Method and device for identifying suspicious transaction account, storage medium and server |
| CN111258750A (en) * | 2020-01-07 | 2020-06-09 | 支付宝(杭州)信息技术有限公司 | Data volume processing method and system, and quota allocation method and system |
| CN112016797A (en) * | 2020-07-15 | 2020-12-01 | 北京淇瑀信息科技有限公司 | KNN-based resource quota adjusting method and device and electronic equipment |
| CN113626681A (en) * | 2020-05-08 | 2021-11-09 | 北京达佳互联信息技术有限公司 | Account information processing method and device, electronic equipment and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160210707A1 (en) * | 2009-01-20 | 2016-07-21 | Domos, Llc | Computer-implemented systems and methods for simulating large scale automatic data combinations |
| CN106534164A (en) * | 2016-12-05 | 2017-03-22 | 公安部第三研究所 | Cyberspace user identity-based effective virtual identity description method in computer |
| CN108280755A (en) * | 2018-02-28 | 2018-07-13 | 阿里巴巴集团控股有限公司 | The recognition methods of suspicious money laundering clique and identification device |
-
2018
- 2018-09-29 CN CN201811151443.5A patent/CN109389501A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160210707A1 (en) * | 2009-01-20 | 2016-07-21 | Domos, Llc | Computer-implemented systems and methods for simulating large scale automatic data combinations |
| CN106534164A (en) * | 2016-12-05 | 2017-03-22 | 公安部第三研究所 | Cyberspace user identity-based effective virtual identity description method in computer |
| CN108280755A (en) * | 2018-02-28 | 2018-07-13 | 阿里巴巴集团控股有限公司 | The recognition methods of suspicious money laundering clique and identification device |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110675196A (en) * | 2019-09-27 | 2020-01-10 | 中国工商银行股份有限公司 | User identification method and device, electronic equipment and storage medium |
| CN110706111A (en) * | 2019-10-15 | 2020-01-17 | 支付宝(杭州)信息技术有限公司 | Method and device for identifying suspicious transaction account, storage medium and server |
| CN110706111B (en) * | 2019-10-15 | 2022-11-01 | 支付宝(杭州)信息技术有限公司 | Method and device for identifying suspicious transaction account, storage medium and server |
| CN111258750A (en) * | 2020-01-07 | 2020-06-09 | 支付宝(杭州)信息技术有限公司 | Data volume processing method and system, and quota allocation method and system |
| CN111258750B (en) * | 2020-01-07 | 2023-06-20 | 支付宝(杭州)信息技术有限公司 | Data volume processing method and system, and credit allocation method and system |
| CN113626681A (en) * | 2020-05-08 | 2021-11-09 | 北京达佳互联信息技术有限公司 | Account information processing method and device, electronic equipment and storage medium |
| CN113626681B (en) * | 2020-05-08 | 2023-09-26 | 北京达佳互联信息技术有限公司 | Account information processing method and device, electronic equipment and storage medium |
| CN112016797A (en) * | 2020-07-15 | 2020-12-01 | 北京淇瑀信息科技有限公司 | KNN-based resource quota adjusting method and device and electronic equipment |
| CN112016797B (en) * | 2020-07-15 | 2024-03-01 | 北京淇瑀信息科技有限公司 | KNN-based resource quota adjustment method and device and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109389501A (en) | A kind of calculating equipment, computing system | |
| CN110070364A (en) | Method and apparatus, storage medium based on the fraud of graph model detection clique | |
| CN110782335B (en) | Method, device and storage medium for processing credit data based on artificial intelligence | |
| CN109145050B (en) | Computing device | |
| CN105389488B (en) | Identity identifying method and device | |
| US20100100443A1 (en) | User classification apparatus, advertisement distribution apparatus, user classification method, advertisement distribution method, and program used thereby | |
| CN110020873A (en) | Artificial customer service switching method, relevant apparatus, equipment and computer-readable medium | |
| CN102567534B (en) | Interactive product user generated content intercepting system and intercepting method for the same | |
| CN108989581A (en) | A kind of consumer's risk recognition methods, apparatus and system | |
| CN111754337A (en) | Method and system for identifying credit card maintenance contract group | |
| CN109284589A (en) | Across the social networks entity identities analytic method of one kind | |
| CN109767225A (en) | A kind of network payment fraud detection method based on self study sliding time window | |
| CN114398512A (en) | Big data-based voice portrait analysis method for communication operator business customer | |
| CN107368499A (en) | A kind of client's tag modeling and recommendation method and device | |
| CN108090787A (en) | A kind of call bill data depth based on Apriori algorithm is excavated and the method for user's behavior prediction | |
| CN110347566A (en) | For carrying out the method and device of measures of effectiveness to registration air control model | |
| CN114066420A (en) | Work order distribution method, system, electronic equipment and computer readable medium | |
| CN106791230A (en) | Telephone number recognition methods and device | |
| CN106293354A (en) | Shortcut menu self adaptation display control method, server and portable terminal | |
| CN111062422B (en) | Method and device for identifying set-way loan system | |
| CN109711984B (en) | Pre-loan risk monitoring method and device based on collection urging | |
| CN105225116A (en) | The recognition methods of transactional operation and server | |
| CN110288467B (en) | Data mining method and device, electronic equipment and storage medium | |
| CN111126503A (en) | Training sample generation method and device | |
| CN105824920A (en) | Information display method and method for stopping information display |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190226 |
|
| RJ01 | Rejection of invention patent application after publication |