CN106097095A - Determine the method and device of credit - Google Patents

Abstract

The invention discloses a kind of method and device determining credit；Method includes: obtaining geographical position sequence based on data source, described geographical position sequence includes that user is in the geographical position residing for different time；Geographical position in the sequence of described geographical position is clustered and obtains place；The described geographical position corresponding based on each described place and time determine the geographical attribute in each described place；Geographical attribute based on each described place and corresponding time build geographical attribute sequence；Geographical attribute corresponding for each place in described geographical attribute sequence and time are carried out the first mapping and processes the credit feature of at least one dimension obtaining described user；The credit feature of at least one dimension based on described user carries out the second mapping and processes the credit obtaining described user.Implement the present invention, it is possible to the information in the sequence of geographical position of excavating determines the credit of user comprehensively.

Description

Determine the method and device of credit

Technical field

The present invention relates to field of computer technology, particularly relate to a kind of method and device determining credit.

Background technology

Credit investigation system is the credibility record set up for user, is available to Ge Jia bank, data subject, finance prison Pipe mechanism, judicial department and other government organs use.The activity of this shared credit information is exactly reference.Credit investigation system one Aspect is the instrument prevented financial risk, and plays the effect safeguarding financial stability, on the other hand also plays propelling social credibility The effect of Establishing.

Credit currently for evaluation user is based primarily upon collage-credit data (such as Bank Account Number flowing water, consumer record, credit Record etc.), user shopping platform behavioral data with income situation evaluation user credit.

But, correlation technique evaluation user credit exists (cannot evaluate the letter of all groups to the coverage rate of crowd is low With) and collage-credit data do not affect the accuracy problems of evaluation credit comprehensively.

Summary of the invention

The present invention is that the problems referred to above at least solving correlation technique existence provide a kind of method and device determining credit.

The technical scheme is that and be achieved in that:

According to the first aspect of the invention, it is provided that a kind of method determining credit, described method includes:

Obtaining geographical position sequence based on data source, described geographical position sequence includes that user is on the ground residing for different time Reason position；

Geographical position in the sequence of described geographical position is clustered and obtains place；

The described geographical position corresponding based on each described place and time determine the geographical attribute in each described place；

Geographical attribute based on each described place and corresponding time build geographical attribute sequence；

Geographical attribute corresponding for each place in described geographical attribute sequence and time carry out the first mapping process obtain The credit feature of at least one dimension of described user；

The credit feature of at least one dimension based on described user carries out the second mapping and processes the letter obtaining described user With.

According to an aspect of the present invention, it is provided that a kind of device determining credit, described device includes:

Data acquisition module, for obtaining geographical position sequence based on data source, described geographical position sequence includes user In the geographical position residing for different time；

Geographical attribute constructing module, obtains place for being clustered in the geographical position in the sequence of described geographical position；

Described geographical attribute constructing module, is additionally operable to the described geographical position corresponding based on each described place and the time is true The geographical attribute in fixed each described place；Geographical attribute based on each described place and corresponding time build geographical attribute sequence；

Described geographical attribute constructing module, be additionally operable to by geographical attribute corresponding for each place in described geographical attribute sequence with And the time carries out the first mapping and processes the credit feature of at least one dimension obtaining described user；

Credit mapping block, the credit feature at least one dimension based on described user carries out the second mapping process Obtain the credit of described user.

The method have the advantages that the attribute using the data in geographical position of user to determine different location, base Attribute in different location maps the credit feature of (the first mapping processes) user, and then maps (the second mapping processes) user's Credit, on the one hand achieve the multi-angle to user's life time cover thus improve further credit scoring accuracy and can By property, on the other hand widen the most further and can evaluate the border that the crowd of credit covers.

Accompanying drawing explanation

Fig. 1 is an optional hardware architecture diagram of the device determining credit in the embodiment of the present invention；

Fig. 2-1 is the optional application scenarios schematic diagram determining information in the embodiment of the present invention；

Fig. 2-2 is the optional application scenarios schematic diagram determining information in the embodiment of the present invention；

Fig. 3 is an optional schematic flow sheet of the method determining credit in the embodiment of the present invention；

Fig. 4 is another the optional schematic flow sheet of the method determining credit in the embodiment of the present invention；

Fig. 5 is another the optional schematic flow sheet of the method determining credit in the embodiment of the present invention；

Fig. 6 is another the optional schematic flow sheet of the method determining credit in the embodiment of the present invention；

Fig. 7 be in the embodiment of the present invention one of matrix elasticity partition-merge optionally realize schematic diagram；

Fig. 8 be in the embodiment of the present invention multidimensional data matrix be split as position data module and multidimensional location data module Schematic diagram；

Fig. 9 is the optional schematic flow sheet extracting static geographical attribute in the embodiment of the present invention；

Figure 10 is the optional schematic flow sheet extracting Dynamic Geographic attribute in the embodiment of the present invention；

Figure 11 is an optional structural representation of the device determining credit in the embodiment of the present invention.

Detailed description of the invention

First the problem that there is the mode of correlation technique evaluation user credit illustrates, and inventor is implementing the present invention During find, the credit evaluating user in correlation technique at least there is problems in that

1) crowd that collage-credit data covers is limited

As a example by tradition reference mode, it is based primarily upon finance activities on the line such as passing credit record of bank's flowing water and user Data evaluation credit, the data class of use is less, data volume is little and low frequency, and collage-credit data is low to the coverage rate of crowd, Owing to current a considerable amount of crowds do not have collage-credit data, these crowds not having collage-credit data cannot be evaluated credit.

2) accuracy of credit evaluation is limited

As a example by behavioral data based on the Internet evaluation credit, based on user's behavioral data evaluation letter in the Internet With, such as right at the behavioral data (as done shopping, collect, paying, the data of the behavior such as goods browse) of shopping platform based on user User carries out credit evaluation.

Again as a example by income situation based on user evaluation credit, based on personal profession income, industry average income, region The statistics class data such as average income assess the credit grade of certain user, collect and explicitly to use that income/expenditure data are come right The repaying ability of user is estimated, and then calculates credit risk.

The analysis of summary can be found out with volume, and correlation technique is at bank's flowing water, credit record and shopping etc. and money The aspect of correlated activation collects collage-credit data, and collected collage-credit data exists the defect that kind is single, data volume is few, due to only Certain customers can collect collage-credit data (such as to the collage-credit data that cannot collect the above-mentioned type with substantial portion of rural subscribers), The coverage rate causing credit evaluation is low, and has the credit evaluation of user because collage-credit data does not affects accuracy comprehensively Problem.

The collage-credit data that correlation technique is collected is user to be engaged in the activity directly related with money such as shopping and finance enter Row data collection obtains, and these activities only occupy the least ratio in the life cycle of user, therefore only to user The moving collection collage-credit data of the very small scale of life cycle.

Inventor finds in the practice of the invention, in the life cycle of user with shopping and finance etc. and golden Money does not has activity or the information of direct correlation, such as geographical position (such as residence, place of working, public place of entertainment), the outdoor fortune of user The features relevant to the credit of user such as dynamic loan repayment capacity and the refund wish that can indirectly reflect user, and the geography of user The time (being in time during this geographical position) of position and correspondence is can be for the major part of whole mobile phone users Life cycle is collected easily.

As a example by collecting geographical position based on mobile terminal, the popularity rate of mobile terminal nearly reaches 100%, beyond relevant Technology can be collected the ratio of the user of collage-credit data, (and corresponding to the geographical position of some region of mobile phone users Time) be collected can be considered that the geographical position of the whole crowds to this area is collected, mobile phone users use base Also carry out beyond user in the time such as social networking application of applying of location-based service (LBS, Location Based Service) Finance, the time of consumption activity, use geographical position and the time, relatively correlation technique of application by obtaining mobile phone users Realize collection to subscriber lifecycle more multi-activity, geographic position datas based on these activities (include geographical position with And the time of correspondence) feature relevant to the credit of user can be obtained more comprehensively.

Thus, once obtain the position data of user, then can (give birth to beyond user by most of life cycle based on user The life cycle is engaged in finance, the time of consumption activity, is used for doing shopping because the life cycle of user only has a small amount of time And finance activities) in the residing static geographical attribute in geographical position, between diverse geographic location the Dynamic Geographic attribute of movement Evaluation user credit, the coverage rate of the user of evaluation credit is wider compared with the coverage rate of correlation technique reference user (mobile to gather As a example by terminal use uses the geographical position of wechat, for substantial portion of wechat user such as urban residents and middle-aged and elderly people, only Position data can be collected and collage-credit data cannot be collected), therefore, the geographical position gathered based on the most of life cycle user Putting the credit that data are evaluated, the credit of the collage-credit data evaluation relatively gathered at the fraction life cycle of user can be more comprehensive Accurately.

Before being further elaborated the present invention, noun and term to relating in the embodiment of the present invention are said Bright, the noun related in the embodiment of the present invention and term are applicable to following explanation.

1) collage-credit data, for evaluating the data of credit of user, including the finance activities data of user (such as Bank Account Number Flowing water, credit record), the behavioral data of shopping platform (as done shopping, collect, paying, the data of the behavior such as goods browse) and receiving Enter situation (zone leveling income as residing for the personal income of user, user institute engaged in trade average income, user).

2) credit, can be continuous print score value (credit scoring), it is also possible to be discrete credit grade, can be used for weighing and uses The credit risk at family and fulfil the ability repaid the loan, the credit of user is the highest, then the ability repaid the loan is the highest, promise breaking Probability is the lowest.

3) geographical position, user is location when using based on location-based service, and longitude and latitude etc. can be used arbitrarily may be used Characterize in the way of calibration position.

4) place, obtains after clustering geographical position, and place can be a geographical position, it is also possible to be a region.

5) geographical position sequence, is the sequence that basic element is constituted with " time m-geographical position ", and geographical position refers to user Use based on location during location-based service, the time refer to user's time in this position (can be a certain moment, it is possible to Think that one is the time period).Exemplarily, use " geographical location marker-time " such form recording geographical position and Time.

5) geographical attribute, including:

Static geographical attribute (static geographical attribute): the classification corresponding to several places that user often haunts belongs to Property.

Dynamic geographical attribute (Dynamic Geographic attribute): user's motion track pattern between multiple places is (such as place B-place, A-place C), or space-time migration model (such as, " every morning 6 adheres to outdoor activity ", " every night about 8 come off duty Driving is gone home " etc.)；In the classification system of the different mode pre-build, to same pattern (motion track pattern and space-time Migration model) carry out the division (description) of multiple dimension and give respective labels, obtain motion track pattern or space-time migrates mould The multidimensional of formula describes.

6) application (App): be often referred to the application software on equipment (such as smart mobile phone) in the narrow sense, also refers to all and calculates All application softwaries and son thereof outside the upper division operation system of machine equipment (containing PC, mobile terminal, cloud computing sever platform etc.) are soft Part (such as plug-in unit).

Below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that mentioned herein Embodiment only in order to explain the present invention, is not intended to limit the present invention.It addition, embodiment provided below is for implementing The section Example of the present invention, rather than the whole embodiments implementing the present invention are provided, in the case of not conflicting, the present invention implements The technical scheme that example is recorded can mode in any combination be implemented.

" first second the 3rd " is only that difference is similar it should be noted that term involved by the embodiment of the present invention Object, do not represent the particular sorted for object, it is possible to understand that ground, " first second the 3rd " is permissible in the case of allowing Exchange specific order or precedence.Should be appreciated that the object that " first second the 3rd " is distinguished in the appropriate case can be mutual Change, so that embodiments of the invention described herein can be real with the order in addition to those here illustrating or describing Execute.

The embodiment of the present invention can be provided as determining the method for credit and determining the device of credit, in actual application, determines Each functional module in the device of credit can by the hardware resource of equipment (such as terminal unit, server or server cluster), Resource, the communication resource (communicating as realized the various mode such as optical cable, honeycomb for support) cooperative achievement is calculated such as processor etc..Figure The 1 optional hardware architecture diagram illustrating equipment 10, including processor 11, input/output interface 13 (example One or more as in display, keyboard, touch screen, Speaker Microphone), storage medium 14 and network interface 12, group Part can connect communication through system bus 15.

Certainly, the embodiment of the present invention is not limited to offer method and hardware.Such as, as shown in Fig. 2-1, in actual application, The embodiment of the present invention can be provided as performing credit and determine that the software function module of method (includes a series of determining for performing credit The executable instruction of method), to be coupled to existing any application such as social networking application, credit application etc., answering for user Querying individual credit during with, certainly, except realizing in the way of software function module, as shown in Fig. 2-2, the present invention is real Execute example and also can be provided separately as credit evaluation platform, in a specific way such as application programming interfaces (API, Application Interface), plug-in unit or software development kit (SDK, Soft Development Toolkit) mode provide public credit Calling of inquiry service, makes a credit inquiry for enterprise, entity and individual.

It is pointed out that the software function module that the embodiment of the present invention is provided can be embedded into comprehensive social networks number According to reference marking system in, be combined the credit determining user, a side with other credit scoring strategies in reference marking system Face, credit scoring model based on position data is the important supplement of all multi-models based on other social network data, permissible Significantly improve accuracy and the reliability of reference marking system output result；On the other hand, from the point of view of certain class user, if should The social network data of class user is the most sparse, do not cover, if now geographic position data has certain abundance, permissible Rely primarily on this type of user credit evaluation of output of credit scoring model based on position data.Certainly, based on geographical position number According to credit scoring model when being used alone, it may have excellent credit scoring ability, for credit operation, can take out Taking the credit scoring higher head user candidate crowd as credit, the crowd expanding general favour finance covers border.

The embodiment of the present invention determines that the credit of user can include three phases in the following ways:

The first, geographical position sequence is formed.

The second, intermediate object program namely the geographical attribute for determining user credit is obtained based on geographical position sequence, including Static geographical attribute and Dynamic Geographic attribute.

3rd, based on intermediate object program carry out mapping process (exemplarily, in the embodiment of the present invention follow-up to carry out successively One mapping processes and the second mapping processes as a example by twice mapping processes and illustrates) obtain the credit of user.

Below in conjunction with the optional schematic flow sheet of method of the determination credit shown in Fig. 3 to the above three stage Process illustrates, and determines that the method for the credit of user at least comprises the following steps in figure 3.

Step 101, obtains geographical position sequence based on data source.

Geographical position sequence includes that user is in the geographical position residing for different time.

Step 102, clusters the geographical position in the sequence of geographical position and obtains place.

Step 103, the geographical position corresponding based on each place and time determines the geographical attribute in each place.

Step 104, geographical attribute based on each place and corresponding time build geographical attribute sequence.

Step 105, carries out the first mapping process by geographical attribute corresponding for place each in geographical attribute sequence and time Obtain the credit feature of at least one dimension of user.

Step 106, the credit feature of at least one dimension based on user carries out the second mapping and processes the letter obtaining user With.

Exemplarily, data source can use polytype equipment, such as: mobile phone, panel computer, Wearable (intelligence Energy wrist-watch, intelligent glasses) and car-mounted terminal etc., the said equipment has locating module, and is provided with the authority to terminal positioning Such that it is able to collect geographical position and the time of relative users.Or, at terminal operating based on location-based service (such as social networking application The navigation feature etc. searched in nearby friends function, map application).

In certain embodiments, in a step 101, from a data source, the geographical position sequence of user is only obtained, and Parse the geographical attribute in each geographical position in the sequence of geographical position in step 103, to build geographical attribute at step 104 Sequence is as the intermediate object program of user's map user credit.Such as from the data acquisition geographical position, location of the smart mobile phone of user Sequence, and each geographical position resolved in geographical position sequence obtains corresponding geographical attribute.

In certain embodiments, in a step 101, for accurately determining the credit of user, from multiple data sources more comprehensively (such as smart mobile phone, panel computer and car-mounted terminal) obtains geographical position sequence one by one, then formed for waiting in a step 102 to solve The geographical position sequence of analysis, and have when forming geographical attribute sequence in step 102 to step 105 and be mapped as the credit of user Multiple processing mode, illustrative to different processing modes below.

Mode one) form multiple intermediate object program from the geographical position sequence one_to_one corresponding that multiple data sources obtain, fusion is many Individual intermediate object program map user credit.

For the multiple geographical position sequence obtained from multiple data source correspondences, directly to each geographical position sequence respectively Carry out resolving the geographical attribute obtaining each geographical position, form the geographical attribute sequence corresponding with each geographical position sequence, as The intermediate object program of map user credit, utilize the intermediate object program of corresponding each data source carry out mapping process (first mapping process and Second mapping processes) obtain user credit.

As shown in Figure 4, the geographical position sequence each data source collected is respectively processed and obtains phase one example The geographical attribute sequence answered: resolve from the geographical position sequence of data source 1 obtain intermediate object program 1 (namely geographical attribute sequence, Such as the static geographical form such as sequence of attributes, Dynamic Geographic sequence of attributes), as an intermediate object program of map user credit；With Reason, resolves the geographical position sequence pair from data source 2, data source 3 and should obtain intermediate object program 2 and intermediate object program 3, as mapping The intermediate object program of user credit.

Continue Fig. 4 to be illustrated, after the intermediate object program obtaining corresponding each data source, by intermediate object program (intermediate object program 1, intermediate object program 2 and intermediate object program 3), geographical attribute that namely in geographical attribute sequence, each place is corresponding and time utilization Mapping model carries out mapping process (carrying out the first mapping successively to process and the second mapping process), and the credit obtaining user credit is commented Divide result (it is of course also possible to using the forms such as the grading of credit).

In certain embodiments, when be combined with the intermediate object program that multiple data source obtains carry out mapping process time, it is considered to The reliability of data, degree of accuracy and sampling density (data source data acquisition frequency to geographical position) to different pieces of information source Difference, in order to ensure the credibility of the user credit determined, to the centre from different pieces of information source in credit mapping model Result distribution weight, is namely that (sign is used for reflecting the different geographical sequence of attributes distribution weights that pending first mapping processes Penetrate the significance level of user credit), merge the intermediate object program map user credit of corresponding multiple data source, credit mapping model can To use the modes such as linear weighted function, complex nonlinear weighting (such as neutral net) to merge the middle junction of corresponding multiple data source Really.The weight that geographical attribute sequence pair is answered is by (namely determining that geographical sequence of attributes is used geographical position based on derived data source Derived data source) reliability and at least one degree of accuracy determine.

Still as a example by Fig. 4, if the reliability of data source 1, data source 2 and data source 3 reduces successively, to for intermediate object program 1, intermediate object program 2 and intermediate object program 3 corresponding distribution weight 1, weight 2 and weight 3, wherein weight 1 > weight 2 > weight 3 so that in Between result 1, intermediate object program 2 and intermediate object program 3 influence degree of user credit is sequentially reduced, it is ensured that determine user credit can Reliability.

Mode two) it is overlapped from the geographical position sequence that multiple data sources obtain, utilize the geographical position sequence after superposition Row form intermediate object program, utilize intermediate object program map user credit.

Above-mentioned multiple data sources can be different types of multiple data source, and multiple data sources here can be from multiple Dimension divides.Such as, divide from the dimension of hardware device, different hardware devices is divided into different types of number According to source.The most such as, divide from the dimension of software, the corresponding different types of data source of different software such as App.The most such as, The dimension of data collection points (operation triggering geographic position data collection that user implements) different from same application divides For different types of data source.

For the multiple geographical position sequence obtained from multiple data source correspondences, by the identical time in the sequence of geographical position Geographical position be overlapped, geographical position cluster in the geographical position sequence after superposition for place, is obtained the ground in each place Reason attribute, geographical attribute based on each place (static geographical attribute and Dynamic Geographic attribute) and time structure geographical attribute Sequence, as the intermediate object program for map user credit, utilizes intermediate object program to carry out mapping process in credit mapping model Obtain user credit.

In certain embodiments, in order to avoid the geographical position from corrupt data source (or degree of accuracy the highest data source) Put the negative effect that sequence pair determines the accuracy of user credit, be overlapped multiple geographical position sequence being formed to be resolved During geographical position sequence (also referred to as geographical position synthetic data sequence), it is corresponding different in the synthetic data sequence of geographical position The geographical position distribution weight of data source, according to weight, the reliability of each data source data, degree of accuracy and sampling density are at least One of determine so that the appraisal result of the user credit that geographical position sequence pair based on data source output more reliably determines Impact bigger, it is ensured that the credibility of the user credit mapped out.

An example as superposition geographical position sequence, it is assumed that have 10 data sources can collect the geographical position of user Put, in the geographical position having 3 data sources (such as smart mobile phone, car-mounted terminal and intelligent watch) to collect user at 7 o'clock, that After the geographical position sequence of 3 data sources output is overlapped, 7 o'clock correspondence have 3 geographical position of user Data, the packing density in the geographical position of same time increases.

Another example, as it is shown in figure 5, the geographical position sequence each data source collected is based on sequential superposition, obtains Geographical position synthetic data sequence, geographical position synthetic data sequence each time includes that multiple data source collected in this time The geographical position of user, thus the packing density in the geographical position in each time is more than data mapping.Based on geographical position Put synthetic data sequence analysis and go out the geographical attribute in each place, such as static geographical attribute and Dynamic Geographic attribute, if static geographical Attribute and Dynamic Geographic attribute occupy the classification of multiple dimension, then formation multidimensional quiet Dynamic Geographic sequence of attributes and multidimensional are statically Reason sequence of attributes, as intermediate object program, carries out credit mapping in credit mapping model based on intermediate object program and obtains user credit, Exemplarily, credit mapping model utilizes linear and complex nonlinear classification and regression algorithm, obtains final credit and comment Point.

Mode three) by geographical position sequence construct multidimensional data matrix (HDLM) from multiple data sources, from many dimensions Location data block (LDB, Location Data Block) and multidimensional location data block (MDB, Multiple is extracted according to matrix Dimensional Location Data Block), to the geographical position sequence in location data block and multidimensional location data block Carry out dissection process and obtain geographical attribute sequence (such as shapes such as multidimensional static state geography sequence of attributes, Dynamic and Multi dimensional geographical attribute sequences Formula) as the intermediate object program for map user credit, utilize credit mapping model that intermediate object program carries out mapping process and obtain User credit.

One example is as it is shown in fig. 7, the geographical position sequence that multiple data sources exported, according to each geographical position (with reality Line square frame identifies) sequential (sequencing of the most each geographical position correspondence time) alignment, if the output of data source Geographical position sequence in corresponding geographical position sometime lack (this is because data source does not collect use in this time The geographical position at family), then utilize the value (such as zero, identify with dashed rectangle) of acquiescence or the calculated value of interpolation algorithm to fill For the geographical position of disappearance, each geographical position sequence after alignment and filling being processed builds multidimensional data square as row vector Battle array.

The corresponding geographical position sequence from a data source of every a line of multidimensional data matrix shown in Fig. 8 (is passed through The alignment of sequential, the filling in disappearance geographical position and interpolation processing), first the geographical position sequence of multiple data sources is carried out Matrix elasticity partition-merge processes, and obtains building the row vector (also referred to as primary features sequence) of multidimensional data matrix, then base The geographical attribute extracting each place (obtaining geographical position cluster) in multidimensional data matrix (includes static geographical attribute and moves State geographical attribute) to build geographical attribute sequence (also can be considered secondary features sequence), as in map user credit Between result.Finally geographical attribute sequence is classified and regression treatment through linear and complex nonlinear in credit mapping model (as used regression algorithm), obtains the credit scoring result of user.

Multiple geographical position sequence is carried out matrix elasticity partition-merge obtain the place of primary features sequence to above-mentioned again Reason illustrates.

Usually, in multiple data sources, only one of which data source can collect geographical position sometime, accordingly Ground, in multidimensional data matrix there is Effective Numerical in every string (vectorial) the most only certain one-dimensional geographical position, this column vector (can fill default value) that the data in the geographical position of other dimensions are missing from, exists a large amount of only one in multidimensional data matrix Individual dimension has the column vector of value, and such column vector exists the feature of continuous distribution, thus forms several big blocks also It is exactly to position data block.

It addition, in multidimensional data matrix in addition to there is big block, there is also block of cells i.e. multidimensional location data Block, block of cells is made up of the continuous print column vector of the Effective Numerical that all there is geographical position in multiple dimensions.Such as, if existing 10 data sources collect the position of user in continuous three moment simultaneously, 10 dimensions (each dimension correspondence one in this moment Individual data source) geographical position constitute a column vector in multidimensional data matrix, each column vector includes the ground of 10 dimensions Reason position, the column vector that in multidimensional data matrix, continuous three times are corresponding forms a multidimensional location data block.

In certain embodiments, as shown in Figure 8, the location data in multidimensional data matrix are identified by such a way Block and multidimensional location data block: scan by column identification multidimensional data matrix column vector, identify in poly-dimensional block data and only have The continuation column vector of effective geographic position data of one dimension is location data block；Identify in poly-dimensional block data and have at least The continuation column vector of effective geographic position data of two dimensions is that multidimensional positions data block.Whole multidimensional data matrix is divided For location data block and the sequence (primary features sequence) of multidimensional location data block alternative splicing of staggered splicing, wherein position number Inconsistent (the collection in each column vector corresponding data source of quantity according to the column vector included by block, multidimensional location data block Time), therefore the length of the time that location data block and multidimensional location data block are covered is inconsistent.

In certain embodiments, for the location data block identified from multidimensional data matrix, each is positioned data block In geographical position sequence based on sequential superposition, using the geographical position sequence that obtains after superposition as step 103 ground to be resolved Reason position sequence, goes out the geographical attribute in each place, forms quiet Dynamic Geographic attribute based on geographical position synthetic data sequence analysis Sequence and static state geography sequence of attributes, as intermediate object program, carry out credit in credit mapping model based on intermediate object program and map To user credit.

It addition, in order to ensure the credibility of user credit determined, weight can be distributed for different location data blocks, Weight determines with at least one reliability and degree of accuracy of positioning data block corresponding data source, exemplarily, and location data block The reliability in weight and corresponding data source and degree of accuracy positive correlation, the reliability of data source is the highest with degree of accuracy, positions the most accordingly The weight of data block is the highest, so that user's letter that geographical position sequence pair based on data source output more reliably determines The impact of appraisal result bigger, it is ensured that the credibility of the user credit mapped out.

In certain embodiments, for the multidimensional location data block identified from multidimensional data matrix, owing to multidimensional positions Data block is that not only this time is corresponding owing to multiple data sources are formed in the geographical position that the same time collects user simultaneously The packing density in geographical position is higher than the packing density in geographical position in the data block of location, but also illustrates that user is now in certain A little special scenes or implement some specific behavior.

Such as, the vehicle mounted guidance App of user collects geographical position, and the mobile phone of user is commented on App masses and also adopted simultaneously Collection, to geographical position, illustrates that user is the most driving and searching where to have a meal.If now certain good friend of this user Mobile phone also collect geographic position data, and the geographical position of good friend is the most synchronize with the position of this user, then be likely to It is that user drives to carry good friend and goes together to have a meal somewhere.

Correspondingly, when determining in step 103 based on multidimensional location data block structure geographical attribute sequence, multidimensional is positioned Geographical position sequence mapping in data (includes multiple dimension to the semantic category system of the scene characterized residing for user or behavior Classification) in, obtain each geographical position Dynamic Geographic attribute in multiple dimensions.

The geographical attribute sequence of correspondence is formed also based on geographical position sequence again in aforesaid three kinds of mode either types The process of map user credit illustrates, and the geographical attribute in geographical position includes static geographical attribute and Dynamic Geographic attribute, Correspondingly, geographical attribute sequence includes static geographical sequence of attributes and Dynamic Geographic sequence of attributes, separately below to combining static state Geographical attribute map user credit and combine Dynamic Geographic attribute map user credit explanation.

1) static geographical attribute

In one embodiment, carry out the geographical position in the sequence of geographical position clustering (as based on distribution density poly- Class or cluster based on Euclidean distance) it is place, place can be a geographical position, it is also possible to be by multiple geographical position structure The region (region such as formed by multiple geographical position) become.

The labelling using the place after cluster replaces the labelling in corresponding geographical position in geographical position sequence, when obtaining place Between sequence (sequence that the binary combination of place and time is constituted), as geographical position sequence: the geographical position 1-time 1, Reason position 2-time 2, geographical position 3-time 3, the geographical position 4-time 4, if geographical position 1 and geographical position 2 cluster are ground Point 1, geographical position 3 and geographical position 4 cluster are place 2, then replace geographical position 1 and geographical position 2 with place 1, utilize ground Point 2 replacement geographical position 3 and geographical position 4, the ground point sequence of formation is: when place 1-time 1, place 1-time 2, place 2- Between 3, the place 2-time 4.

After forming place time series, to time of each place in the time series of place and correspondence from least one Individual dimension carries out semantic category classification, obtains the static geographical attribute of at least one dimension corresponding of each place in the time series of place (during multiple dimension, then forming multidimensional static state geographical attribute), the time shape that static geographical attribute based on each place, place are corresponding Become static geographical sequence of attributes, such as static geographical attribute 1-time 1, static geographical attribute 2-time 2 such form.

Form an example of static geographical sequence of attributes as it is shown in figure 9, for the geographical position of random length (being designated as n) Sequence, by density-based algorithms (such as DBSCAN etc.) or clustering algorithm (such as K-means based on Euclidean distance Deng), (being designated as m, usual m is much smaller than the list in n) place (or region) to obtain several.Based on position geographical to n dimension, m dimension place Put the labelling in geographical position in sequence to be replaced, obtain n dimension place time series.

The mode using semantic category classification carries out semantics recognition to the place in the time series of n dimension place, it is thus achieved that static Geographical attribute sequence: m ties up " place-classification " list.Every a line in m dimension " place-classification " list has 2 data item, and first Individual data item is place ID, and second data item is place classification obtained by semantics recognition, such as residence, work The classifications such as ground, amusement and leisure ground.

When it is pointed out that the various places point in the time series of place carries out semantics recognition classifies, it is possible to use one (semantics recognition grader is for place each in the time series of place for semantic category grader or multiple semantics recognition grader Carry out the mathematical model of semantics recognition classification).Utilize multiple semantic category grader (namely from multiple dimensions) to each ground When point carries out semantics recognition classification, correspondingly, " place-classification " list of m dimension has the most just been extended to many column vectors namely matrix Form, each column vector each place corresponding classification being had under certain taxonomic hierarchies, example such as table 1 institute Show:

Table 1

As it can be seen from table 1 same place has the label of different classifications, and some under different taxonomic hierarchieses The label of the classification under dimension can lack (such as, because data source does not collect corresponding geographical position or because class Type excessively obscures and cannot definitely classify).

2) credit mapping processes

Continue to determining that the process at the static geographical attribute of at least one dimension of each place illustrates.Geographical attribute sequence Geographical attribute and time that in row, each place is corresponding carry out the first mapping process, exemplarily, based on each place statically Reason attribute and user are in the time in all kinds of place, by the Nonlinear Classifier built in advance and regression model, determine at least Reflection user's loan repayment capacity one dimension, direct or indirect and the credit feature of refund wish, exemplarily, including: user Income level, level of consumption, job specification (such as day shift, night shift or in shifts), occupation type (such as high-tech enterprise, School, institutional settings etc.), quality of life and life health degree (overtime work for example whether stay up late the most for a long time, live or work Pressure is big).

Based on the feature that obtains of mapping, then by compressive classification and regression model, credit feature carried out the second mapping process Obtain final credit scoring result.

The process of above-mentioned map user credit based on static geographical attribute is if provided as credit mapping block, except can To be used alone and outside export credit appraisal result, it is also possible to by special for multiple dimensions of reflection user's loan repayment capacity and refund wish Levy and export as intermediate object program, for the follow-up Dynamic Geographic attribute intermediate object program determining each place, for combining the quiet of each place State geographical attribute and the credit scoring of Dynamic Geographic attribute map user, obtain credit more comprehensive, accurate, reliable and comment Estimate.

2) Dynamic Geographic attribute

Dynamic Geographic attribute uses the space-time migration model of user and the type of motion track pattern to describe, to two kinds of moulds The extraction of formula and determine that the classification of pattern illustrates.

Input data shown in Fig. 9 and intermediate object program are used for being formed Dynamic Geographic attribute, and based on credit mapping model Carrying out mapping and obtain user credit, input data and intermediate object program include that aforesaid n dimension geographical position sequence, n tie up the place time Sequence and m tie up " place-classification " list (or matrix).

2.1) motion track pattern

See Figure 10, determine that motion track pattern depends on n dimension geographical position sequence, in one embodiment, Ke Yitong Cross such mode and determine motion track pattern, by the geographical position in the sequence of geographical position based at least one partition of the scale (such as in time scale and space scale) is multiple segmentation, a movement of the geographical position correspondence user in each segmentation Track, user's motion track pattern on different scale that the motion track on different scale is corresponding.Such as in residential quarters Yardstick on, user may take a walk in community, or travels to and fro between the supermarket shopping in community, or carries out various entertainment in community Movable etc., according to different application demands, pre-establish the motion track mode type table on a series of different scale, pass through structure Build grader and stamp the classification of at least one dimension in motion track mode type table to the motion track pattern on different scale Label.Such as table 2 below example:

Table 2

2.2) space-time migration model

N is tieed up locations and regions time series and m dimension " place-classification " list combines, extract user different The different subsequence (such as B-place, place A-place C) migrated between place, extracts frequently the subsequence extracted The subsequence of pattern, namely the frequency of occurrences meets that pre-conditioned (if the frequency of occurrences is higher than frequency threshold, or the frequency of occurrences is High predetermined quantity) subsequence, as user at the space-time migration model of place rank.

It addition, the space-time migration model (such as residence-hospital-residence, or CBD between different types of place Region-high-tech park--CBD region, the suburbs etc.) in, there is also some and be under multiple category division systems frequently simultaneously The subsequence of pattern, lasts the migration between these space-time migration models and the time of staying in each place, place, migrates Initial complete the moment etc. with migration and combine, just constitute " multidimensional space-time migration series ".By structural classification device, by these Multidimensional space-time migration series is mapped in " the space-time migration semantic category " pre-established, and is formed user's living habit or life The quantitative expression of pattern, such as, " often work then go to bar street to entertain 2:00 AM at 9 in evening then go home sleep ", Or " working in morning often late and on the way Dou Shi smooth traffic district " etc., above-mentioned quantitative expression may be used for map user Credit.

3) credit mapping processes

Based on motion track pattern derived above and space-time migration model, by the Nonlinear Classifier built in advance and return Return model, carry out the first mapping process based on motion track pattern and space-time migration model, can obtain multiple dimension, directly Or indirectly reflect user's loan repayment capacity and the associated credit feature of refund wish.

Exemplarily, including: the living habit of user, work habit, sports health custom, (such as every day adheres to outdoor fortune Dynamic, indoor timing body-building etc. weekly), quality of life and health status (the most whether having serious disease etc.), philosophy of life (such as Strict punctual, work drive foot, life are careless and sloppy freely etc.), (such as " working day lives in CBD district to income level with quality of the life Territory and return the villa in the suburbs weekend and live and leisure ", or " during work, often both at home and abroad aircraft is gone on business and is often gone to the beach weekend And sea " etc.), level of consumption etc..

Process, based on carrying out the first mapping, the credit feature of user obtained, then by compressive classification and regression model to The credit feature at family carries out the second mapping and processes the final credit scoring result obtaining user.

Above-mentioned credit mapping model based on Dynamic Geographic attribute is if provided as credit mapping block, except can be independent Use and export credit appraisal result outside, it is also possible to using reflection user's loan repayment capacity and refund wish multiple dimensional characteristics as Intermediate object program exports, for the follow-up Dynamic Geographic attribute intermediate object program determining each place, for combining the static geographical of each place Attribute and the credit scoring of Dynamic Geographic attribute map user, obtain credit evaluation more comprehensive, accurate, reliable.

Again the device of the determination credit that the embodiment of the present invention provides is illustrated, see the determination credit shown in Figure 11 One optional structural representation of device, including: data acquisition module 10, geographical attribute constructing module 20, credit map mould Block 30, direct sequential density laminating module 40 and matrix elasticity partition-merge module 50, be combined with position data and merge Process and determine that each module is illustrated by the processing procedure of user credit.

Amalgamation mode one

In conjunction with Fig. 4, data acquisition module 10, for obtaining n geographical position from data source (data source 1 to data source n) Sequence, geographical position sequence includes that user is in the geographical position residing for different time.

Geographical attribute constructing module 20, obtains place for being clustered in the geographical position in n geographical position sequence；Based on Geographical position that each place is corresponding and time determine the geographical attribute in each place；Geographical attribute based on each place and correspondence Time builds geographical attribute sequence, for each geographical position sequence formed n intermediate object program (include geographical attribute sequence 1 to Geographical attribute sequence n).

Credit mapping block 30, is additionally operable to based on n intermediate object program, by geography corresponding for place each in geographical attribute sequence Attribute and time carry out the first mapping and process the credit feature of at least one dimension obtaining user；Based on user at least one The credit feature of individual dimension carries out the second mapping and processes the credit obtaining user.

Amalgamation mode two

Data acquisition module 10 is for obtaining geographical position sequence, namely geographical position from data source 1 to data source n correspondence Sequence 1 to geographical position sequence n.

Directly sequential density laminating module 40, time identical in general's at least geographical position sequence 1 to geographical position sequence n Between corresponding geographical position be overlapped, obtain geographical position sequence to be resolved namely geographical position synthetic data sequence.

Alternatively, direct sequential density laminating module 40, it is additionally operable in geographical position sequence to be resolved for correspondence not Distributing weight with the geographical position of data source, the weight distributed is that at least one reliability based on data source and degree of accuracy are true Fixed.

Geographical attribute constructing module 20, obtains ground for being clustered in the geographical position in the synthetic data sequence of geographical position Point；The geographical position corresponding based on each place and time determine the geographical attribute in each place；Geographical attribute based on each place And the correspondence time builds geographical attribute sequence, form (one) intermediate object program.

Credit mapping block 30, is additionally operable to based on intermediate object program, by geographical genus corresponding for place each in geographical attribute sequence Property and the time carry out the first mapping and process the credit feature of at least one dimension obtaining user；Based on user at least one The credit feature of dimension carries out the second mapping and processes the credit obtaining user.

Amalgamation mode three

1) multidimensional data matrix-split

Matrix elasticity partition-merge module 50, obtains geographical position sequence for data source 1 to data source n correspondence, i.e. Reason position sequence 1 to geographical position sequence n.Multidimensional data matrix is built based on sequence 1 to geographical position, geographical position sequence n.

Matrix elasticity partition-merge module 50 builds multidimensional data matrix by such mode: geographical based at least two The sequencing of geographical position correspondence time in position sequence, it is right to be carried out in the geographical position in the sequence of at least two geographical position Neat and filling processes, and the geographical position sequence after processing builds multidimensional data matrix as row vector.

Matrix elasticity partition-merge module 50, for identifying each column vector in multidimensional data matrix, based on each column vector Including the dimension in effective geographical position, location data block and the multidimensional that multidimensional data matrix-split is alternative splicing is positioned data Block, adopts and identifies location data block and multidimensional location data block in such a way: identify and only have one in poly-dimensional block data The continuation column vector in effective geographical position of dimension is location data block；Identify and poly-dimensional block data has at least two dimension The continuation column vector in effective geographical position be that multidimensional positions data block.

2) static geographical attribute is extracted

Geographical attribute constructing module 20 uses the place after cluster to replace corresponding geographical position in geographical position sequence, obtains Place time series；The time that each place in the time series of place and each place are corresponding is carried out from least one dimension Semantic category is classified, and the classification obtaining at least one dimension corresponding of each place in the time series of place is static geographical attribute.

3) Dynamic Geographic attribute is extracted

Geographical attribute constructing module 20 is additionally operable to extract user based on place time series and static state geography sequence of attributes Migration subsequence between different location, static geographical sequence of attributes is static geographical attribute based on each place and correspondence Time build obtain；It is dynamic for extract the frequency of occurrences meeting the pre-conditioned space-time migration model that subsequence is user that migrates Geographical attribute.

Alternatively, geographical attribute constructing module 20 is additionally operable to the geographical position in the sequence of geographical position based at least one Partition of the scale is at least two segmentation；The motion track pattern that geographical position in each segmentation is corresponding is entered from least one dimension Row classification, the classification obtaining motion track pattern at least one dimension corresponding is dynamic geographical attribute.

4) credit maps

Credit mapping block 30, is additionally operable to static geographical attribute corresponding for place each in geographical attribute sequence with dynamic At least one geographical attribute carry out the first mapping process, obtain reflecting the credit feature of at least one dimension of user；Will reflection The credit feature of at least one dimension of user carries out the second mapping and processes the credit obtaining user.

Alternatively, credit mapping block 30, it is additionally operable at least two geographical attribute sequence processed into pending first mapping Row distribution weight, the weight that geographical attribute sequence pair is answered is that at least one reliability based on derived data source and degree of accuracy are true Fixed, derived data source by output for determining the data source in geographical the used geographical position of sequence of attributes.

In conjunction with Fig. 1 to determining that the device of information implementation in actual applications illustrates.

Implementation 1) mobile terminal

The device of the determination credit that the embodiment of the present invention provides may be embodied as the mobile end with hardware configuration shown in Fig. 1 End, mobile terminal the method implementing above-mentioned determination credit by operation application program or software function module.

Such as, it is possible to provide for using the software function module of the programming language exploitations such as C/C++, Java (to include a series of It is available for the instruction that processor performs), it is embedded in various mobile terminal Apps based on systems such as Android or iOS (the most micro- Letter etc.), thus directly use the calculating resource (processor) of mobile terminal self to obtain the geographical attribute of user, and based on geography The credit scoring of property calculation user, and periodically or non-periodically by various network communication modes by data, intermediate object program Or final result sends long-range server to, or preserve in mobile terminal this locality.

Implementation 2) server end

The embodiment of the present invention can provide is write as single application software or large-scale soft based on programming languages such as C/C++, Java Software function module (including a series of instruction being available for processor execution) in part system, runs on server end, will connect The original geographical position of the mobile terminal from single or numerous users received, treated intermediate object program or final credit The result of scoring, integrates with the result of the historical data on server, intermediate object program or credit scoring and is calculated renewal The result of credit scoring, other application programs run in export server end the most real-time or non real-time or software function Module uses, it is also possible to write server-side database or file store.

Implementation 3) distributed credit evaluation platform

The embodiment of the present invention also provides for the distributed parallel computing platform constituted into multiple servers, carries mutual net Network (Web) interface or other kinds user interface, form the geographical location information for individual, colony or enterprise and excavate and letter Use Evaluation Platform.Existing packet batch can be uploaded to platform to obtain various result of calculations (in the middle of such as by user Result and the final result of credit scoring), it is also possible to calculate and update knot to this platform in real time real-time data stream transmitting Really (final result of such as intermediate object program and credit scoring).

Implementation 4) server-side application interface (API, Application Interface) and plug-in unit

The embodiment of the present invention can be provided as the API of server end, software development kit (SDK, Soft Development Toolkit) or plug-in unit, the server-side application developer for other calls, and is embedded in types of applications program.

Implementation 5) mobile device client end AP I and plug-in unit

The embodiment of the present invention can be provided as API, SDK or the plug-in unit of mobile device end, for other mobile terminal application program Developer calls, and is embedded in types of applications program.

Implementation 6) high in the clouds open service

Excavating and credit evaluation platform at the available geographical location information of the embodiment of the present invention, the embodiment of the present invention also may be used Being provided as geographical location information to excavate and API, SDK of credit evaluation platform and plug-in unit etc., packing is packaged into and is available for inside and outside enterprise The open cloud service used of personnel.Or by various results (intermediate object program and the final result of credit scoring) in a suitable form It is illustrated on various terminal presentation facility, for individual, colony or enterprises and institutions' inquiry.

In sum, the embodiment of the present invention has the advantages that

1) the user's geographic position data (unequal interval gathered in mobile (social) Apps such as such as wechat can be utilized Time series) obtain the credit scoring of user, the coverage rate of data acquisition is high, it is possible to most user is carried out credit and comments Point.

2) shopping or the balance data that employ non-immediate (non-explicit) determine the credit scoring of user as collage-credit data, Owing to user can be covered the most of the time gathering geographic position data of the life cycle of user, it is possible to big portion based on user The credit of user is updated comprehensively evaluating by the geographical attribute dividing the place residing for life cycle, relatively merely with directly purchasing The scheme of the credit scoring of thing or balance data (data directly related with the finance of user, transaction) assessment user to The evaluation of family credit is more accurate.Meanwhile, directly can do shopping or the balance data scheme to carrying out credit scoring with utilizing It is used in combination such that it is able to support to use rich and varied data to solve reference problem.

3) obtainable data volume is bigger, and the frequency is higher, covers the closeest wider, to different crowd to the life time of user Covering wider, contribute to realizing general favour finance.

4) use the scheme from multi-data source acquisition geographic position data, support plurality of devices, device and the nothing of data source Stitch linking and merge, on the one hand achieve the covering of the multi-angle to user's life time thus improve the essence of credit scoring further Parasexuality and reliability, on the other hand widened the border that crowd covers the most further, contributes to realizing general favour finance.

It will be appreciated by those skilled in the art that: all or part of step realizing said method embodiment can pass through journey The hardware that sequence instruction is relevant completes, and aforesaid program can be stored in a computer read/write memory medium, and this program exists During execution, perform to include the step of said method embodiment；And aforesaid storage medium includes: flash memory device, deposit at random Access to memory (RAM, Random Access Memory), read only memory (ROM, Read-Only Memory), magnetic disc or The various medium that can store program code such as CD.

Or, if the above-mentioned integrated unit of the present invention is using the form realization of software function module and as independent product When selling or use, it is also possible to be stored in a computer read/write memory medium.Based on such understanding, the present invention implements The part that correlation technique is contributed by the technical scheme of example the most in other words can embody with the form of software product, This computer software product is stored in a storage medium, including some instructions with so that a computer installation is (permissible It is personal computer, server or network equipment etc.) perform all or part of of method described in each embodiment of the present invention. And aforesaid storage medium includes: flash memory device, RAM, ROM, magnetic disc or CD etc. are various can store program code Medium.

The above, the only detailed description of the invention of the present invention, but protection scope of the present invention is not limited thereto, and any Those familiar with the art, in the technical scope that the invention discloses, can readily occur in change or replace, should contain Cover within protection scope of the present invention.Therefore, protection scope of the present invention should be as the criterion with described scope of the claims.

Claims (22)

1. the method determining credit, it is characterised in that described method includes:

Obtaining geographical position sequence based on data source, described geographical position sequence includes that user is in the geographical position residing for different time Put；

Geographical position in the sequence of described geographical position is clustered and obtains place；

The described geographical position corresponding based on each described place and time determine the geographical attribute in each described place；

Geographical attribute based on each described place and corresponding time build geographical attribute sequence；

Geographical attribute corresponding for each place in described geographical attribute sequence and time are carried out the first mapping process and obtains described The credit feature of at least one dimension of user；

The credit feature of at least one dimension based on described user carries out the second mapping and processes the credit obtaining described user.

Method the most according to claim 1, it is characterised in that described obtain geographical position sequence based on data source, including:

At least two geographical position sequence is obtained from data source correspondence described at least two；

The geographical position corresponding identical time in the sequence of at least two geographical position is overlapped, obtain to be resolved describedly Reason position sequence.

Method the most according to claim 2, it is characterised in that described method also includes:

Described geographical position sequence to be resolved is the geographical position distribution weight in corresponding different pieces of information source, the power distributed Heavily determine at least one reliability based on described data source, degree of accuracy and sampling density.

Method the most according to claim 1, it is characterised in that described obtain geographical position sequence based on data source, including:

Based at least two geographical position sequence construct multidimensional data matrix obtained from least two data source correspondence；

Identify each column vector in described multidimensional data matrix, include the dimension in effective geographical position based on each described column vector, Location data block and multidimensional that described multidimensional data matrix-split is alternative splicing are positioned data block.

Method the most according to claim 4, it is characterised in that

Described based at least two geographical position sequence construct multidimensional data matrix obtained from least two data source correspondence, bag Include:

Based on the sequencing of geographical position correspondence time in the sequence of described at least two geographical position, by least two geography position Putting the geographical position in sequence to carry out aliging and filling process, the geographical position sequence after processing builds described as row vector Multidimensional data matrix.

Method the most according to claim 4, it is characterised in that

The described dimension including effective geographical position based on each described column vector, by described multidimensional data matrix-split for alternately spelling The location data block connect and multidimensional location data block, including:

The continuation column vector identifying effective geographical position only in described poly-dimensional block data with a dimension is described location Data block；

The continuation column vector identifying effective geographical position in described poly-dimensional block data with at least two dimension is described many Dimension location data block.

Method the most according to claim 1, it is characterised in that the described described geographical position corresponding based on each described place And the time determines the geographical attribute in each described place, including:

Use the described place after cluster to replace corresponding geographical position in the sequence of described geographical position, obtain place time series；

The time that each described place in the time series of described place and each described place are corresponding is entered from least one dimension Lang justice category classification, the classification obtaining at least one dimension corresponding of each described place in the time series of described place is static Described geographical attribute.

Method the most according to claim 7, it is characterised in that described method also includes:

Described user is extracted between different described places based on described place time series and static state geography sequence of attributes Migrating subsequence, described static geographical sequence of attributes is static geographical attribute based on each described place and the time structure of correspondence Build and obtain；

It is dynamic for extract the frequency of occurrences meeting the pre-conditioned space-time migration model that described migration subsequence is described user Described geographical attribute.

Method the most according to claim 1, it is characterised in that the described described geographical position corresponding based on each described place And the time determines the geographical attribute in each described place, including:

It is at least two segmentation by the geographical position in the sequence of described geographical position based at least one partition of the scale；

The motion track pattern that geographical position in each described segmentation is corresponding is classified from least one dimension, obtains described The classification of motion track pattern at least one dimension corresponding is dynamic described geographical attribute.

Method the most according to claim 1, it is characterised in that described that each place in described geographical attribute sequence is corresponding Geographical attribute and the time carry out the first mapping and process the credit feature of at least one dimension obtaining described user, including:

At least one static geographical attribute corresponding for each place in described geographical attribute sequence and dynamic geographical attribute are entered Row the first mapping processes, and obtains reflecting the credit feature of described at least one dimension of user.

11. methods according to claim 10, it is characterised in that described method also includes:

For geographical attribute sequence distribution weight, described geographical attribute sequence described at least two that pending described first mapping processes The weight that row are corresponding be reliability based on derived data source, degree of accuracy and and at least one sampling density determine, described source Data source by output for determining the data source in described the used geographical position of geographical attribute sequence.

12. 1 kinds of devices determining credit, it is characterised in that described device includes:

Data acquisition module, for obtaining geographical position sequence based on data source, described geographical position sequence includes that user is not With the geographical position residing for the time；

Geographical attribute constructing module, obtains place for being clustered in the geographical position in the sequence of described geographical position；

Described geographical attribute constructing module, is additionally operable to the described geographical position corresponding based on each described place and the time determines respectively The geographical attribute in described place；Geographical attribute based on each described place and corresponding time build geographical attribute sequence；

Credit mapping block, for carrying out first by geographical attribute corresponding for each place in described geographical attribute sequence and time Mapping processes the credit feature of at least one dimension obtaining described user；

Described credit mapping block, the credit feature being additionally operable at least one dimension based on described user is carried out at the second mapping Reason obtains the credit of described user.

13. devices according to claim 12, it is characterised in that

Described data acquisition module, is additionally operable to obtain at least two geographical position sequence from data source correspondence described at least two；

Described device also includes: directly sequential density laminating module, for by the identical time in the sequence of at least two geographical position Corresponding geographical position is overlapped, and obtains described geographical position sequence to be resolved.

14. devices according to claim 13, it is characterised in that

Described direct sequential density laminating module, is additionally operable in described geographical position sequence to be resolved as corresponding different pieces of information Source geographical position distribution weight, the weight distributed be reliability based on described data source, degree of accuracy and employing density extremely One of few determine.

15. devices according to claim 12, it is characterised in that

Described device also includes: matrix elasticity partition-merge module, for based on obtain from least two data source correspondence to Few two geographical position sequence construct multidimensional data matrixes；

Identify each column vector in described multidimensional data matrix, include the dimension in effective geographical position based on each described column vector, Location data block and multidimensional that described multidimensional data matrix-split is alternative splicing are positioned data block.

16. devices according to claim 15, it is characterised in that

Described matrix elasticity partition-merge module, is additionally operable to based on geographical position in the sequence of described at least two geographical position corresponding The sequencing of time, carries out the geographical position in the sequence of at least two geographical position aliging and filling process, after processing Geographical position sequence as row vector build described multidimensional data matrix.

17. devices according to claim 15, it is characterised in that

Described matrix elasticity partition-merge module, is additionally operable to identify in described poly-dimensional block data and only has the effective of a dimension The continuation column vector in geographical position is described location data block；

The continuation column vector identifying effective geographical position in described poly-dimensional block data with at least two dimension is described many Dimension location data block.

18. devices according to claim 12, it is characterised in that

Described geographical attribute constructing module is also used for the described place after cluster and replaces in the sequence of described geographical position corresponding Geographical position, obtains place time series；

The time that each described place in the time series of described place and each described place are corresponding is entered from least one dimension Lang justice category classification, the classification obtaining at least one dimension corresponding of each described place in the time series of described place is static Described geographical attribute.

19. devices according to claim 18, it is characterised in that

Described geographical attribute constructing module is additionally operable to extract institute based on described place time series and static state geography sequence of attributes Stating user migration subsequence between different described places, described static geographical sequence of attributes is based on each described place quiet The time of state geographical attribute and correspondence builds and obtains；

It is dynamic for extract the frequency of occurrences meeting the pre-conditioned space-time migration model that described migration subsequence is described user Described geographical attribute.

20. devices according to claim 12, it is characterised in that

Described geographical attribute constructing module is additionally operable to the geographical position in the sequence of described geographical position based at least one yardstick It is divided at least two segmentation；

The motion track pattern that geographical position in each described segmentation is corresponding is classified from least one dimension, obtains described The classification of motion track pattern at least one dimension corresponding is dynamic described geographical attribute.

21. devices according to claim 12, it is characterised in that

Described credit mapping block, is additionally operable to static geographical attribute corresponding for each place in described geographical attribute sequence and moves At least one geographical attribute of state carries out the first mapping process, obtains reflecting the credit feature of described at least one dimension of user.

22. devices according to claim 21, it is characterised in that

Described credit mapping block, is additionally operable to as geographical attribute sequence described at least two of pending described first mapping process Distribution weight, the weight that described geographical attribute sequence pair is answered is reliability based on derived data source, degree of accuracy and sampling density At least one determine, described derived data source by output for determining the data in described the used geographical position of geographical attribute sequence Source.