CN108804462A - Method, apparatus and server are recommended in a kind of advertisement - Google Patents
Method, apparatus and server are recommended in a kind of advertisement Download PDFInfo
- Publication number
- CN108804462A CN108804462A CN201710304206.7A CN201710304206A CN108804462A CN 108804462 A CN108804462 A CN 108804462A CN 201710304206 A CN201710304206 A CN 201710304206A CN 108804462 A CN108804462 A CN 108804462A
- Authority
- CN
- China
- Prior art keywords
- user
- attribute
- characteristic
- characteristic attribute
- feature
- 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
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0251—Targeted advertisements
- G06Q30/0269—Targeted advertisements based on user profile or attribute
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0277—Online advertisement
Landscapes
- Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Finance (AREA)
- Economics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the invention discloses a kind of advertisements to recommend method, apparatus and server, and by obtaining the behavioral data of the first user, the behavioral data includes the data information with the relevant at least one characteristic attribute of ability to pay;The data information of each of first user characteristic attribute is handled respectively, obtains the characteristic value of each of first user characteristic attribute;Determine the weight of each characteristic attribute;According to the characteristic value of each characteristic attribute of first user and corresponding weight, the appraisal result of first user is determined;The user gradation belonging to first user is determined based on the appraisal result, and recommend the mode of advertisement corresponding with identified user gradation to first user, assessment of the behavioral data to the ability to pay of user of the user based on the ability to pay for being more prone to reaction user is realized, and then ensure that effective popularization of advertisement in computer application.
Description
Technical field
The present invention relates to field of computer technology, and in particular to method, apparatus and server are recommended in a kind of advertisement.
Background technology
With the rapid development of network technology, computer application has become the important work that people carry out advertisement recommendation
Tool.Computer application generates dislike during carrying out advertisement recommendation, in order to reduce advertisement side to the advertisement received
The generation of the case where mood and then the promotion effect of influence advertisement, it will usually avoid recommending to prop up with it to advertisement side as possible
Pay the unmatched advertisement of ability.For example, when the ability to pay of advertisement side is 5000 yuan, if recommending to the advertisement side
Advertisement needed for ability to pay be 50,000 yuan, the advertisement at this time may largely be thought by the advertisement side
It is harassing and wrecking information, the popularization for causing the advertisement side to generate dislike mood, influence advertisement.
The prior art is typically the payment energy that the data such as payroll, bank's flowing water directly using user determine user
Power, and then propagated and the matched information of its ability to pay to the user.But it has been investigated that, it determines in this manner
The user's ability to pay gone out is often more prone to the earning power of reaction user, between the ability to pay of real user simultaneously
Without too big correlation.
In view of this, providing a kind of advertisement recommends method, apparatus and server, to realize commenting to the ability to pay of user
Estimate, and then ensure effective popularization of advertisement in computer application, is a problem to be solved.
Invention content
In view of this, method, apparatus and server are recommended in a kind of advertisement of offer of the embodiment of the present invention, to realize to user's
The assessment of ability to pay, and then ensure effective popularization of advertisement in computer application.
To achieve the above object, the embodiment of the present invention provides the following technical solutions:
A kind of advertisement recommendation method, including:
The behavioral data of the first user is obtained, the behavioral data includes and the relevant at least one feature of ability to pay
The data information of attribute;
Based on the behavioral data of at least one second user, respectively to each of first user characteristic attribute
Data information is handled, and the characteristic value of each of first user characteristic attribute is obtained;
Utilize each of the characteristic value of each of first user characteristic attribute and each described second user
The characteristic value of the characteristic attribute determines the weight of each characteristic attribute;
According to the characteristic value of each characteristic attribute of first user and corresponding weight, each of first user is obtained
The scoring of characteristic attribute, and the scoring of each characteristic attribute according to first user, determine the scoring knot of first user
Fruit;
Determine the user gradation belonging to first user based on the appraisal result, and to first user recommend with
The corresponding advertisement of identified user gradation.
A kind of advertisement recommendation apparatus, including:
Behavioral data acquiring unit, the behavioral data for obtaining the first user, the behavioral data include and pay
The data information of the relevant at least one characteristic attribute of ability;
Characteristic value determination unit is used for the behavioral data based at least one second user, respectively to first user
Each of the data information of the characteristic attribute handled, obtain the feature of each of first user characteristic attribute
Value;
Weight determining unit, for utilizing each of first user characteristic value of the characteristic attribute and each institute
The characteristic value of each of second user characteristic attribute is stated, determines the weight of each characteristic attribute;
First appraisal result computing unit, for according to the characteristic value of each characteristic attribute of first user and corresponding
Weight obtains the scoring of each characteristic attribute of first user, and the scoring of each characteristic attribute according to first user,
Determine the appraisal result of first user;
First advertisement recommendation unit, for determining the user gradation belonging to first user based on the appraisal result,
And recommend advertisement corresponding with identified user gradation to first user.
A kind of advertisement recommendation server, including memory and processor, the memory is for storing program, the processing
Device calls described program, described program to be used for:
The behavioral data of the first user is obtained, the behavioral data includes and the relevant at least one feature of ability to pay
The data information of attribute;
Based on the behavioral data of at least one second user, respectively to each of first user characteristic attribute
Data information is handled, and the characteristic value of each of first user characteristic attribute is obtained;
Utilize each of the characteristic value of each of first user characteristic attribute and each described second user
The characteristic value of the characteristic attribute determines the weight of each characteristic attribute;
According to the characteristic value of each characteristic attribute of first user and corresponding weight, each of first user is obtained
The scoring of characteristic attribute, and the scoring of each characteristic attribute according to first user, determine the scoring knot of first user
Fruit;
Determine the user gradation belonging to first user based on the appraisal result, and to first user recommend with
The corresponding advertisement of identified user gradation.
The embodiment of the invention discloses a kind of advertisements to recommend method, apparatus and server, by the row for obtaining the first user
For data, the behavioral data includes the data information with the relevant at least one characteristic attribute of ability to pay;Respectively to institute
The data information for stating each of first user characteristic attribute is handled, and obtains each of first user feature
The characteristic value of attribute;Determine the weight of each characteristic attribute;According to the characteristic value of each characteristic attribute of first user
And corresponding weight, determine the appraisal result of first user;It is determined belonging to first user based on the appraisal result
User gradation, and recommend to first user mode of corresponding with identified user gradation advertisement, realize and be based on
It is more prone to assessment of the behavioral data to the ability to pay of user of the user of the ability to pay of reaction user, and then ensure that
Effective popularization of advertisement in computer application.
Description of the drawings
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technology description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
The embodiment of invention for those of ordinary skill in the art without creative efforts, can also basis
The attached drawing of offer obtains other attached drawings.
Fig. 1 is that method flow diagram is recommended in a kind of advertisement provided by the embodiments of the present application;
Fig. 2 is provided by the embodiments of the present application a kind of for each characteristic attribute, to this feature of first user
The data information of this feature attribute of the data information of attribute and the behavioral data of at least one second user unitizes
Processing, determines the characteristic value of this feature attribute of first user, and this feature attribute of each second user
The method flow diagram of characteristic value;
Fig. 3 is that a kind of data information of this feature attribute determining first user provided by the embodiments of the present application indicates
Numerical value method flow diagram;
Fig. 4 is a kind of characteristic value using each of first user characteristic attribute provided by the embodiments of the present application
And the characteristic value of each of each described second user characteristic attribute, determine the side of the weight of each characteristic attribute
Method flow chart;
Fig. 5 is provided by the embodiments of the present application a kind of according to obtained change corresponding with each characteristic attribute respectively
Change value determines the method flow diagram of the weight of each characteristic attribute;
Fig. 6 is a kind of structural schematic diagram of advertisement recommendation apparatus provided by the embodiments of the present application;
Fig. 7 is a kind of detailed construction schematic diagram of characteristic value determination unit provided by the embodiments of the present application;
Fig. 8 is a kind of detailed construction schematic diagram of weight determining unit provided by the embodiments of the present application;
Fig. 9 is a kind of structural schematic diagram of advertisement recommendation server provided by the embodiments of the present application.
Specific implementation mode
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation describes, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
Embodiment:
The embodiment of the present application provides a kind of advertisement recommendation method, is applied to the first computer application, first computer
Using for third-party application.It is only the preferred side that the application of method is recommended in a kind of advertisement provided by the embodiments of the present application above
Formula does not limit herein.
Fig. 1 is that method flow diagram is recommended in a kind of advertisement provided by the embodiments of the present application.
As shown in Figure 1, this method includes:
S101, user data set is obtained, the user data set includes the row of the first user and at least one second user
For data, the behavioral data includes the data information with the relevant at least one characteristic attribute of ability to pay.
Optionally, the behavioral data of each user (the first user/second user) includes the ability to pay with the user
The data information of relevant at least one characteristic attribute.The corresponding characteristic attribute of behavioral data of each user and other each use
The corresponding characteristic attribute of behavioral data at family is identical.If it is determined that the number of the corresponding characteristic attribute of the behavioral data of a user
It is believed that when breath need not obtain, by way of sky/preset value, institute can be indicated by the way that the data information of a characteristic attribute is arranged
The data information for stating a characteristic attribute is not acquired, accordingly in the implementation procedure that method is recommended in subsequent advertisement, it may be determined that
The data information of one characteristic attribute indicates numerical value, and the numerical value is 0.
S102, the behavioral data based at least one second user, respectively to each of first user feature
The data information of attribute is handled, and the characteristic value of each of first user characteristic attribute is obtained;
Optionally, the behavioral data based at least one second user, respectively to each of first user spy
The data information of sign attribute is handled, and the characteristic value of each of first user characteristic attribute is obtained, including:For
Each characteristic attribute, the behavior number of data information and at least one second user to this feature attribute of first user
According to including the data information of this feature attribute carry out unitized processing, obtain the feature of this feature attribute of first user
Value.
Optionally, for each characteristic attribute, data information to this feature attribute of first user and at least
The data information for this feature attribute that the behavioral data of one second user includes carries out unitized processing, obtains described first and uses
The characteristic value of this feature attribute at family, including:For each characteristic attribute, to the number of this feature attribute of first user
It is believed that the data information of this feature attribute of breath and the behavioral data of at least one second user carries out unitized processing, really
The characteristic value of this feature attribute of fixed first user, and each characteristic value of this feature attribute of the second user.
Optionally, provided by the embodiments of the present application a kind of for each characteristic attribute, to the spy of first user
The data information for levying this feature attribute of the data information of attribute and the behavioral data of at least one second user carries out unification
Change is handled, and determines the characteristic value of this feature attribute of first user, and this feature attribute of each second user
Characteristic value method, Fig. 2 please be participate in.
As shown in Fig. 2, this method includes:
S201, it is directed to each characteristic attribute, determines what the data information of this feature attribute of first user indicated
Numerical value;
Optionally, include with the relevant at least one characteristic attribute of ability to pay:Fisrt feature attribute, second feature category
Property, third feature attribute, fourth feature attribute, fifth feature attribute, sixth feature attribute, seventh feature attribute, eighth feature
Attribute, ninth feature attribute, tenth feature attribute and/or the 11st characteristic attribute.
Wherein, the instruction of fisrt feature attribute logged in the first historical time section second computer application number it is most
Hardware device;The instruction of second feature attribute logs in each hardware of the second computer application in the first historical time section
Equipment;Third feature attribute instruction user in the first historical time section is generated each based on second computer application
The total amount of Transaction Information;The instruction of fourth feature attribute is answered using pagerank algorithms and user based on the second computer
The red packet quantity received and dispatched in the first historical time section, the use in the first historical time section being calculated
Importance value of the family in second computer application (importance value can be described as pagerank values);Fifth feature
Attribute instruction user applies the total amount of the red packet sent out in the first historical time section based on the second computer;The
Six characteristic attributes instruction user applies the red packet sent out in the first historical time section to touch based on the second computer
The number of users reached;The seventh feature attribute instruction user is applied based on the second computer in the first historical time section
Difference between the total amount of the red packet inside sent out and the total amount of the red packet of income;Eighth feature attribute indicates that user is based on institute
It states second computer and applies the transnational trip number realized in the first historical time section;Ninth feature attribute indicates user
The trans-city trip number in the country realized in the first historical time section is applied based on the second computer;Tenth feature category
Property instruction user each go on a journey that (trip includes based on what the second computer was applied in the first historical time section
Transnational trip and domestic trans-city trip) average trip distance;The instruction of 11st characteristic attribute is calculated according to described second applies
The cell that the user that the information reported determines lives in the first historical time section.
It is only at least one feature related to ability to pay that behavioral data provided by the embodiments of the present application includes above
The preference information of attribute, inventor can be according to the demand of oneself arbitrarily settings and the relevant at least one characteristic attribute of ability to pay
Particular content, then this does not limit.
Optionally, it is with the data information of the relevant at least one characteristic attribute of ability to pay in the embodiment of the present application
What the acquisition of information reported according to the second computer application arrived.It is third-party application that the second computer, which is applied, described
First computer application is same as the second computer application, is calculated or, first computer application is different from described second
Machine application.When first computer application applies identical with the second computer, first computer application is institute
State the functional unit in second computer application.
Preferably, the instruction of the 11st characteristic attribute calculates the user for the information determination that application reports in institute according to described second
The cell lived in the first historical time section is stated, now to the data information of the 11st characteristic attribute of the behavioral data of the first user
Acquisition modes illustrate:
Receive the information (described information includes target information) for the first user that second computer application reports, the target
Information includes that the first user for reporting of second computer application is based on the second computer in the first historical time section
Using all information of registering registered, the information of registering includes registering position and to register the time.From the target information
Rejecting the time of registering meets the information of registering of preset first time range, and obtains at least one first and registers information;From described
At least one first, which registers, determines that the flow of personnel probability for position of registering is minimum in information and at least one second registers information;From
Second, which registers, determines that a third is registered information in information, and first user is in the first historical time section described the
The number of days that three positions of registering for registering information are registered is most.
Optionally, the calculation of flow of personnel probability of position of registering includes:Obtain what second computer application reported
All users (first user and at least one second user) are in the second historical time section based on described the
All thirds that two computer applications are registered are registered information, and are answered based on the second computer in third historical time section
It is registered information with register all four.
Determine all thirds register information and all 4th register information all positions of registering, for each label
To position, determine that all users corresponding with the position of registering collect as the first user in registering information from all thirds, and from
All four, which register, determines in information all users corresponding with the position of registering as second user collection;Described first is calculated to use
The intersection (quantity of existing identical user between the first user collection and second user collection) of family collection and second user collection,
And calculate the union (total number of users of the first user collection and second user collection that first user gathers second user collection
Amount), and first is obtained as a result, subtracting obtained second knot of first result by 1 using the intersection divided by the union
Flow of personnel probability of the fruit as the position of registering.
Optionally, the second historical time section and third historical time section are two adjacent periods, when such as the second history
Between section be in June, 2016, third historical time section be in July, 2016.
Preferably, the first historical time section includes the second historical time section of institute and third historical time section, if for example, second
Historical time section is in June, 2016, and third historical time section is in July, 2016, and the first historical time section is 1-7 in 2016
Month.
Optionally, the behavioral data for obtaining the first user, including:Determine the primitive behavior data of the first user;It picks
Except the abnormal data in the primitive behavior data, the behavioral data of the first user is obtained.
It should be noted that:What the primitive behavior data of the first determining user received is answered by the second computer
With the information of first user reported.
In order to make it easy to understand, being now explained for characteristic attribute:
For the fisrt feature attribute and second feature attribute, in the primitive behavior data for determining the first user
The first information, the first information include that the first user for reporting of second computer application is every in the first historical time section
Once log in the facility information that the second computer applies used hardware device.By the non-mobile phone of instruction in the first information
The facility information of equipment is rejected as abnormal data from the primitive behavior data.And then according to rejecting abnormalities data after
The first information obtains the data information of fisrt feature attribute and the data information of second feature attribute.
For the third feature attribute, the second information in the primitive behavior data of the first user is determined, it is described
Second information includes each friendship that the first user is generated in the first historical time section based on second computer application
Easy information.Each Transaction Information that transaction amount in second information is unsatisfactory for preset transaction amount threshold value is different as one
Regular data is rejected from the primitive behavior data.And then according to the second acquisition of information third feature after rejecting abnormalities data
The data information of attribute.
For the fourth feature attribute, fifth feature attribute, sixth feature attribute and seventh feature attribute, determine
Third information in the primitive behavior data of first user, the third information include the first user in first historical time
The red packet information of each red packet based on second computer application transmitting-receiving in section.The red packet amount of money sent out is unsatisfactory for presetting
Red packet amount of money threshold value each red packet information as an abnormal data, rejected from the primitive behavior data.
It should be noted that:If the first user is in the first historical time section based on second computer application hair
The total amount of the red packet gone out do not reach it is preset when sending out the red packet amount of money, determine first user fourth feature attribute,
The data information of five characteristic attributes, sixth feature attribute and seventh feature attribute need not obtain.And then according to rejecting abnormalities
The data information of third acquisition of information fourth feature attribute after data, data information, the sixth feature category of fifth feature attribute
The data information of property and the data information of seventh feature attribute.
For eighth feature attribute, ninth feature attribute and tenth feature attribute, the raw line of the first user is determined
For the 4th information in data, the 4th information includes that the first user is based on described second in the first historical time section
The transnational trip data and domestic trans-city trip data that computer application reports.The every of error will be positioned caused by abnormal cause
The transnational trans-city trip data in trip data/country of item is rejected as an abnormal data from the primitive behavior data.In turn
According to the data information of the 4th acquisition of information eighth feature attribute after rejecting abnormalities data, the data of ninth feature attribute are believed
The data information of breath and tenth feature attribute.
For the 11st characteristic attribute, determine the 5th information in the primitive behavior data of the first user (on i.e.
State target information), the 5th information includes the first user for reporting of second computer application in the first historical time section
The interior all information of registering registered based on the second computer application, the information of registering include the position and when registering of registering
Between.It can " rejecting from the target information time of registering meets the information of registering of preset first time range, and obtains by above-mentioned
At least one first registers information " step, regard the process of the rejecting abnormalities data from the 5th information of primitive behavior data as.Into
And according to the data information of the 11st characteristic attribute of the 5th acquisition of information after rejecting abnormalities data.
S202, the numerical value indicated according to the data information of this feature attribute of first user determine that described first uses
Sequence of the family in all users, all users include first user and at least one second user;
Optionally, after the numerical value that the data information for determining this feature attribute of first user indicates, described in determination
The numerical value that the data information of this feature attribute of each second user at least one second user indicates;And it will be identified
The numerical value of all users's (the first user and all second users) sorts according to sequence from big to small, and determines that described first uses
Sequence of the family in all users, and each second user is in the middle sequence of all users.
The sequence tagmeme of S203, the number based on all users and first user in all users, obtains
To the characteristic value of this feature attribute of first user, and each characteristic value of this feature attribute of the second user.
Optionally, the sequence tagmeme according to the number of all users and first user in all users,
Determine the characteristic value of this feature attribute of first user.
Optionally, for each second user, the characteristic value of this feature attribute of the second user is determined
Mode is:According to the sequence tagmeme of the number and the second user of all users in all users, determine this second
The characteristic value of the characteristic attribute of user.
S103, the characteristic value using each of first user characteristic attribute and each second user
The characteristic value of each characteristic attribute determines the weight of each characteristic attribute;
S104, the characteristic value according to each characteristic attribute of first user and corresponding weight obtain described first and use
The scoring of each characteristic attribute at family, and the scoring of each characteristic attribute according to first user, determine first user's
Appraisal result;
S105, the user gradation belonging to first user is determined based on the appraisal result, and to first user
Recommend advertisement corresponding with identified user gradation.
Optionally, it is preset with the user gradation of multiple and different ranks, each user gradation is corresponding with corresponding scoring range;
After the appraisal result for determining the first user, by the user corresponding to the scoring range belonging to the appraisal result of first user
Grade is determined as the user gradation belonging to first user.
Optionally, it is preset with advertisement corresponding with each user gradation respectively, is determining the use belonging to first user
After the grade of family, recommend advertisement corresponding with the first user owning user grade to first user.
Further, in a kind of advertisement recommendation method provided by the embodiments of the present application, further include:Utilize the second user
Each characteristic attribute characteristic value and respective weights, obtain the scoring of each characteristic attribute of the second user, and according to described
The scoring of each characteristic attribute of second user, determines the appraisal result of the second user;Institute is determined based on the appraisal result
The user gradation belonging to second user is stated, and recommends advertisement corresponding with identified user gradation to the second user.
Optionally, at least one second user in a kind of advertisement recommendation method provided by the embodiments of the present application
Each second user can carry out role exchange with first user, so that by described the second of carry out role exchange
User recommends the first user of method in the process of implementation as the advertisement, by the first user conduct of carry out role exchange
The second user of method in the process of implementation is recommended in the advertisement.
The embodiment of the invention discloses a kind of advertisements to recommend method, by obtaining the behavioral data of the first user, the row
Include the data information with the relevant at least one characteristic attribute of ability to pay for data;Respectively to the every of first user
The data information of a characteristic attribute is handled, and the characteristic value of each of first user characteristic attribute is obtained;
Determine the weight of each characteristic attribute;According to the characteristic value of each characteristic attribute of first user and corresponding weight,
Determine the appraisal result of first user;The user gradation belonging to first user is determined based on the appraisal result, and
The mode for recommending advertisement corresponding with identified user gradation to first user realizes and is based on being more prone to react
Assessment of the behavioral data of the user of the ability to pay of user to the ability to pay of user, and then ensure that wide in computer application
The effective popularization accused.
Optionally, the data information table of a kind of this feature attribute determining first user provided by the embodiments of the present application
The method of the numerical value shown refers to Fig. 3.As shown in figure 3, this method includes:
S301, be directed to each characteristic attribute, determine this feature attribute of first user data information whether be
Nonumeric information, if not, executing step S302;If so, executing step S303;
Optionally, be above embodiment provide fisrt feature attribute, second feature attribute, third feature attribute, the 4th
Characteristic attribute, fifth feature attribute, sixth feature attribute, seventh feature attribute, eighth feature attribute, ninth feature attribute,
It is illustrated for ten characteristic attributes and/or the 11st characteristic attribute.The fisrt feature attribute, second feature attribute and the tenth
The data information of one characteristic attribute is nonumeric information;The third feature attribute, fourth feature attribute, fifth feature category
Property, sixth feature attribute, seventh feature attribute, eighth feature attribute, ninth feature attribute and tenth feature attribute numerical value letter
Breath is numerical information.
S302, the data information of this feature attribute of first user is converted to numerical information, and from transformed
Corresponding numerical value is transferred in the data information of this feature attribute of first user;
Optionally, the data information of the fisrt feature attribute is converted to the mode of data information, including:By described
The valence of the most hardware device of number that second computer application is logged in the first historical time section of one characteristic attribute instruction
Lattice are determined as the data information of fisrt feature attribute.
The data information of second feature attribute is converted to the mode of data information, including:Second feature attribute is indicated
In the first historical time section log in second computer application each hardware device average price be determined as second spy
Levy the data information of attribute.
Optionally, the price of hardware device can be captured by way of spiders.It is only the embodiment of the present application above
Crawl hardware device can be arranged in the preferred embodiment of the price of the crawl hardware device of offer, inventor according to the demand personnel of oneself
Price mode, then this does not limit.
Optionally, the data information of the 11st characteristic attribute is converted to the mode of data information, including:By the tenth
It is special that the price for the cell that the user of one characteristic attribute instruction lives in the first historical time section is determined as the described 11st
Levy the data information of attribute.
Optionally, by searching for the side of the price of the preset cell lived in the first historical time section with user
Formula realizes the determination of the price for the cell lived in the first historical time section to user.It is only that the application is real above
The preferred embodiment of the price for the cell that the determination user for applying example offer lives in the first historical time section, inventor can root
The concrete mode of the price for the cell for determining that user lives in the first historical time section is arbitrarily set according to the demand of oneself,
This is not limited again.
S303, corresponding numerical value is transferred from the data information of this feature attribute of first user.
Optionally, corresponding numerical value is transferred from the data information of this feature attribute of first user (that is, transferring institute
State the value of the data information of this feature attribute of the first user), this feature attribute as identified first user
The numerical value that data information indicates.
By above-mentioned to a kind of data letter for this feature attribute determining first user provided by the embodiments of the present application
Cease being further described for the method for the numerical value indicated so that a kind of advertisement recommendation method provided by the embodiments of the present application is more clear
It is clear, complete, be convenient for those skilled in the art understand that.
In order to make it easy to understand, existing to a kind of number based on all users provided by the embodiments of the present application and described
Sequence tagmeme of first user in all users obtains the characteristic value of this feature attribute of first user, and each
The concrete methods of realizing of the characteristic value of this feature attribute of the second user is described in detail.
Optionally, it is preset with characteristic value calculation formulaWherein, the ftotalIndicate all users
Number, the fm(rank)Indicate sequence of the numerical value of the data information expression of the characteristic attribute of user in all users
Tagmeme, the pmIndicate characteristic value (the alternatively referred to as characteristic attribute of the user of the characteristic attribute of the user
Percentile).
It should be noted that:F in the characteristic value calculation formulam(rank)Sequence of the user of expression in all users
Tagmeme is:The sequence of the numerical value indicated according to the data information of characteristic attribute from big to small arrange obtained.
Optionally, the sequence tagmeme according to the number of all users and first user in all users,
The method for determining the characteristic value of this feature attribute of first user, including:The characteristic value calculation formula is called, according to institute
The sequence tagmeme of the number and first user of all users in all users is stated, is calculated first user's
The characteristic value of this feature attribute.
Optionally, for each second user, the characteristic value of this feature attribute of the second user is determined
Mode is:According to the sequence tagmeme of the number and the second user of all users in all users, determine this second
The characteristic value of this feature attribute of user.
In the embodiment of the present application, it is preferred that the number and the second user according to all users is in institute
There is the sequence tagmeme in user, determines the characteristic value of the characteristic attribute of the second user, including:The characteristic value is called to calculate public
Formula, according to the sequence tagmeme of the number and the second user of all users in all users, be calculated this second
The characteristic value of this feature attribute of user.
By above-mentioned to a kind of number according to all users provided by the embodiments of the present application and first use
Sequence tagmeme of the family in all users, determines the characteristic value of this feature attribute of first user, and each described the
The method of the characteristic value of this feature attribute of two users is further described so that a kind of advertisement provided by the embodiments of the present application pushes away
The method of recommending is more clear, completely, be convenient for those skilled in the art understand that.
Optionally, a kind of feature using each of first user characteristic attribute provided by the embodiments of the present application
The characteristic value of each of value and each second user characteristic attribute, determines the weight of each characteristic attribute
Method flow diagram refers to Fig. 4.
As shown in figure 4, this method includes:
S401, for same characteristic attribute first user characteristic value and each second user feature
Value carries out exponential transform, obtains the changing value with same space discrimination;
Optionally, based on the characteristic value of each of the identified first user characteristic attribute and each described second
The characteristic value of each of user characteristic attribute determines the weight of each characteristic attribute, including:For each spy
For levying attribute, this feature attribute of characteristic value and each second user to this feature attribute of first user
Characteristic value carry out exponential transform, obtain the changing value with same space discrimination corresponding with this feature attribute.It needs to note
Meaning, the obtained changing value with same space discrimination corresponding with this feature attribute includes multiple numerical value, institute
Stating multiple numerical value each characteristic value corresponding with this feature attribute, (each characteristic value herein is:This feature category of first user
Property characteristic value and each second user this feature attribute characteristic value) correspond.
Optionally, this feature of the characteristic value to this feature attribute of first user and each second user
The characteristic value of attribute carries out exponential transform, obtains the side of the changing value with same space discrimination corresponding with this feature attribute
Method, including:Call exponential transform formula fnew=efx, wherein the f is characterized value, and the x is characterized coefficient, fnewFor variation
Value;Determine characteristic coefficient corresponding with this feature attribute;Based on the exponential transform formula called, identified feature system is utilized
The feature of this feature attribute of the characteristic value of this feature attribute of several and described first user and each second user
Value, is calculated changing value f with same space discrimination corresponding with this feature attributenew。
In the embodiment of the present application, it is preferred that the corresponding characteristic coefficient of characteristic attribute is set in advance according to actual conditions
It is fixed, for each two characteristic attribute, the corresponding characteristic coefficient of the two characteristic attributes can it is identical can not also be identical,
This is not limited.
S402, according to obtained changing value corresponding with each characteristic attribute respectively, determine each feature
The weight of attribute.
Optionally, provided by the embodiments of the present application a kind of according to obtained corresponding with each characteristic attribute respectively
Changing value determines the method flow diagram of the weight of each characteristic attribute, refers to Fig. 5.
As shown in figure 5, this method includes:
S501, according to obtained changing value corresponding with each characteristic attribute respectively, determine the feature described two-by-two
Related coefficient between attribute;
In the embodiment of the present application, it is preferred that determine the mode of the related coefficient between the characteristic attribute described two-by-two, packet
It includes:Call preset related coefficient calculation formulaWherein, X is characterized the corresponding changing values of attribute X, and Y is spy
The corresponding changing values of attribute Y are levied, co ν (X, Y) indicate the covariance of characteristic attribute X and characteristic attribute Y, σXIt is characterized attribute X's
Standard variance, σYIt is characterized the standard variance of attribute Y, p is characterized the related coefficient between attribute X and characteristic attribute Y;Based on institute
Related coefficient calculation formula is stated, calculates separately to obtain the related coefficient between each two characteristic attribute.
S502, based on the related coefficient between the identified characteristic attribute described two-by-two, determine each characteristic attribute
Weight.
Optionally, based on the related coefficient between the identified characteristic attribute described two-by-two, it may be determined that each characteristic attribute
Related coefficient between other each characteristic attributes respectively, and then it is directed to each characteristic attribute, it calculates and this feature attribute pair
The related coefficient answered and.
Wherein, for each characteristic attribute, calculate related coefficient corresponding with this feature attribute and mode be:Determining should
The characteristic attribute related coefficient between other each characteristic attributes respectively, by the sum of identified related coefficient, as with
The corresponding related coefficient of this feature attribute and.
Correspondingly, when calculate related coefficient corresponding with each characteristic attribute respectively and after, to a calculated institute of institute
Some related coefficients and progress read group total, obtain final related coefficient;It, will be with this feature category for each characteristic attribute
Property corresponding related coefficient and account for the proportion of the final related coefficient, be determined as the weight of this feature attribute.Wherein, with the spy
The corresponding related coefficient of sign attribute and the proportion for accounting for the final related coefficient are:The corresponding related coefficient of this feature attribute is removed
With the final related coefficient, obtained result.
If for example, there are 3 characteristic attributes, respectively characteristic attribute 1, characteristic attribute 2 and characteristic attribute 3, wherein special
It is a to levy the related coefficient between attribute 1 and characteristic attribute 2, and the related coefficient between specific properties 1 and characteristic attribute 3 is b, special
It is c to levy the related coefficient between attribute 2 and characteristic attribute 3.So, related coefficient corresponding with characteristic attribute 1 and be a+b;With
2 corresponding related coefficient of characteristic attribute and be a+c;Related coefficient corresponding with characteristic attribute 3 and be b+c;Final related coefficient
For (a+b)+(a+c)+(b+c);The weight of characteristic attribute 1 isThe weight of characteristic attribute 2 isThe weight of characteristic attribute 3
By above-mentioned to provided by the embodiments of the present application a kind of based on each of the identified first user feature category
Property characteristic value and each of each described second user characteristic attribute characteristic value, determine each characteristic attribute
The method of weight be further described so that a kind of advertisement provided by the embodiments of the present application recommends method to be more clear, completely,
Convenient for those skilled in the art understand that.
Method is described in detail in aforementioned present invention disclosed embodiment, diversified forms can be used for the method for the present invention
Device realize that therefore the invention also discloses a kind of devices, and specific embodiment is given below and is described in detail.
Fig. 6 is a kind of structural schematic diagram of advertisement recommendation apparatus provided by the embodiments of the present application.
As shown in fig. 6, the device includes:
Behavioral data acquiring unit 61, the behavioral data for obtaining the first user, the behavioral data include and prop up
Pay the data information of the relevant at least one characteristic attribute of ability;
Characteristic value determination unit 62 is used for the behavioral data based at least one second user, uses respectively described first
The data information of each of family characteristic attribute is handled, and the spy of each of first user characteristic attribute is obtained
Value indicative;
Weight determining unit 63, for using each of first user characteristic value of the characteristic attribute and each
The characteristic value of each of the second user characteristic attribute determines the weight of each characteristic attribute;
First appraisal result computing unit 64, for according to the characteristic value of each characteristic attribute of first user and corresponding
Weight, obtain the scoring of each characteristic attribute of first user, and commenting according to each characteristic attribute of first user
Point, determine the appraisal result of first user;
First advertisement recommendation unit 65, for determining the user etc. belonging to first user based on the appraisal result
Grade, and recommend advertisement corresponding with identified user gradation to first user.
Further, in a kind of advertisement recommendation apparatus provided by the embodiments of the present application, further include:
Second appraisal result computing unit, the characteristic value for each characteristic attribute using the second user and corresponding power
Weight, obtains the scoring of each characteristic attribute of the second user, and the scoring of each characteristic attribute according to the second user, really
The appraisal result of the fixed second user;
Second advertisement recommendation unit, for determining the user gradation belonging to the second user based on the appraisal result,
And recommend advertisement corresponding with identified user gradation to the second user.
In the embodiment of the present application, it is preferred that the characteristic value determination unit is specifically used for:For each feature category
Property, the spy that the behavioral data of data information and at least one second user to this feature attribute of first user includes
The data information of sign attribute carries out unitized processing, obtains the characteristic value of this feature attribute of first user.
A kind of alternative construction of characteristic value determination unit 62 provided in an embodiment of the present invention refers to Fig. 7.
As shown in fig. 7, characteristic value determination unit 62, including:
First determination unit 71 determines this feature attribute of first user for being directed to each characteristic attribute
The numerical value that data information indicates;
Second determination unit 72, the numerical value for being used to be indicated according to the data information of this feature attribute of first user,
Determine sequence of first user in all users, all users include first user and described at least one
Second user;
Third determination unit 73, for based on all users number and first user in all users
Sequence tagmeme, obtain the characteristic value of this feature attribute of first user.
In the embodiment of the present application, it is preferred that first determination unit 71 is specifically used for:
For each characteristic attribute, if the data information of this feature attribute of first user is nonumeric information,
The data information of this feature attribute of first user is converted into numerical information, and from transformed first user's
Corresponding numerical value is transferred in the data information of this feature attribute;
If the data information of this feature attribute of first user is numerical information, from this feature of first user
Corresponding numerical value is transferred in the data information of attribute.
In the embodiment of the present application, it is preferred that the behavioral data acquiring unit 61, including:
Primitive behavior data determination unit, the primitive behavior data for determining the first user;
Culling unit obtains the behavior of first user for rejecting the abnormal data in the primitive behavior data
Data.
A kind of alternative construction of weight determining unit 63 provided in an embodiment of the present invention refers to Fig. 8.
As shown in figure 8, the weight determining unit 63, including:
Changing value determination unit 81, the characteristic value for first user for same characteristic attribute and each institute
The characteristic value for stating second user carries out exponential transform, obtains the changing value with same space discrimination;
Weight determination subelement 82, for according to obtained changing value corresponding with each characteristic attribute respectively,
Determine the weight of each characteristic attribute.
In the embodiment of the present application, it is preferred that the weight determination subelement 82, including:
Related coefficient determination unit, for according to obtained changing value corresponding with each characteristic attribute respectively,
Determine the related coefficient between the characteristic attribute described two-by-two;
Characteristic attributes weight determination unit is used for based on the related coefficient between the identified characteristic attribute described two-by-two,
Determine the weight of each characteristic attribute.
The embodiment of the present application also provides a kind of advertisement recommendation servers.It is one kind provided by the embodiments of the present application such as Fig. 9
The structural schematic diagram of advertisement recommendation server, advertisement recommendation server include:Processor 91 and memory 92.
Wherein processor 91, memory 92, communication interface 93 complete mutual communication by communication bus 94.
Optionally, communication interface 93 can be the interface of communication module, such as the interface of gsm module.Processor 91, for holding
Line program.
Processor 91 may be a central processor CPU or specific integrated circuit ASIC (Application
Specific Integrated Circuit), or be arranged to implement the integrated electricity of one or more of the embodiment of the present invention
Road.
Memory 92, for storing program.
Program may include program code, and said program code includes computer-managed instruction.In embodiments of the present invention,
Program may include the corresponding program of above-mentioned user interface editing machine.
Memory 92 may include high-speed RAM memory, it is also possible to further include nonvolatile memory (non-volatile
Memory), a for example, at least magnetic disk storage.
Wherein, program can be specifically used for:
The behavioral data of the first user is obtained, the behavioral data includes and the relevant at least one feature of ability to pay
The data information of attribute;
Based on the behavioral data of at least one second user, respectively to each of first user characteristic attribute
Data information is handled, and the characteristic value of each of first user characteristic attribute is obtained;
Utilize each of the characteristic value of each of first user characteristic attribute and each described second user
The characteristic value of the characteristic attribute determines the weight of each characteristic attribute;
According to the characteristic value of each characteristic attribute of first user and corresponding weight, each of first user is obtained
The scoring of characteristic attribute, and the scoring of each characteristic attribute according to first user, determine the scoring knot of first user
Fruit;
Determine the user gradation belonging to first user based on the appraisal result, and to first user recommend with
The corresponding advertisement of identified user gradation.
The embodiment of the invention discloses a kind of advertisement recommendation apparatus and servers, by the behavior number for obtaining the first user
According to the behavioral data includes the data information with the relevant at least one characteristic attribute of ability to pay;Respectively to described
The data information of each of the one user characteristic attribute is handled, and obtains each of first user characteristic attribute
Characteristic value;Determine the weight of each characteristic attribute;According to the characteristic value and phase of each characteristic attribute of first user
The weight answered determines the appraisal result of first user;The use belonging to first user is determined based on the appraisal result
Family grade, and to the mode of first user recommendation advertisement corresponding with identified user gradation, realize based on more
Tend to react assessment of the behavioral data to the ability to pay of user of the user of the ability to pay of user, and then ensure that calculating
Effective popularization of advertisement in machine application.
To sum up:
The embodiment of the invention discloses a kind of advertisements to recommend method, apparatus and server, by the row for obtaining the first user
For data, the behavioral data includes the data information with the relevant at least one characteristic attribute of ability to pay;Respectively to institute
The data information for stating each of first user characteristic attribute is handled, and obtains each of first user feature
The characteristic value of attribute;Determine the weight of each characteristic attribute;According to the characteristic value of each characteristic attribute of first user
And corresponding weight, determine the appraisal result of first user;It is determined belonging to first user based on the appraisal result
User gradation, and recommend to first user mode of corresponding with identified user gradation advertisement, realize and be based on
It is more prone to assessment of the behavioral data to the ability to pay of user of the user of the ability to pay of reaction user, and then ensure that
Effective popularization of advertisement in computer application.
Each embodiment is described by the way of progressive in this specification, the highlights of each of the examples are with other
The difference of embodiment, just to refer each other for identical similar portion between each embodiment.For device disclosed in embodiment
For, since it is corresponded to the methods disclosed in the examples, so description is fairly simple, related place is said referring to method part
It is bright.
Professional further appreciates that, unit described in conjunction with the examples disclosed in the embodiments of the present disclosure
And algorithm steps, can be realized with electronic hardware, computer software, or a combination of the two, in order to clearly demonstrate hardware and
The interchangeability of software generally describes each exemplary composition and step according to function in the above description.These
Function is implemented in hardware or software actually, depends on the specific application and design constraint of technical solution.Profession
Technical staff can use different methods to achieve the described function each specific application, but this realization is not answered
Think beyond the scope of this invention.
The step of method described in conjunction with the examples disclosed in this document or algorithm, can directly be held with hardware, processor
The combination of capable software module or the two is implemented.Software module can be placed in random access memory (RAM), memory, read-only deposit
Reservoir (ROM), electrically programmable ROM, electrically erasable ROM, register, hard disk, moveable magnetic disc, CD-ROM or technology
In any other form of storage medium well known in field.
The foregoing description of the disclosed embodiments enables those skilled in the art to implement or use the present invention.
Various modifications to these embodiments will be apparent to those skilled in the art, as defined herein
General Principle can be realized in other embodiments without departing from the spirit or scope of the present invention.Therefore, of the invention
It is not intended to be limited to the embodiments shown herein, and is to fit to and the principles and novel features disclosed herein phase one
The widest range caused.
Claims (14)
1. method is recommended in a kind of advertisement, which is characterized in that including:
The behavioral data of the first user is obtained, the behavioral data includes and the relevant at least one characteristic attribute of ability to pay
Data information;
Based on the behavioral data of at least one second user, respectively to the data of each of first user characteristic attribute
Information is handled, and the characteristic value of each of first user characteristic attribute is obtained;
It is described using each of the characteristic value of each of first user characteristic attribute and each second user
The characteristic value of characteristic attribute determines the weight of each characteristic attribute;
According to the characteristic value of each characteristic attribute of first user and corresponding weight, each feature of first user is obtained
The scoring of attribute, and the scoring of each characteristic attribute according to first user, determine the appraisal result of first user;
The user gradation belonging to first user is determined based on the appraisal result, and is recommended to first user with institute really
The corresponding advertisement of fixed user gradation.
2. according to the method described in claim 1, it is characterized in that, further including:
Using the characteristic value and respective weights of each characteristic attribute of the second user, each feature category of the second user is obtained
The scoring of property, and the scoring of each characteristic attribute according to the second user, determine the appraisal result of the second user;
The user gradation belonging to the second user is determined based on the appraisal result, and is recommended to the second user with institute really
The corresponding advertisement of fixed user gradation.
3. according to the method described in claim 1, it is characterized in that, the behavioral data based at least one second user,
The data information of each of first user characteristic attribute is handled respectively, obtains each of described first user
The characteristic value of the characteristic attribute, including:
For each characteristic attribute, the data information to this feature attribute of first user and at least one second user
The data information of behavioral data this feature attribute for including carry out unitized processing, obtain this feature category of first user
The characteristic value of property.
4. according to the method described in claim 3, it is characterized in that, described be directed to each characteristic attribute, to first use
The data for this feature attribute that the data information of this feature attribute at family and the behavioral data of at least one second user include are believed
Breath carries out unitized processing, obtains the characteristic value of this feature attribute of first user, including:
For each characteristic attribute, the numerical value that the data information of this feature attribute of first user indicates is determined;
According to the numerical value that the data information of this feature attribute of first user indicates, determine that first user is useful in institute
Sequence in family, all users include first user and at least one second user;
The sequence tagmeme of number and first user in all users based on all users, obtains described first
The characteristic value of this feature attribute of user.
5. according to the method described in claim 4, it is characterized in that, it is described be directed to each characteristic attribute, determine described first
The numerical value that the data information of this feature attribute of user indicates, including:
For each characteristic attribute, if the data information of this feature attribute of first user is nonumeric information, by institute
The data information for stating this feature attribute of the first user is converted to numerical information, and from the spy of transformed first user
It levies in the data information of attribute and transfers corresponding numerical value;
If the data information of this feature attribute of first user is numerical information, from this feature attribute of first user
Data information in transfer corresponding numerical value.
6. according to the method described in claim 5, it is characterized in that, it is described obtain the first user behavioral data, including:
Determine the primitive behavior data of the first user;
The abnormal data in the primitive behavior data is rejected, the behavioral data of first user is obtained.
7. according to the method described in claim 1, it is characterized in that, described utilize each of first user feature category
Property characteristic value and each of each described second user characteristic attribute characteristic value, determine each characteristic attribute
Weight, including:
The characteristic value of characteristic value and each second user for first user of same characteristic attribute refers to
Transformation of variables obtains the changing value with same space discrimination;
According to obtained changing value corresponding with each characteristic attribute respectively, the power of each characteristic attribute is determined
Weight.
8. the method according to the description of claim 7 is characterized in that it is described according to it is obtained respectively with each feature category
Property corresponding changing value, determine the weight of each characteristic attribute, including:
According to obtained changing value corresponding with each characteristic attribute respectively, determine between the characteristic attribute described two-by-two
Related coefficient;
Based on the related coefficient between the identified characteristic attribute described two-by-two, the weight of each characteristic attribute is determined.
9. a kind of advertisement recommendation apparatus, which is characterized in that including:
Behavioral data acquiring unit, for obtaining the behavioral data of the first user, the behavioral data includes and ability to pay
The data information of relevant at least one characteristic attribute;
Characteristic value determination unit is used for the behavioral data based at least one second user, respectively to the every of first user
The data information of a characteristic attribute is handled, and the characteristic value of each of first user characteristic attribute is obtained;
Weight determining unit, for utilizing each of first user characteristic value of the characteristic attribute and each described the
The characteristic value of each of the two users characteristic attribute determines the weight of each characteristic attribute;
First appraisal result computing unit is used for the characteristic value according to each characteristic attribute of first user and corresponding power
Weight, obtains the scoring of each characteristic attribute of first user, and the scoring of each characteristic attribute according to first user, really
The appraisal result of fixed first user;
First advertisement recommendation unit, for determining the user gradation belonging to first user based on the appraisal result, and to
First user recommends advertisement corresponding with identified user gradation.
10. device according to claim 9, which is characterized in that further include:
Second appraisal result computing unit, the characteristic value and respective weights of each characteristic attribute for utilizing the second user,
The scoring of each characteristic attribute of the second user, and the scoring of each characteristic attribute according to the second user are obtained, is determined
The appraisal result of the second user;
Second advertisement recommendation unit, for determining the user gradation belonging to the second user based on the appraisal result, and to
The second user recommends advertisement corresponding with identified user gradation.
11. device according to claim 9, which is characterized in that the characteristic value determination unit is specifically used for:For every
One characteristic attribute, to the behavioral data of the data information and at least one second user of this feature attribute of first user
Including the data information of this feature attribute carry out unitized processing, obtain the feature of this feature attribute of first user
Value.
12. according to the devices described in claim 11, which is characterized in that the characteristic value determination unit, including:
First determination unit determines the data letter of this feature attribute of first user for being directed to each characteristic attribute
Cease the numerical value indicated;
Second determination unit, the numerical value for being used to be indicated according to the data information of this feature attribute of first user, determines institute
Sequence of first user in all users is stated, all users include that first user and described at least one second are used
Family;
Third determination unit is used for the sequence of number and first user in all users based on all users
Tagmeme obtains the characteristic value of this feature attribute of first user.
13. device according to claim 9, which is characterized in that the weight determining unit, including:
Changing value determination unit, for the characteristic value of first user for same characteristic attribute and each described second
The characteristic value of user carries out exponential transform, obtains the changing value with same space discrimination;
Weight determination subelement, for according to obtained changing value corresponding with each characteristic attribute respectively, determining every
The weight of a characteristic attribute.
14. a kind of advertisement recommendation server, which is characterized in that including memory and processor, the memory is for storing journey
Sequence, the processor call described program, described program to be used for:
The behavioral data of the first user is obtained, the behavioral data includes and the relevant at least one characteristic attribute of ability to pay
Data information;
Based on the behavioral data of at least one second user, respectively to the data of each of first user characteristic attribute
Information is handled, and the characteristic value of each of first user characteristic attribute is obtained;
It is described using each of the characteristic value of each of first user characteristic attribute and each second user
The characteristic value of characteristic attribute determines the weight of each characteristic attribute;
According to the characteristic value of each characteristic attribute of first user and corresponding weight, each feature of first user is obtained
The scoring of attribute, and the scoring of each characteristic attribute according to first user, determine the appraisal result of first user;
The user gradation belonging to first user is determined based on the appraisal result, and is recommended to first user with institute really
The corresponding advertisement of fixed user gradation.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710304206.7A CN108804462B (en) | 2017-05-03 | 2017-05-03 | Advertisement recommendation method and device and server |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710304206.7A CN108804462B (en) | 2017-05-03 | 2017-05-03 | Advertisement recommendation method and device and server |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108804462A true CN108804462A (en) | 2018-11-13 |
CN108804462B CN108804462B (en) | 2021-08-27 |
Family
ID=64054398
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710304206.7A Active CN108804462B (en) | 2017-05-03 | 2017-05-03 | Advertisement recommendation method and device and server |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108804462B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111639972A (en) * | 2020-05-29 | 2020-09-08 | 中国联合网络通信集团有限公司 | Multimedia data delivery method and device |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102075573A (en) * | 2011-01-07 | 2011-05-25 | 北京邮电大学 | Method and system for automatically recommending learning partner in distance education social network service (ESNS) |
CN102163310A (en) * | 2010-02-22 | 2011-08-24 | 深圳市腾讯计算机系统有限公司 | Information pushing method and device based on credit rating of user |
US20120136861A1 (en) * | 2010-11-25 | 2012-05-31 | Samsung Electronics Co., Ltd. | Content-providing method and system |
CN102637170A (en) * | 2011-02-10 | 2012-08-15 | 北京百度网讯科技有限公司 | Question pushing method and system |
CN103927673A (en) * | 2014-03-24 | 2014-07-16 | 腾讯科技(深圳)有限公司 | Method and device for recommending service information |
-
2017
- 2017-05-03 CN CN201710304206.7A patent/CN108804462B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102163310A (en) * | 2010-02-22 | 2011-08-24 | 深圳市腾讯计算机系统有限公司 | Information pushing method and device based on credit rating of user |
US20120136861A1 (en) * | 2010-11-25 | 2012-05-31 | Samsung Electronics Co., Ltd. | Content-providing method and system |
CN102075573A (en) * | 2011-01-07 | 2011-05-25 | 北京邮电大学 | Method and system for automatically recommending learning partner in distance education social network service (ESNS) |
CN102637170A (en) * | 2011-02-10 | 2012-08-15 | 北京百度网讯科技有限公司 | Question pushing method and system |
CN103927673A (en) * | 2014-03-24 | 2014-07-16 | 腾讯科技(深圳)有限公司 | Method and device for recommending service information |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111639972A (en) * | 2020-05-29 | 2020-09-08 | 中国联合网络通信集团有限公司 | Multimedia data delivery method and device |
Also Published As
Publication number | Publication date |
---|---|
CN108804462B (en) | 2021-08-27 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106875110A (en) | Operational indicator layered calculation method and device, distributed computing method and system | |
CN108009915A (en) | A kind of labeling method and relevant apparatus of fraudulent user community | |
Connor et al. | Exploring the cost effectiveness of land conservation auctions and payment policies | |
CN106779457A (en) | A kind of rating business credit method and system | |
CN107563633A (en) | A kind of performance indicators examination appraisal procedure, equipment and storage medium | |
CN108133013A (en) | Information processing method, device, computer equipment and storage medium | |
CN110689457A (en) | Intelligent reception method for online clients in real estate industry, server and storage medium | |
CN111681102B (en) | Credit prediction method, apparatus, device and storage medium | |
CN108416630A (en) | A kind of determination method and device of target audience | |
CN107895324A (en) | Insurance examination & verification apparatus and method | |
CN109600344A (en) | Identify the method, apparatus and electronic equipment of risk group | |
CN107886361A (en) | A kind of method and server for assessing ad conversion rates prediction model | |
CN110489691A (en) | Page assembly display methods and terminal device | |
CN105740434B (en) | Network information methods of marking and device | |
CN107993056A (en) | With reference to weekly wage and the emolument delivery system and method for monthly pay | |
CN107767152A (en) | Product purchase intention analysis method and server | |
CN111506876A (en) | Data prediction analysis method, system, equipment and readable storage medium | |
CN108171423A (en) | A kind of administrative examination and approval Effectiveness Evaluation System and method | |
CN110175438A (en) | Share account detection method and relevant device | |
Zeytoonnejad Mousavian et al. | Determinants of foreign direct investment inflow to the agricultural sector: A panel-data analysis | |
Solakoglu et al. | Exchange rate volatility and exports: A firm-level analysis | |
CN110288488A (en) | Medical insurance Fraud Prediction method, apparatus, equipment and readable storage medium storing program for executing | |
CN110264368A (en) | Method and device for generating emotion index of target cryptocurrency | |
CN108804462A (en) | Method, apparatus and server are recommended in a kind of advertisement | |
CN110766441A (en) | Resource object processing method and device, storage medium and computer equipment |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |