CN105894028A - User identification method and device - Google Patents
User identification method and device Download PDFInfo
- Publication number
- CN105894028A CN105894028A CN201610197077.1A CN201610197077A CN105894028A CN 105894028 A CN105894028 A CN 105894028A CN 201610197077 A CN201610197077 A CN 201610197077A CN 105894028 A CN105894028 A CN 105894028A
- Authority
- CN
- China
- Prior art keywords
- user
- feature
- weight
- property value
- value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a user identification method and device. The method in one embodiment comprises steps of: acquiring the attribute value of at least one characteristic of a user to be identified from a user information set collected in advance; acquiring, from a model established in advance, weights matching each of the at least one characteristic of the user to be identified and the attribute value of each of the at least one characteristic of the user to be identified, wherein the model includes characteristics, candidate attribute values associated with the characteristics, and weights associated with the characteristics and the candidate attribute values, the weights are determined according to the proportion of users in a pre-stored target user information set and a basic user information set and with the attribute values of characteristics equal to the candidate attribute values; acquiring the sum of the weights matching the characteristic of the attribute values of characteristics of the user to be identified; and determining whether the user to be identified is a potential target user according to the sum of the weights. The method may accurately identify more potential target users.
Description
Technical field
The application relates to field of computer technology, is specifically related to user's Portrait brand technology field, especially
Relate to user identification method and device.
Background technology
Flourish along with internet, precisely analyzes each user's by user's representation data
The demand of attribute and relation is more and clearer and more definite.User's portrait is the virtual representations of real user,
It is built upon the targeted customer's model on a series of True Data.Go to understand by user's investigation
They, according to the difference of their target, behavior and viewpoint, are divided into different classes by user
Type, then extracts characteristic feature from each type, give some demography key elements,
Scenes etc. describe, and are the formation of user's representation data.User's portrait makes enterprise to lead to
Cross internet and advantageously obtain user's feedback information the most widely, for the most precisely, quickly
The important business informations such as user behavior custom, consumption habit are analyzed on ground, it is provided that enough data
Basis.
At present, user's representation data has more successful at aspects such as information recommendation, message propelling movements
Application experience.Before carrying out information recommendation, message propelling movement, need at basic user information collection
Conjunction identifies potential target user so that information recommendation, message propelling movement etc. can more have for
Property ground carry out.Prior art identifying, the method for potential target user is normally based on user and uses
The frequency of predetermined prod is identified.
But, the usual scale of potential target user identified by above-mentioned prior art is less,
Have some limitations, it is impossible to identify more potential target user exactly.
Summary of the invention
The purpose of the application is to propose a kind of user identification method and device, solves above back of the body
The technical problem that scape technology segment is mentioned.
First aspect, this application provides a kind of user identification method, it is characterised in that described
Method includes: obtain at least one of user to be identified from the user profile set collected in advance
The property value of feature;Obtain from the model pre-build and described in user to be identified at least one
Each feature in individual feature and the weight of the property value coupling of feature, wherein, described model bag
Include following information: feature and the candidate value of described feature association and described feature and institute
Stating the weight of candidate value association, described weight is by the targeted customer's information collection prestored
Close and the property value of feature described in basic user information aggregate is equal to the use of described candidate value
The accounting at family carries out contrasting and determines;Obtain at least one feature described of user to be identified
The weight sum of the property value coupling of each feature and feature;With the size of described weight sum it is
Foundation, identifies whether user to be identified is potential target user.
In certain embodiments, described by the targeted customer's information aggregate prestored and basis are used
The property value of feature described in the information aggregate of family enters equal to the accounting of the user of described candidate value
Row contrast, including: obtain described targeted customer's information aggregate and described basic user information respectively
The property value of feature described in set equal to the user of described candidate value in set shared
Ratio;Obtain the described ratio of described targeted customer's information aggregate and described basic user information aggregate
The absolute value of the difference of example, as the weight associated with described feature and described property value.
In certain embodiments, described by the targeted customer's information aggregate prestored and basis are used
The property value of feature described in the information aggregate of family enters equal to the accounting of the user of described candidate value
Row contrast, also includes: according to the quantity with the candidate value of described feature association, revises institute
State weight, wherein, between weight and the quantity of described candidate value of correction, there is positive correlation
Relation.
In certain embodiments, described model also includes precondition information, in described model
The weight that described and described feature and described property value associate is and described precondition, described spy
The weight of described property value of seeking peace association, wherein, described weight is by meeting institute to prestore
State the genus of feature described in targeted customer's information aggregate of precondition and basic user information aggregate
Property value carry out contrasting equal to the accounting of user of described candidate value and determine;Treat described in and
Identify that user is the user to be identified meeting predetermined precondition;And it is described from pre-building
Model obtains and each feature at least one feature described of user to be identified and feature
The weight of property value coupling, including: obtain predetermined preposition with described from the model pre-build
Condition, user to be identified at least one feature described in each feature and the property value of feature
The weight of coupling.
In certain embodiments, described method also includes: after determining potential target user,
Obtain the identification successful user set becoming targeted customer in potential target user;By to described
Identify associated by each feature in successful user set and described basic user information aggregate is every
The accounting of the user of individual candidate value contrasts, and recalculates and updates in described model
Described weight.
In certain embodiments, described obtain from the user profile set collected in advance to be identified
The property value of at least one feature of user, including: the user profile set to collecting in advance is entered
At least one in the following process of row: former property value discrete in user profile set is updated to
For representing the property value of each predetermined interval scope of former property value;By in user profile set
The property value of user is the default value that the property value of empty feature is set to preset;For each institute
State feature, delete the corresponding weight property value less than predetermined threshold;From through described process it
After user profile set in obtain the property value of at least one feature of user to be identified.
Second aspect, this application provides a kind of customer identification device, and described device includes: special
Levy information acquisition unit, for obtaining user to be identified from the user profile set collected in advance
The property value of at least one feature;Weight Acquisition unit, for from the model pre-build
Obtain and each feature at least one feature described of user to be identified and the property value of feature
The weight of coupling, wherein, described model includes following information: feature and described feature association
The weight that associates with described feature and described candidate value of candidate value, described weight
It is by feature described in the targeted customer's information aggregate prestored and basic user information aggregate
Property value carries out contrasting equal to the accounting of user of described candidate value and determines;Add and unit,
For obtaining each feature at least one feature described in user to be identified and the attribute of feature
The weight sum of value coupling;Recognition unit, for the size of described weight sum as foundation,
Identify whether user to be identified is potential target user.
In certain embodiments, described device also includes: weight determining unit, for for institute
State each feature at least one feature and each property value with described feature association, pass through
The property value of feature in the targeted customer's information aggregate prestored and basic user information aggregate is equal to
The accounting of the user of candidate value carries out contrast and determines and described feature and described candidate value
The weight of association, described weight determining unit includes: ratio obtains subelement, for obtaining respectively
Take the attribute of feature described in described targeted customer's information aggregate and described basic user information aggregate
The ratio that value is shared in set equal to the user of described candidate value;Weight determines subelement,
For obtaining described targeted customer's information aggregate and the described ratio of described basic user information aggregate
The absolute value of difference, as the weight associated with described feature and described property value.
In certain embodiments, described weight determining unit also includes: weight correction subelement,
For according to the quantity with the candidate value of described feature association, revise described weight, wherein,
Positive correlation is there is between the weight and the quantity of described candidate value that revise.
In certain embodiments, described model also includes precondition information, in described model
The weight that described and described feature and described property value associate is and described precondition, described spy
The weight of described property value of seeking peace association, wherein, described weight is that described weight determining unit leads to
Cross the targeted customer's information aggregate meeting described precondition prestored and basic user information collection
The property value of feature described in conjunction carries out contrast really equal to the accounting of the user of described candidate value
Fixed;And described user to be identified is the user to be identified meeting predetermined precondition;And
Described Weight Acquisition unit is additionally operable to from the model pre-build obtain and described predetermined preposition bar
Part, user to be identified at least one feature described in each feature and the property value of feature
The weight joined.
In certain embodiments, described device also includes: success sample acquisition unit, is used for
After determining potential target user, obtain the identification becoming targeted customer in potential target user
Successful user set;Weight updating block, for by described identification successful user set and
The user of each candidate value associated by each feature in described basic user information aggregate
Accounting contrast, the described weight recalculating and updating in described model.
In certain embodiments, described characteristic acquisition unit includes: pretreatment subelement,
For the user profile set collected in advance is carried out at least one in following process: by user
Former property value discrete in information aggregate is updated to each predetermined interval for representing former property value
The property value of scope;By the property value that the property value of user in user profile set is empty feature
It is set to the default value preset;For each described feature, delete corresponding weight less than predetermined
The property value of threshold value;Characteristic information extraction subelement, for from through described pretreatment subelement
Process after user profile set in obtain the attribute of at least one feature of user to be identified
Value.
The user identification method of the application offer and device, by from the above-mentioned model pre-build
Middle acquisition and each feature at least one feature described of user to be identified and the attribute of feature
The weight of value coupling, and with the size of described weight sum as foundation, identify that user to be identified is
No for potential target user, based on the property value of more feature, user can be identified,
It is thus possible to identify more potential target user exactly.
Accompanying drawing explanation
By reading retouching in detail with reference to made non-limiting example is made of the following drawings
Stating, other features, purpose and advantage will become more apparent upon:
Fig. 1 is that the application can apply to exemplary system architecture figure therein;
Fig. 2 is the flow chart of an embodiment of the user identification method according to the application;
Fig. 3 is showing of the data process of an embodiment of the user identification method according to the application
Example schematic diagram;
Fig. 4 is the structural representation of an embodiment of the customer identification device according to the application;
Fig. 5 is adapted for the structure of the computer system of the server for realizing the embodiment of the present application
Schematic diagram.
Detailed description of the invention
With embodiment, the application is described in further detail below in conjunction with the accompanying drawings.It is appreciated that
, specific embodiment described herein is used only for explaining related invention, rather than to this
Bright restriction.It also should be noted that, for the ease of describe, accompanying drawing illustrate only with
About the part that invention is relevant.
It should be noted that in the case of not conflicting, the embodiment in the application and embodiment
In feature can be mutually combined.Describe this below with reference to the accompanying drawings and in conjunction with the embodiments in detail
Application.
Fig. 1 shows the reality of user identification method or the customer identification device that can apply the application
Execute the exemplary system architecture 100 of example.
As it is shown in figure 1, system architecture 100 can include terminal device 101,102,103,
Network 104 and server 105.Network 104 is in order at terminal device 101,102,103 and
The medium of communication link is provided between server 105.Network 104 can include various connection class
Type, the most wired, wireless communication link or fiber optic cables etc..
User can use terminal device 101,102,103 by network 104 and server 105
Alternately, to receive or to send message etc..Can be provided with on terminal device 101,102,103
Various client application, class of such as calling a taxi application, map search are served by.
Terminal device 101,102,103 can be the various electronic equipments of support information communication,
Include but not limited to smart mobile phone, panel computer, pocket computer on knee and desktop computer
Etc..
Server 105 can be to provide the server of various service, such as to terminal device 101,
102, the class application of calling a taxi on 103, map search the user profile of transmission such as are served by and carry out
Storage, analysis etc. process, it is possible to give corresponding user according to result PUSH message.
It should be noted that the user identification method that the embodiment of the present application is provided is generally by servicing
Device 105 performs.Correspondingly, during customer identification device is generally disposed at server 105.
It should be understood that the number of terminal device, network and the server in Fig. 1 is only signal
Property.According to realizing needs, can have any number of terminal device, network and server.
A reality of the user identification method according to the application is shown with continued reference to Fig. 2, Fig. 2
Execute the flow process 200 of example.
As in figure 2 it is shown, the user identification method of the present embodiment comprises the following steps:
Step 201, obtains user to be identified at least from the user profile set collected in advance
The property value of one feature.
In the present embodiment, user identification method runs on electronic equipment thereon (such as Fig. 1
Shown server) can locally or remotely obtain from the user profile set collected in advance
Take the property value of at least one feature of user to be identified.Wherein, above-mentioned user profile set can
To be user's representation data;Above-mentioned user to be identified can be one or more, above-mentioned at least one
Individual feature can include but not limited to: sex, age, income level, schooling, institute
In industry, consumption habit, personal interest etc. may affect the feature identifying result to be identified
One or more features.
In some optional implementations of the present embodiment, electronic equipment can be first in advance
The user profile set collected carries out at least one in following process: by user profile set
Discrete former property value is updated to the attribute of each predetermined interval scope for representing former property value
Value;It is set to preset by the property value that the property value of user in user profile set is empty feature
Default value;For each features described above, delete the corresponding weight attribute less than predetermined threshold
Value.Then, from the user profile set after above-mentioned process, obtain user's to be identified
The property value of at least one feature.Wherein, as a example by age characteristics, discrete former property value leads to
It is often concrete age numerical value (such as 21,22,30), former property value is updated to for table
Show the property value of each predetermined interval scope of former property value, afterwards, associated by age characteristics
Property value such as may include that 20~25,25~29, more than 30.By this implementation,
The aspect such as identification, hit rate, stability can have and improve significantly.
Step 202, obtains above-mentioned at least with user to be identified from the model pre-build
Each feature in individual feature and the weight of the property value coupling of feature.
Wherein, above-mentioned model includes following information: the candidate that feature associates with features described above belongs to
The weight that property value associates with features described above and above-mentioned candidate value, above-mentioned weight is by right
The property value etc. of features described above in the targeted customer's information aggregate prestored and basic user information aggregate
Accounting in the user of above-mentioned candidate value carries out contrasting and determines.
In the present embodiment, the feature in above-mentioned model and property value can use signature identification respectively
Represent with property value mark.Such as, ith feature i represents, the jth associated by feature i
Individual property value j represents, wherein, i, j are positive integer, i ∈ 1 ... M}, j ∈ 1 ... Ni+ 1},
M is the quantity of feature, N at least one feature above-mentionediIt is characterized property value associated by i
Quantity.If the weight associated with the property value j associated by feature i and feature i in model is Sij,
Electronic equipment can according to each feature at least one feature above-mentioned and the property value of feature,
The S mated with each property value of each feature and feature is mated in above-mentioned modelij, obtain
Mate with the property value of each feature at least one feature above-mentioned of user to be identified and feature
Weight.
Wherein, SijIt is beforehand through to the targeted customer's information aggregate prestored and basic user information
In set, the property value of the feature i accounting of user equal to property value j carries out contrasting and determines.Its
In, the targeted customer in targeted customer's information aggregate can be mobile phone arrived actually used certain
The user of product (such as certain application of calling a taxi), the user in basic user information aggregate is not have
It is determined as the general user of potential target user.As a example by this feature of the level of consumption, this spy
Levy associated property value and may include that high, medium and low.Electronic equipment can obtain the most respectively
Take targeted customer's information aggregate and and basic user information aggregate in the level of consumption be high user account for
User's accounting and the level of consumption in than, the level of consumption being are low user's accounting, such as:
The most respectively by accounting and base that the level of consumption in targeted customer's information aggregate is high user
User's accounting that in plinth user profile set, the level of consumption is high contrasts, if differed greatly,
Then the weight associated by property value height of level of consumption feature is the most relatively large.For in property value
With low, it be also adopted by said method and obtain associated weight respectively.
In some optional implementations of the present embodiment, use above by the target prestored
In family information aggregate and basic user information aggregate, the property value of features described above belongs to equal to above-mentioned candidate
The accounting of the user of property value carries out contrast and may include that and obtain above-mentioned targeted customer's information collection respectively
Close and in above-mentioned basic user information aggregate, the property value of features described above is equal to above-mentioned candidate value
The ratio shared in set of user;Obtain above-mentioned targeted customer's information aggregate and above-mentioned basis
The absolute value of the difference of the aforementioned proportion of user profile set, as with features described above and above-mentioned attribute
The weight of value association.
Additionally, due to a user is in the weight of the feature association of many property values with at few property value
If the weight of feature association is identical, the weight of many property values feature should be bigger.Therefore, may be used
Selection of land, above by the targeted customer's information aggregate prestored and basic user information aggregate
The accounting of the user that the property value stating feature is equal to above-mentioned candidate value carries out contrast and can also wrap
Include: according to the quantity of the candidate value associated with features described above, revise above-mentioned weight, wherein,
Positive correlation is there is between the weight and the quantity of above-mentioned candidate value that revise.Such as, if
Above-mentioned former weight is S 'ij, the most revised SijCan be: log (Ni)×S’ij.By above-mentioned right
The correction of weight so that become positive correlation between the quantity of the above-mentioned candidate value of weight in model
Relation, so that the weight in model is more rationally, accurately.
Step 203, obtain each feature at least one feature above-mentioned of user to be identified and
The weight sum of the property value coupling of feature.
In the present embodiment, electronic equipment can be by the user to be identified of acquisition in step 202
Each feature and the weight of the property value coupling of feature at least one feature above-mentioned add
With, it is thus achieved that above-mentioned weight sum, and should can be used to indicate that user to be identified is that potential target is used
The significance at family, possibility.
Whether step 204, with the size of above-mentioned weight sum as foundation, identify user to be identified
For potential target user.
In the present embodiment, above-mentioned weight sum can be carried out by electronic equipment with the threshold value preset
Relatively, if above-mentioned weight sum is more than this threshold value, then it is latent for can identifying user to be identified
Targeted customer.If additionally, user to be identified has multiple, electronic equipment can be according to above-mentioned
Weight sum the most once chooses predetermined quantity user to be identified, uses as potential target
Family.
In some optional implementations of the present embodiment, above-mentioned model also includes precondition
Information, the weight associated with features described above and above-mentioned property value in above-mentioned model for above-mentioned before
Put condition, weight that features described above associates with above-mentioned property value.Wherein, above-mentioned weight is to pass through
To the targeted customer's information aggregate meeting above-mentioned precondition prestored and basic user information aggregate
The property value of middle features described above carries out contrast equal to the accounting of the user of above-mentioned candidate value and determines
's.And, above-mentioned user to be identified is the user to be identified meeting predetermined precondition.And,
Above-mentioned obtain from the model pre-build with at least one feature above-mentioned of user to be identified
The weight of the property value coupling of each feature and feature may include that from the model pre-build
Obtain with above-mentioned predetermined precondition, user to be identified at least one feature above-mentioned in each
The weight of the property value coupling of feature and feature.
Wherein, above-mentioned precondition can be such as regional condition, user profile acquisition time bar
Part etc..Such as, the ratio of the feature such as the level of consumption of the user of different cities, schooling
Distribution usually has discrepant, if according to the targeted customer's information aggregate by a line city
With the weight of the acquisition of information of basic user information aggregate, the user of small city is identified, effect
Fruit is generally lower than the degree of accuracy being identified the user in a line city.Therefore, real by this
Existing mode so that the precondition of user to be verified and generate model based on user profile collection
The potential condition closed is identical so that the degree of accuracy that user identifies is higher.
In some optional implementations of the present embodiment, the user identification method of the present embodiment
Can also include: after determining potential target user, obtain in potential target user and become
The identification successful user set of targeted customer;By to above-mentioned identification successful user set and above-mentioned
The user of each candidate value associated by each feature in basic user information aggregate accounts for
Ratio contrasts, the above-mentioned weight recalculating and updating in above-mentioned model.With targeted customer it is
As a example by using the user of certain product, electronic equipment can certain time after step 204
After Duan (after such as 1 month), first pass through and obtain the potential target identified in step 204
User employed in this time period the user of this product, is identified successful user set.
Then by the method for step 202, to above-mentioned identification successful user set and above-mentioned basic user
It is right that the accounting of the user of each candidate value associated by each feature in information aggregate is carried out
Ratio, the above-mentioned weight recalculating and updating in above-mentioned model.By this implementation, to mould
Type has carried out tuning, improves the degree of accuracy of subsequent user identification.
One of the user identification method according to the present embodiment is shown below with reference to Fig. 3, Fig. 3
Illustrative diagram.
As it is shown on figure 3, electronic equipment can targeted customer's information aggregate based on basic data layer
301 and basic user information aggregate 302, in model layer by preposition in precondition set 303
The screening (also can not pass through precondition) of condition obtains basic user sample 304 and target is used
The sample 305 at family;Then by carrying out user characteristics as shown in Figure 3 and extracting, contrast ratio
Distribution calculating, weight calculation etc. process (i.e. the step 202 of the present embodiment), obtain model 306.
When carrying out user and identifying, special by user user 307 to be identified carried out as shown in Figure 3
Levy the process (i.e. the step 201 of the present embodiment) of extraction;And based on the user to be identified being drawn into
The property value of feature and above-mentioned model, carry out the targeted customer's significance shown in Fig. 3 and calculate (i.e.
The step 203 of the present embodiment);Finally obtaining the potential target user shown in Fig. 3, to gather 308 (logical
Cross the step 204 of the present embodiment).
The user identification method that the present embodiment provides, by obtaining from the above-mentioned model pre-build
Take and each feature at least one feature above-mentioned of user to be identified and the property value of feature
The weight joined, and with the size of above-mentioned weight sum as foundation, identifies that whether user to be identified is
Potential target user, can be identified user based on the property value of more feature, from
And more potential target user can be identified exactly.
With further reference to Fig. 4, as to the realization of method shown in above-mentioned each figure, the application provides
One embodiment of a kind of customer identification device, this device embodiment and the method shown in Fig. 2
Embodiment is corresponding, and this device specifically can apply in server.
As shown in Figure 4, the customer identification device 400 that the present embodiment provides includes: characteristic information
Acquiring unit 401, Weight Acquisition unit 402, add and unit 403 and recognition unit 404.
Wherein, characteristic acquisition unit 401 is for obtaining from the user profile set collected in advance
The property value of at least one feature of user to be identified;Weight Acquisition unit 402 is for from advance
The model set up obtains with each feature at least one feature above-mentioned of user to be identified and
The weight of the property value coupling of feature, wherein, above-mentioned model includes following information: feature and
The power that the candidate value of features described above association associates with features described above and above-mentioned candidate value
Weight, above-mentioned weight is by the targeted customer's information aggregate prestored and basic user information aggregate
The property value of middle features described above carries out contrast equal to the accounting of the user of above-mentioned candidate value and determines
's;Add and each at least one feature above-mentioned obtaining user to be identified of unit 403
The weight sum of the property value coupling of feature and feature;Recognition unit 404 is for above-mentioned weight
The size of sum is foundation, identifies whether user to be identified is potential target user.
In the present embodiment, characteristic acquisition unit 401, Weight Acquisition unit 402, add and
The concrete process of unit 403 and recognition unit 404 can be respectively with reference in Fig. 2 correspondence embodiment
The related description of step 201, step 202, step 203 and step 204, the most superfluous at this
State.
In some optional implementations of the present embodiment, characteristic acquisition unit 401 can
To include: pretreatment subelement 4011 and characteristic information extraction subelement 4012.Wherein, in advance
Process subelement 4011 for the user profile set collected in advance is carried out in following process
At least one: former property value discrete in user profile set is updated to for representing belonging to originally property
The property value of each predetermined interval scope of value;By the property value of user in user profile set it is
The property value of empty feature is set to the default value preset;For each features described above, it is right to delete
The weight answered is less than the property value of predetermined threshold.Characteristic information extraction subelement 4012 is for from warp
User profile set after crossing the process of above-mentioned pretreatment subelement obtains user's to be identified
The property value of at least one feature.The concrete technology effect processing and being brought of this implementation
Fruit refers in Fig. 2 correspondence embodiment the relevant portion of corresponding optional implementation in step 201
Explanation, do not repeat them here.
In some optional implementations of the present embodiment, the customer identification device of the present embodiment
Can also include: weight determining unit 405, every for at least one feature above-mentioned
Individual feature and each property value associated with features described above, by the targeted customer's information prestored
In set and basic user information aggregate, the property value of feature is equal to accounting for of the user of candidate value
The weight associated with features described above and above-mentioned candidate value is determined than carrying out contrasting.Wherein, power
Heavily determine that unit 405 may include that ratio obtains subelement 4051, above-mentioned for obtaining respectively
In targeted customer's information aggregate and above-mentioned basic user information aggregate, the property value of features described above is equal to
The ratio that the user of above-mentioned candidate value is shared in set;Weight determines subelement 4052,
For obtaining above-mentioned targeted customer's information aggregate and the aforementioned proportion of above-mentioned basic user information aggregate
The absolute value of difference, as the weight associated with features described above and above-mentioned property value.This realization side
Formula specifically process the corresponding optional realization side referred in Fig. 2 correspondence embodiment in step 202
The related description of formula, does not repeats them here.
Based on a upper implementation, in some optional implementations of the present embodiment, weight
Determine that unit 405 can also include: weight correction subelement 4053, for basis and above-mentioned spy
Levying the quantity of the candidate value of association, revise above-mentioned weight, wherein, the weight of correction is with upper
State and there is positive correlation between the quantity of candidate value.This implementation concrete process and
It is the most optional that its technique effect brought refers in Fig. 2 correspondence embodiment in step 202
The explanation of the relevant portion of implementation, does not repeats them here.
In some optional implementations of the present embodiment, above-mentioned model can also include preposition
The weight that conditional information, above-mentioned and features described above in above-mentioned model and above-mentioned property value associate can
Think the weight associated with above-mentioned precondition, features described above and above-mentioned property value.Wherein, on
Stating weight can be that above-mentioned weight determining unit is by the mesh meeting above-mentioned precondition prestored
In mark user profile set and basic user information aggregate, the property value of features described above is equal to above-mentioned time
The accounting selecting the user of property value carries out contrasting and determines.And, above-mentioned user to be identified is full
The user to be identified of the predetermined precondition of foot.And, Weight Acquisition unit 402 can be also used for
From the model pre-build obtain with above-mentioned predetermined precondition, user to be identified above-mentioned to
Each feature in a few feature and the weight of the property value coupling of feature.This implementation
The concrete technique effect processing and being brought refers to corresponding optional reality in Fig. 2 correspondence embodiment
The explanation of the relevant portion of existing mode, does not repeats them here.
In some optional implementations of the present embodiment, the customer identification device of the present embodiment
Can also include: success sample acquisition unit 406, for determine potential target user it
After, obtain the identification successful user set becoming targeted customer in potential target user;Weight is more
New unit 407, for by above-mentioned identification successful user set and above-mentioned basic user information
The accounting of the user of each candidate value associated by each feature in set contrasts,
The above-mentioned weight recalculated and update in above-mentioned model.This implementation concrete process and
The technique effect brought refers to the relevant of corresponding optional implementation in Fig. 2 correspondence embodiment
The explanation of part, does not repeats them here.
The customer identification device that the present embodiment provides, is built from advance by Weight Acquisition unit 402
Vertical above-mentioned model obtains and each feature at least one feature above-mentioned of user to be identified
The weight mated with the property value of feature, and by recognition unit 404 by add with unit 403 in terms of
The size of the weight sum calculated is foundation, identifies whether user to be identified is potential target user,
Based on the property value of more feature, user can be identified such that it is able to know exactly
Do not go out more potential target user.
Below with reference to Fig. 5, it illustrates the server that is suitable to for realizing the embodiment of the present application
The structural representation of computer system 500.
As it is shown in figure 5, computer system 500 includes CPU (CPU) 501, its
Can be according to the program being stored in read-only storage (ROM) 502 or from storage part 508
It is loaded into the program in random access storage device (RAM) 503 and performs various suitable action
And process.In RAM 503, also storage has system 500 to operate required various program sums
According to.CPU 501, ROM 502 and RAM 503 are connected with each other by bus 504.Input
/ output (I/O) interface 505 is also connected to bus 504.
It is connected to I/O interface 505: include the storage part 506 of hard disk etc. with lower component;And
Communications portion 507 including the NIC of such as LAN card, modem etc..Communication
Part 507 performs communication process via the network of such as internet.Driver 508 is also according to needing
I/O interface 505 to be connected to.Detachable media 509, such as disk, CD, magneto-optic disk,
Semiconductor memory etc., is arranged on driver 508, in order to read from it as required
The computer program gone out is mounted into storage part 506 as required.
Especially, according to embodiment of the disclosure, the process described above with reference to flow chart is permissible
It is implemented as computer software programs.Such as, embodiment of the disclosure and include a kind of computer journey
Sequence product, it includes the computer program being tangibly embodied on machine readable media, above-mentioned meter
Calculation machine program comprises the program code for performing the method shown in flow chart.In such enforcement
In example, this computer program can be downloaded and installed from network by communications portion 507,
And/or be mounted from detachable media 509.At this computer program by CPU
(CPU), during 501 execution, the above-mentioned functions limited in the present processes is performed.
Flow chart in accompanying drawing and block diagram, it is illustrated that according to the various embodiment of the application system,
Architectural framework in the cards, function and the operation of method and computer program product.This point
On, each square frame in flow chart or block diagram can represent a module, program segment or code
A part, a part for above-mentioned module, program segment or code comprise one or more for
Realize the executable instruction of the logic function of regulation.It should also be noted that at some as replacement
In realization, the function marked in square frame can also be sent out to be different from the order marked in accompanying drawing
Raw.Such as, two square frames succeedingly represented can essentially perform substantially in parallel, they
Sometimes can also perform in the opposite order, this is depending on involved function.It is also noted that
It is, the square frame in each square frame in block diagram and/or flow chart and block diagram and/or flow chart
Combination, can realize by the special hardware based system of the function or operation that perform regulation,
Or can realize with the combination of specialized hardware with computer instruction.
Being described in the embodiment of the present application involved unit can be real by the way of software
Existing, it is also possible to realize by the way of hardware.Described unit can also be arranged on process
In device, for example, it is possible to be described as: a kind of processor includes characteristic acquisition unit, weight
Acquiring unit, add and unit and recognition unit.Wherein, the title of these unit is in certain feelings
Being not intended that the restriction to this unit itself under condition, such as, characteristic acquisition unit is all right
It is described as " from the user profile set collected in advance, obtaining at least one of user to be identified
The unit of the property value of feature ".
As on the other hand, present invention also provides a kind of nonvolatile computer storage media,
This nonvolatile computer storage media can be in above-described embodiment included in said apparatus
Nonvolatile computer storage media;Can also be individualism, be unkitted allocate in terminal non-
Volatile computer storage medium.Above-mentioned nonvolatile computer storage media storage have one or
The multiple program of person, when said one or multiple program are performed by an equipment so that above-mentioned
Equipment: obtain at least one feature of user to be identified from the user profile set collected in advance
Property value;At least one spy above-mentioned with user to be identified is obtained from the model pre-build
The weight of the property value coupling of each feature in levying and feature, wherein, above-mentioned model include with
Lower information: candidate value that feature associates with features described above and features described above and above-mentioned time
Select the weight that property value associates, above-mentioned weight be by the targeted customer's information aggregate prestored and
In basic user information aggregate, the property value of features described above is equal to the user's of above-mentioned candidate value
Accounting carries out contrasting and determines;Obtain at least one feature above-mentioned of user to be identified each
The weight sum of the property value coupling of feature and feature;With the size of above-mentioned weight sum as foundation,
Identify whether user to be identified is potential target user.
Above description is only the preferred embodiment of the application and saying institute's application technology principle
Bright.It will be appreciated by those skilled in the art that invention scope involved in the application, do not limit
In the technical scheme of the particular combination of above-mentioned technical characteristic, also should contain simultaneously without departing from
In the case of foregoing invention design, above-mentioned technical characteristic or its equivalent feature are combined
And other technical scheme formed.Such as features described above and (but not limited to) disclosed herein
The technical characteristic with similar functions is replaced mutually and the technical scheme that formed.
Claims (12)
1. a user identification method, it is characterised in that described method includes:
At least one feature of user to be identified is obtained from the user profile set collected in advance
Property value;
Obtain from the model pre-build and at least one feature described in user to be identified
The weight of the property value coupling of each feature and feature, wherein, described model includes following information:
Feature and the candidate value of described feature association and described feature and described candidate value
The weight of association, described weight is by the targeted customer's information aggregate prestored and basic user
The property value of feature described in information aggregate is carried out equal to the accounting of the user of described candidate value
Contrast determines;
Obtain each feature at least one feature described of user to be identified and the attribute of feature
The weight sum of value coupling;
With the size of described weight sum as foundation, identify whether user to be identified is potential target
User.
Method the most according to claim 1, it is characterised in that described by prestoring
The property value of feature described in targeted customer's information aggregate and basic user information aggregate is equal to described
The accounting of the user of candidate value contrasts, including:
Obtain respectively described in described targeted customer's information aggregate and described basic user information aggregate
The ratio that the property value of feature is shared in set equal to the user of described candidate value;
Obtain described targeted customer's information aggregate and the described ratio of described basic user information aggregate
The absolute value of difference, as the weight associated with described feature and described property value.
Method the most according to claim 2, it is characterised in that described by prestoring
The property value of feature described in targeted customer's information aggregate and basic user information aggregate is equal to described
The accounting of the user of candidate value contrasts, and also includes:
According to the quantity with the candidate value of described feature association, revise described weight, wherein,
Positive correlation is there is between the weight and the quantity of described candidate value that revise.
4. according to the arbitrary described method of claim 1-3, it is characterised in that described model is also
Including precondition information, described and described feature and described property value in described model associate
Weight be the weight associated with described precondition, described feature and described property value, wherein,
Described weight is by the targeted customer's information aggregate meeting described precondition prestored and base
The property value of feature described in plinth user profile set is equal to accounting for of the user of described candidate value
Determine than carrying out contrasting;And
Described user to be identified is the user to be identified meeting predetermined precondition;And
Described acquisition and at least one feature described in user to be identified from the model pre-build
In each feature and feature property value coupling weight, including:
Obtain and described predetermined precondition, the institute of user to be identified from the model pre-build
State the weight of the property value coupling of each feature at least one feature and feature.
5. according to the arbitrary described method of claim 1-3, it is characterised in that described method is also
Including:
After determining potential target user, obtain in potential target user and become targeted customer
Identification successful user set;
Each by described identification successful user set and described basic user information aggregate
The accounting of the user of each candidate value associated by feature contrasts, and recalculates and more
Described weight in new described model.
6. according to the arbitrary described method of claim 1-3, it is characterised in that described from advance
The user profile set collected obtains the property value of at least one feature of user to be identified, bag
Include:
User profile set to collecting in advance carries out at least one in following process: by user
Former property value discrete in information aggregate is updated to each predetermined interval for representing former property value
The property value of scope;By the property value that the property value of user in user profile set is empty feature
It is set to the default value preset;For each described feature, delete corresponding weight less than predetermined
The property value of threshold value;
At least the one of user to be identified is obtained from the user profile set after described process
The property value of individual feature.
7. a customer identification device, it is characterised in that described device includes:
Characteristic acquisition unit, waits to know for obtaining from the user profile set collected in advance
The property value of at least one feature of other user;
Weight Acquisition unit, for obtaining the institute with user to be identified from the model pre-build
State the weight of the property value coupling of each feature at least one feature and feature, wherein, institute
State model and include following information: feature and the candidate value of described feature association are with described
The weight that feature associates with described candidate value, described weight is by using the target prestored
The property value of feature described in family information aggregate and basic user information aggregate belongs to equal to described candidate
The accounting of the user of property value carries out contrasting and determines;
Adding and unit, each at least one feature described obtaining user to be identified is special
The weight sum of the property value coupling of feature of seeking peace;
Recognition unit, for the size of described weight sum as foundation, identifying user to be identified
Whether it is potential target user.
Device the most according to claim 7, it is characterised in that described device also includes:
Weight determining unit, for for each feature at least one feature described and with institute
State each property value of feature association, by the targeted customer's information aggregate prestored and basis are used
In the information aggregate of family, the property value of feature carries out contrast really equal to the accounting of the user of candidate value
The fixed weight associated with described feature and described candidate value, described weight determining unit includes:
Ratio obtains subelement, for obtaining described targeted customer's information aggregate and described base respectively
The property value of feature described in plinth user profile set is collecting equal to the user of described candidate value
Ratio shared in conjunction;
Weight determines subelement, is used for obtaining described targeted customer's information aggregate and described basis is used
The absolute value of the difference of the described ratio of family information aggregate, as with described feature and described property value
The weight of association.
Device the most according to claim 8, it is characterised in that described weight determining unit
Also include:
Weight correction subelement, for the quantity of basis with the candidate value of described feature association,
Revise described weight, wherein, exist between weight and the quantity of described candidate value of correction
Positive correlation.
10. according to the arbitrary described device of claim 7-9, it is characterised in that described model
Also including precondition information, described and described feature and described property value in described model close
The weight of connection is the weight associated with described precondition, described feature and described property value, its
In, described weight is that described weight determining unit is by meeting described precondition to prestore
The property value of feature described in targeted customer's information aggregate and basic user information aggregate is equal to described
The accounting of the user of candidate value carries out contrasting and determines;And
Described user to be identified is the user to be identified meeting predetermined precondition;And
Described Weight Acquisition unit be additionally operable to from the model pre-build to obtain with described predetermined before
Put each feature at least one feature described of condition, user to be identified and the attribute of feature
The weight of value coupling.
11. according to the arbitrary described device of claim 7-9, it is characterised in that described device
Also include:
Success sample acquisition unit, for after determining potential target user, obtains potential
Targeted customer becomes the identification successful user set of targeted customer;
Weight updating block, for by using described identification successful user set and described basis
The accounting of the user of each candidate value associated by each feature in the information aggregate of family is carried out
Contrast, the described weight recalculating and updating in described model.
12. according to the arbitrary described device of claim 7-9, it is characterised in that described feature
Information acquisition unit includes:
Pretreatment subelement, for carrying out in following process the user profile set collected in advance
At least one: former property value discrete in user profile set is updated to for represent belong to originally
The property value of each predetermined interval scope of property value;By the property value of user in user profile set
The default value preset it is set to for the property value of empty feature;For each described feature, delete
Corresponding weight is less than the property value of predetermined threshold;
Characteristic information extraction subelement, for after the process through described pretreatment subelement
User profile set in obtain the property value of at least one feature of user to be identified.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610197077.1A CN105894028B (en) | 2016-03-31 | 2016-03-31 | User identification method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610197077.1A CN105894028B (en) | 2016-03-31 | 2016-03-31 | User identification method and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105894028A true CN105894028A (en) | 2016-08-24 |
CN105894028B CN105894028B (en) | 2020-01-10 |
Family
ID=57011752
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610197077.1A Active CN105894028B (en) | 2016-03-31 | 2016-03-31 | User identification method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105894028B (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106294881A (en) * | 2016-08-30 | 2017-01-04 | 五八同城信息技术有限公司 | information identifying method and device |
CN108768743A (en) * | 2018-06-11 | 2018-11-06 | 北京奇艺世纪科技有限公司 | A kind of user identification method, device and server |
CN109377284A (en) * | 2018-11-05 | 2019-02-22 | 连尚(新昌)网络科技有限公司 | Method and electronic equipment for pushed information |
CN110059244A (en) * | 2019-02-01 | 2019-07-26 | 阿里巴巴集团控股有限公司 | Audient's extended method and device |
CN110110084A (en) * | 2019-04-23 | 2019-08-09 | 北京科技大学 | The recognition methods of high quality user-generated content |
CN110334936A (en) * | 2019-06-28 | 2019-10-15 | 阿里巴巴集团控股有限公司 | A kind of construction method, device and the equipment of credit qualification Rating Model |
CN111582906A (en) * | 2020-03-26 | 2020-08-25 | 口碑(上海)信息技术有限公司 | Target user information acquisition method and device and electronic equipment |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104244314A (en) * | 2014-07-30 | 2014-12-24 | 北京拓明科技有限公司 | Potential group client identification method based on Mc interface signaling |
CN104866626A (en) * | 2015-06-15 | 2015-08-26 | 中国移动通信集团黑龙江有限公司 | Method and device for recommending telecommunication service |
CN105610768A (en) * | 2014-11-25 | 2016-05-25 | 阿里巴巴集团控股有限公司 | Method and device for processing network operation |
CN106095916A (en) * | 2016-06-08 | 2016-11-09 | 百度在线网络技术(北京)有限公司 | Information-pushing method and device |
-
2016
- 2016-03-31 CN CN201610197077.1A patent/CN105894028B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104244314A (en) * | 2014-07-30 | 2014-12-24 | 北京拓明科技有限公司 | Potential group client identification method based on Mc interface signaling |
CN105610768A (en) * | 2014-11-25 | 2016-05-25 | 阿里巴巴集团控股有限公司 | Method and device for processing network operation |
CN104866626A (en) * | 2015-06-15 | 2015-08-26 | 中国移动通信集团黑龙江有限公司 | Method and device for recommending telecommunication service |
CN106095916A (en) * | 2016-06-08 | 2016-11-09 | 百度在线网络技术(北京)有限公司 | Information-pushing method and device |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106294881A (en) * | 2016-08-30 | 2017-01-04 | 五八同城信息技术有限公司 | information identifying method and device |
CN108768743A (en) * | 2018-06-11 | 2018-11-06 | 北京奇艺世纪科技有限公司 | A kind of user identification method, device and server |
CN108768743B (en) * | 2018-06-11 | 2021-07-20 | 北京奇艺世纪科技有限公司 | User identification method and device and server |
CN109377284A (en) * | 2018-11-05 | 2019-02-22 | 连尚(新昌)网络科技有限公司 | Method and electronic equipment for pushed information |
CN110059244A (en) * | 2019-02-01 | 2019-07-26 | 阿里巴巴集团控股有限公司 | Audient's extended method and device |
CN110110084A (en) * | 2019-04-23 | 2019-08-09 | 北京科技大学 | The recognition methods of high quality user-generated content |
CN110334936A (en) * | 2019-06-28 | 2019-10-15 | 阿里巴巴集团控股有限公司 | A kind of construction method, device and the equipment of credit qualification Rating Model |
CN110334936B (en) * | 2019-06-28 | 2023-09-29 | 创新先进技术有限公司 | Method, device and equipment for constructing credit qualification scoring model |
CN111582906A (en) * | 2020-03-26 | 2020-08-25 | 口碑(上海)信息技术有限公司 | Target user information acquisition method and device and electronic equipment |
Also Published As
Publication number | Publication date |
---|---|
CN105894028B (en) | 2020-01-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105894028A (en) | User identification method and device | |
CN107908789A (en) | Method and apparatus for generating information | |
US11100421B2 (en) | Customized website predictions for machine-learning systems | |
CN105930425A (en) | Personalized video recommendation method and apparatus | |
CN111966904B (en) | Information recommendation method and related device based on multi-user portrait model | |
CN106503269A (en) | Method, device and server that application is recommended | |
CN105306495B (en) | user identification method and device | |
CN109978033A (en) | The method and apparatus of the building of biconditional operation people's identification model and biconditional operation people identification | |
CN107609890A (en) | A kind of method and apparatus of order tracking | |
CN105721629A (en) | User identifier matching method and device | |
CN107423613A (en) | The method, apparatus and server of device-fingerprint are determined according to similarity | |
CN107545038B (en) | Text classification method and equipment | |
CN107679119A (en) | The method and apparatus for generating brand derivative words | |
CN105956469A (en) | Method and device for identifying file security | |
CN113592605B (en) | Product recommendation method, device, equipment and storage medium based on similar products | |
CN110909222A (en) | User portrait establishing method, device, medium and electronic equipment based on clustering | |
CN108255706A (en) | Edit methods, device, terminal device and the storage medium of automatic test script | |
CN110349013A (en) | Risk control method and device | |
CN107885754B (en) | Method and device for extracting credit variable from transaction data based on LDA model | |
CN106779856A (en) | The method and device that big data customer value is evaluated | |
CN107291774A (en) | Error sample recognition methods and device | |
CN107944026A (en) | A kind of method, apparatus, server and the storage medium of atlas personalized recommendation | |
CN106446844A (en) | Pose estimation method, pose estimation device and computer system | |
CN115114329A (en) | Method and device for detecting data stream abnormity, electronic equipment and storage medium | |
CN107368407A (en) | Information processing method and device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |