CN110532477A - A kind of method and device of information recommendation - Google Patents
A kind of method and device of information recommendation Download PDFInfo
- Publication number
- CN110532477A CN110532477A CN201910826249.0A CN201910826249A CN110532477A CN 110532477 A CN110532477 A CN 110532477A CN 201910826249 A CN201910826249 A CN 201910826249A CN 110532477 A CN110532477 A CN 110532477A
- Authority
- CN
- China
- Prior art keywords
- user
- mentioned
- value
- target
- serial number
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
Abstract
The application proposes that a kind of information recommendation method and device, the above method include: the user's serial number for obtaining target user, and above-mentioned user's serial number is greater than or equal to 0 integer；Obtain the first numerical value corresponding with above-mentioned user's serial number；The user's portrait score for obtaining above-mentioned target user, obtains second value corresponding with above-mentioned user portrait score；Target value in the case where above-mentioned user's serial number is less than first threshold, by the sum of above-mentioned first numerical value and above-mentioned second value, as above-mentioned target user；Above-mentioned target value is exported to above-mentioned target user.Implement the application, the accuracy rate of recommendation information can be improved, to promote network marketing operational effect.
Description
Technical field
This application involves field of computer technology, and in particular to a kind of information recommendation mode and device.
Background technique
With the fast development of Internet technology, shopping at network is increasingly closer in conjunction with people's lives, network battalion
Pin mode is also more and more diversified.And this network marketing is often premised on user completes specified operation, businessman can give successfully
The user for completing specified operation provides certain favor information, and user is allowed to complete to trade with lower price.
In existing network marketing activity, businessman can obtain difficulty by adjusting discount information, control discount amplitude
Size.But as movable user is participated in for, specified operation can only be provided with according to businessman, to get corresponding
Discount information.That is, the discount information that user obtains actually is determined that discount information recommends accuracy rate more by businessman completely
Lowly, the matching degree between user and discount information is low.
Summary of the invention
In view of the above problems, it proposes on the application overcomes the above problem or at least be partially solved in order to provide one kind
State a kind of method and device of information recommendation of problem.
In a first aspect, the embodiment of the present application provides a kind of information recommendation method, the above method includes: to obtain target user
User's serial number, above-mentioned user's serial number be greater than or equal to 0 integer；Obtain the first numerical value corresponding with above-mentioned user's serial number；
The user's portrait score for obtaining above-mentioned target user, obtains second value corresponding with above-mentioned user portrait score；In above-mentioned use
In the case that family serial number is less than first threshold, by the sum of above-mentioned first numerical value and above-mentioned second value, as above-mentioned target user
Target value；Above-mentioned target value is exported to above-mentioned target user.
In one possible implementation, above-mentioned acquisition the first numerical value corresponding with above-mentioned user's serial number, comprising: call
First recommended models handle above-mentioned user's serial number, obtain the first numerical value of above-mentioned target user.
In alternatively possible implementation, above-mentioned acquisition second value corresponding with above-mentioned user portrait score, packet
It includes: the second recommended models being called to handle above-mentioned user portrait score, obtain above-mentioned user's portrait score corresponding second
Numerical value.
In another possible implementation, above-mentioned user draws a portrait score as more than or equal to 0 and less than or equal to 1
Rational；Above-mentioned the second recommended models of calling handle above-mentioned user portrait score, obtain above-mentioned user's portrait score
Corresponding second value, comprising: in the case where above-mentioned user draws a portrait score and directly proportional above-mentioned target value, obtain above-mentioned use
Corresponding first percentile of family portrait score；It calls the second recommended models to handle above-mentioned first percentile, obtains
Second value.
In another possible implementation, above-mentioned user draws a portrait score as more than or equal to 0 and less than or equal to 1
Rational；Above-mentioned the second recommended models of calling handle above-mentioned user portrait score, obtain above-mentioned user's portrait score
Corresponding second value, comprising: in the case where above-mentioned user portrait score and above-mentioned target value are inversely proportional, obtain above-mentioned use
Family portrait score the second percentile corresponding with 1 difference；The second recommended models are called to carry out above-mentioned second percentile
Processing, obtains second value.
In another possible implementation, above-mentioned user's serial number is handled in above-mentioned the first recommended models of calling
Before, the above method further include: obtain the first recommended models, above-mentioned first recommended models are constructed by control parameter, above-mentioned control
Parameter processed includes above-mentioned first threshold, second threshold, fisrt feature numerical value, second feature numerical value and tendency information；Above-mentioned
One threshold value is for indicating minimum participation number；Above-mentioned second threshold is for indicating most participation numbers；Above-mentioned tendency information is used for
The straight slope for determining above-mentioned first recommended models is positive number, alternatively, being negative；Above-mentioned first threshold, above-mentioned second threshold with
And above-mentioned fisrt feature numerical value is used to determine the Linear intercept of above-mentioned first recommended models.
It is above-mentioned by the sum of above-mentioned first numerical value and above-mentioned second value in another possible implementation, as upper
State the target value of target user, comprising: before above-mentioned first numerical value, above-mentioned second value and above-mentioned user's serial number at least
In the case that the sum of accumulated number of one user is more than or equal to above-mentioned fisrt feature numerical value, before above-mentioned user's serial number
The difference of the accumulated number of at least one user and above-mentioned fisrt feature numerical value, the target value as above-mentioned target user；It is no
Then, the target value by the sum of above-mentioned first numerical value and above-mentioned second value, as above-mentioned target user.
In another possible implementation, the above method further include: be greater than or equal in above-mentioned user's serial number above-mentioned
In the case where first threshold, judge whether the corresponding random number of above-mentioned target user is 1；If confirming, above-mentioned target user is corresponding
Random number is 1, and the difference of the accumulated number of at least one user before above-mentioned user's serial number and above-mentioned second feature numerical value is made
For the target value of above-mentioned target user；Otherwise, execution is above-mentioned by the sum of above-mentioned first numerical value and above-mentioned second value, as upper
The step of stating the target value of target user.
Second aspect, the embodiment of the present application provide a kind of information recommending apparatus, comprising: acquiring unit, for obtaining mesh
User's serial number of user is marked, user's serial number is greater than or equal to 0 integer；First obtains unit, for obtain with it is described
Corresponding first numerical value of user's serial number；The first obtains unit is also used to, and obtains user's portrait score of the target user,
Obtain second value corresponding with user portrait score；Recommendation unit, for being less than first threshold in user's serial number
In the case where, the target value by the sum of first numerical value and described second value, as the target user；Output is single
Member, for exporting the target value to the target user.
In one possible implementation, above-mentioned first obtains unit is specifically used for, and calls the first recommended models to upper
It states user's serial number to handle, obtains the first numerical value of above-mentioned target user.
In alternatively possible implementation, above-mentioned first obtains unit is specifically also used to, and calls the second recommended models
Above-mentioned user portrait score is handled, the corresponding second value of above-mentioned user's portrait score is obtained.
In another possible implementation, above-mentioned user draws a portrait score as more than or equal to 0 and less than or equal to 1
Rational；Above-mentioned first obtains unit is specifically also used to, directly proportional in above-mentioned user portrait score and above-mentioned target value
In the case of, obtain corresponding first percentile of above-mentioned user's portrait score；Call the second recommended models to above-mentioned first percentage
Digit is handled, and second value is obtained.
In alternatively possible implementation, above-mentioned user draws a portrait score as more than or equal to 0 and less than or equal to 1
Rational；Above-mentioned first obtains unit is specifically also used to, and is inversely proportional in above-mentioned user portrait score and above-mentioned target value
In the case of, obtain above-mentioned user's portrait score the second percentile corresponding with 1 difference；Call the second recommended models to above-mentioned
Second percentile is handled, and second value is obtained.
In another possible implementation, above-mentioned apparatus further include: the second obtaining unit is recommended for obtaining first
Model, above-mentioned first recommended models by control parameter construct, above-mentioned control parameter include above-mentioned first threshold, second threshold,
Fisrt feature numerical value, second feature numerical value and tendency information；Above-mentioned first threshold is for indicating minimum participation number；Above-mentioned
Two threshold values are for indicating most participation numbers；Above-mentioned tendency information is for determining that the straight slope of above-mentioned first recommended models is positive
Number, alternatively, being negative；Above-mentioned first threshold, above-mentioned second threshold and above-mentioned fisrt feature numerical value are for determining above-mentioned first
The Linear intercept of recommended models.
In another possible implementation, above-mentioned recommendation unit is specifically used for, in above-mentioned first numerical value, above-mentioned second
The sum of accumulated number of at least one user is more than or equal to above-mentioned fisrt feature number before numerical value and above-mentioned user's serial number
In the case where value, by the difference of the accumulated number of at least one user before above-mentioned user's serial number and above-mentioned fisrt feature numerical value,
Target value as above-mentioned target user；
Otherwise, the target value by the sum of above-mentioned first numerical value and above-mentioned second value, as above-mentioned target user.
In another possible implementation, above-mentioned apparatus further include: judging unit, for big in above-mentioned user's serial number
In or be equal to above-mentioned first threshold in the case where, judge whether the corresponding random number of above-mentioned target user is 1；Above-mentioned recommendation unit
It is specifically also used to, if confirming, the corresponding random number of above-mentioned target user is 1, by least one user before above-mentioned user's serial number
The difference of accumulated number and above-mentioned second feature numerical value, the target value as above-mentioned target user；Above-mentioned recommendation unit is specific
It is also used to, otherwise, executes the above-mentioned number of targets by the sum of above-mentioned first numerical value and above-mentioned second value, as above-mentioned target user
The step of value.
The third aspect, the embodiment of the present application provide a kind of information recommending apparatus, comprising: processor, input unit, output
Device and memory, wherein memory is used to store the computer program for supporting server to execute the above method, computer program
Including program instruction, processor is configured for caller instruction, the method for executing above-mentioned first aspect.
Fourth aspect, the embodiment of the present application provide a kind of computer readable storage medium, which is characterized in that the calculating
Machine readable storage medium storing program for executing is stored with computer program, and the computer program is executed by processor to realize described in above-mentioned various aspects
Method.
5th aspect, the embodiment of the present application provides a kind of computer program product comprising program instruction, when it is being counted
When being run on calculation machine, so that computer executes the method as shown in first aspect.
Implement the application, user's portrait score based on target user exports target value to target user, can be improved
Recommendation information accuracy rate, to promote network marketing efficiency of operation effect.
Detailed description of the invention
Technical solution in ord to more clearly illustrate embodiments of the present application or in background technique below will be implemented the application
Attached drawing needed in example or background technique is illustrated.
Fig. 1 is a kind of architecture diagram of information recommendation system provided by the embodiments of the present application；
Fig. 2 is a kind of flow chart of information recommendation provided by the embodiments of the present application；
Fig. 3 is a kind of flow chart of information recommendation method provided by the embodiments of the present application；
Fig. 4 is the flow chart of another information recommendation method provided by the embodiments of the present application；
Fig. 5 is a kind of structural schematic diagram of information recommending apparatus provided by the embodiments of the present application；
Fig. 6 is a kind of structural schematic diagram of the information recommendation entity apparatus of simplification provided by the embodiments of the present application.
Specific embodiment
Below in conjunction with the attached drawing in the embodiment of the present application, the embodiment of the present application is described.
The description and claims of this application and term " first " in the attached drawing, " second " and " third " etc. are
For distinguishing different objects, it is not use to describe a particular order.In addition, " comprising " and " having " and their any deformations,
It is intended to cover and non-exclusive includes.Such as it contains the process, method of a series of steps or units, system, product or sets
It is standby to be not limited to listed step or unit, but optionally further comprising the step of not listing or unit, or optionally
It further include the other step or units intrinsic for these process, methods, product or equipment.
Referenced herein " embodiment " is it is meant that a particular feature, structure, or characteristic described can wrap in conjunction with the embodiments
It is contained at least one embodiment of the application.Each position in the description occur the phrase might not each mean it is identical
Embodiment, nor the independent or alternative embodiment with other embodiments mutual exclusion.Those skilled in the art explicitly and
Implicitly understand, embodiment described herein can be combined with other embodiments.
The term " server " that uses in this application, " unit ", " system " etc. for indicate computer-related entity,
Hardware, firmware, the combination of hardware and software, software or software in execution.For example, server can be but not limited to, processing
Device, data processing platform (DPP) calculate equipment, computer, two or more computers etc..
Referring to Fig. 1, Fig. 1 is a kind of architecture diagram of information recommendation system provided by the embodiments of the present application.As shown in Figure 1,
It include electronic equipment 10 and information recommending apparatus 20 in the system.In one possible implementation, electronic equipment 10 can be with
For portable terminals such as mobile phone, tablet computer, laptops, or the non-portable terminal such as desktop computer, Huo Zhewei
Other can upload the equipment of data.
It should be noted that in information recommendation system shown in Fig. 1: electronic equipment 10 is used for information recommending apparatus
20 provide user's serial number of target user.
Information recommending apparatus 20 for obtaining the corresponding user's serial number of electronic equipment 10, and obtains corresponding with user's serial number
The first numerical value；Information recommending apparatus 20 is also used to obtain user's portrait score of above-mentioned target user, obtains and above-mentioned user
The corresponding second value of portrait score；Information recommending apparatus 20 is also used to the case where above-mentioned user's serial number is less than first threshold
Under, the target value by the sum of above-mentioned first numerical value and above-mentioned second value, as above-mentioned target user；Information recommending apparatus
20, it is also used to export above-mentioned target value to above-mentioned target user.
In alternatively possible implementation, electronic equipment 10 is used to provide target user's to information recommending apparatus 20
User's portrait score.
Referring to Fig. 2, Fig. 2 is a kind of flow chart of information recommendation provided by the embodiments of the present application, information as shown in Figure 2
When information recommending apparatus 20 is drawn a portrait score according to user's serial number and user known to recommended flowsheet, electronic equipment 10 corresponding the is obtained
After one numerical value and second value, need to judge whether the corresponding user's serial number of the user is less than first threshold；If user's sequence
Number it is less than first threshold, the accumulated number of all users before judging the user of information recommending apparatus 20, the first numerical value and the
Whether the sum of two numerical value are greater than fisrt feature numerical value；If the sum of above-mentioned numerical value is less than fisrt feature numerical value, information recommending apparatus
20 target value by the sum of the first numerical value and second value, as the corresponding user of electronic equipment 10；Otherwise, information recommending apparatus
20 by the difference of the accumulated number of all users before the user and fisrt feature numerical value, as the corresponding user's of electronic equipment 10
Target value.
Optionally, information recommending apparatus 20 obtains the mode of user's portrait score, can pass through for information recommending apparatus 20
The user data for receiving the user is calculated；Or receive the electronic equipment hair for being specifically used for calculating user's portrait score
The data sent.
Information recommendation process as shown in Figure 2 it is found that information recommending apparatus 20 be greater than in user's serial number of target user or
Equal to when first threshold, needing further to judge the corresponding random number of the user.To qualified target user,
Information recommending apparatus 20 is by the difference of the accumulated number of all users before the user and second feature numerical value, as electronic equipment
The target value of 10 corresponding users.I.e. for ineligible target user, information recommending apparatus 20 repeats judgement should
Whether the sum of the accumulated number of all users, the first numerical value and second value are greater than the step of fisrt feature numerical value before user
Suddenly, until obtaining the corresponding target value of the target user, after the target value that corresponding user is exported to electronic equipment 10, information
Recommendation apparatus 20 terminates the corresponding information recommendation process of the user.
Referring to Fig. 3, Fig. 3 is a kind of flow diagram of information recommendation method provided by the embodiments of the present application, above-mentioned side
Method includes:
S101, the user's serial number for obtaining target user, above-mentioned user's serial number are greater than or equal to 0 integer.
Specifically, user's serial number of above-mentioned target user is determined by the triggered time of user, the above-mentioned triggered time can be
User clicks the time of specific link, or user enters the time of some system.It is carried out according to the triggered time of user
Sequence, the triggered time earliest corresponding user's serial number 0 of user, the other users that the triggered time is later than the user obtain respectively
User's serial number corresponding with triggered time collating sequence.
For example, the time that user's first enters coupon system of prescribing a time limit is 08:00, the time that user's second enters the system is 09:
30, the triggered time of user is ranked up, user's first and user's second triggered time collating sequence are respectively the 1st and the 30th, then
User's serial number 0 of user's first, user's serial number 29 of user's second.
In another example user's first creates link A, then user's serial number 0 of user's first；User's second and user third are respectively
3 and the 4th are clicked the user of link A, then user's second and third user's serial number are respectively 3 and 4.
Further, information recommending apparatus is stored in advance multiple coding rules, and above- mentioned information recommendation apparatus is according to currently setting
It sets, determines user's serial number of user.
In one possible implementation, above- mentioned information recommendation apparatus according to user triggered time, to user's serial number
Carry out number-of-fragments.
For example, first 100 enter user corresponding user's serial number A000~A099 of system A, the 101st~200th
Into the corresponding user's serial number B000~B099 of user of system A, the 201st~500th user for entering system A is corresponding
User's serial number C000~C299.
In alternatively possible implementation, triggered time and user gradation of the above- mentioned information recommendation apparatus according to user
User's serial number is numbered in information.
For example, user's first and user's second click chain in 12:00:00 after the 100th user clicks link A
A is met, user's second is superior to due to user's first, then user's serial number 100 of user's first, user's serial number 101 of user's second.
In another example user's first and user's second are clicked in 12:00:00 after the 100th user clicks link A
A is linked, since the registion time of user's first is longer, then above- mentioned information recommendation apparatus determines that user's first is superior to user's second,
Then user's serial number 100 of user's first, user's serial number 101 of user's second.
In another possible implementation, above- mentioned information recommendation apparatus carries out triggered time identical user random
Number.It should be understood that above description is intended merely as illustrating, the embodiment of the present application does not make specific limit to the deciding means of user's serial number
It is fixed.
S102, the first numerical value corresponding with above-mentioned user's serial number is obtained.
Specifically, the first recommended models is called to handle above-mentioned user's serial number, the first of above-mentioned target user is obtained
Numerical value.
For example, the slope of above-mentioned first recommended models is 0.01, Linear intercept 0.1, expression formula is G (i)=0.01*i+
0.1, wherein i is user's serial number.User's serial number 10 of above-mentioned target user is then called above-mentioned first recommended models, is obtained
The first numerical value for stating target user is 0.2.
In another example the slope of above-mentioned first recommended models be -0.033, Linear intercept 0.2, expression formula be G (i)=-
0.033*i+0.2, wherein i is user's serial number.User's serial number 0 of above-mentioned target user then calls above-mentioned first to recommend mould
Type, the first numerical value for obtaining above-mentioned target user is 0.2.
In one possible implementation, above-mentioned user's serial number is carried out handling it in above-mentioned the first recommended models of calling
Before, the above method further include: obtain the first recommended models, above-mentioned first recommended models are constructed by control parameter, above-mentioned control
Parameter includes above-mentioned first threshold, second threshold, fisrt feature numerical value, second feature numerical value and tendency information；Above-mentioned first
Threshold value is for indicating minimum participation number；Above-mentioned second threshold is for indicating most participation numbers；Above-mentioned tendency information is for true
The straight slope of fixed above-mentioned first recommended models is positive number, alternatively, being negative；Above-mentioned first threshold, above-mentioned second threshold and
Above-mentioned fisrt feature numerical value is used to determine the Linear intercept of above-mentioned first recommended models.
For example, working as first threshold in control parameter is 3, second threshold 8, fisrt feature numerical value is 0.7, second feature number
Value is 0.9, when above-mentioned tendency information determines that the straight slope of above-mentioned first recommended models is negative, above-mentioned first recommended models
Expression formula is as follows: G (i)=a*i+C, wherein a is the constant less than 0, and i is user's serial number, it is determined that user's serial number M+1
Obtained the first numerical value of user be 0, the first numerical value that the user of user's serial number 0 to M obtains and be above-mentioned fisrt feature
Numerical value.
Above-mentioned fisrt feature numerical value is indicated with 1-b, then the expression formula of above-mentioned first recommended models isThe expression formula of constant C is
Since constant C is about M monotone decreasing, first threshold 3, second threshold 8, fisrt feature numerical value is 0.3, is determined
The corresponding serial number 5 of above-mentioned M.The expression formula that the value of M is substituted into above-mentioned first recommended models, obtains G (i)=- 0.033*i+
0.2, wherein i is user's serial number.
S103, the user's portrait score for obtaining above-mentioned target user, acquisition and above-mentioned user portrait score corresponding second
Numerical value.
Specifically, above-mentioned acquisition second value corresponding with above-mentioned user portrait score, comprising: call the second recommended models
Above-mentioned user portrait score is handled, the corresponding second value of above-mentioned user's portrait score is obtained.
For example, above-mentioned second value is the inverse cumulative distribution function φ of standardized normal distribution-1(x) multiply with standard deviation
Product show that standard normal 97.5% is divided by standardized normal distribution table when standard deviation is 0.23, user's portrait score is 0.975
Digit is 1.96, show that above-mentioned second value is 0.4508.
In one possible implementation, above-mentioned user's portrait score is more than or equal to 0 and less than or equal to 1
Rational；Above-mentioned the second recommended models of calling handle above-mentioned user portrait score, obtain above-mentioned user's portrait score pair
The second value answered, comprising: in the case where above-mentioned user draws a portrait score and directly proportional above-mentioned target value, obtain above-mentioned user
Corresponding first percentile of portrait score；It calls the second recommended models to handle above-mentioned first percentile, obtains
Two numerical value.
Further, when above-mentioned user draws a portrait score for indicating the consuming capacity of user, above-mentioned target value is virtual
When the corresponding rebate value of commodity A, above-mentioned user draw a portrait score and above-mentioned target value it is directly proportional, i.e. the above-mentioned use of some user
Family is drawn a portrait, and score is higher (corresponding consuming capacity is stronger), and the rebate value which obtains is bigger.
For example, the user that standard deviation is 0.23, user's first and user's second draw a portrait score be respectively 0.975 and 0.16 when,
Above-mentioned user's first and the corresponding percentile of second are obtained by standardized normal distribution table respectively, is multiplied with standard deviation, obtains above-mentioned use
Family first and the corresponding second value of user's second are respectively 0.4508 and 0.0368.
In alternatively possible implementation, above-mentioned user draws a portrait score as more than or equal to 0 and less than or equal to 1
Rational；Above-mentioned the second recommended models of calling handle above-mentioned user portrait score, obtain above-mentioned user's portrait score
Corresponding second value, comprising: in the case where above-mentioned user portrait score and above-mentioned target value are inversely proportional, obtain above-mentioned use
Family portrait score the second percentile corresponding with 1 difference；The second recommended models are called to carry out above-mentioned second percentile
Processing, obtains second value.
Further, when above-mentioned user draws a portrait score for indicating the payment interest probabilities marking of user, above-mentioned number of targets
When being worth rebate value corresponding for virtual goods A, above-mentioned user's portrait score and above-mentioned target value are inversely proportional, i.e. some user couple
The above-mentioned user answered draws a portrait, and score is higher (possibility of the user charges is bigger), and the rebate value which obtains is smaller.
For example, when user's portrait score that user's first obtains is 0.76, due to above-mentioned user portrait score and above-mentioned target
Numerical value is inversely proportional, then obtains 0.34 corresponding percentile -0.41 by standardized normal distribution table, be multiplied, obtain with standard deviation 0.23
The corresponding second value of above-mentioned user's first is -0.0943 out.
In another possible implementation, above- mentioned information recommendation apparatus limits the input ginseng of above-mentioned second recommended models
The maximum value and minimum value of above-mentioned user's portrait score is arranged in several ranges, when user's portrait score of some user is greater than
When above-mentioned maximum value, above-mentioned second recommended models is called to handle above-mentioned maximum value；When user's portrait point of some user
When number is less than above-mentioned minimum value, above-mentioned second recommended models is called to handle above-mentioned minimum value.
For example, setting 0.975 for the maximum value of user's portrait score, set the minimum value of user's portrait score to
0.25, when the user of user's first portrait score is 0.98, call above-mentioned second recommended models to above-mentioned maximum value (i.e. 0.975)
It is handled.
In another example when user's first is new user, when obtained user's portrait score is be randomly derived 0.01, i.e., in the use
The user at family draws a portrait in the case that score is less than minimum value, call above-mentioned second recommended models to above-mentioned minimum value (i.e. 0.25) into
Row processing.
S104, above-mentioned user's serial number be less than first threshold in the case where, by above-mentioned first numerical value and above-mentioned second value
The sum of, the target value as above-mentioned target user.
Specifically, at least one user before above-mentioned first numerical value, above-mentioned second value and above-mentioned user's serial number
In the case that the sum of accumulated number is more than or equal to above-mentioned fisrt feature numerical value, by least one use before above-mentioned user's serial number
The difference of the accumulated number at family and above-mentioned fisrt feature numerical value, the target value as above-mentioned target user；Otherwise, by above-mentioned
The sum of one numerical value and above-mentioned second value, the target value as above-mentioned target user.
In one possible implementation, above-mentioned fisrt feature numerical value is the interest concessions space of movable businessman's setting, i.e. institute
The sum of target value for having user to obtain is no more than the fisrt feature numerical value.
For example, above-mentioned target user user's serial number 2 and above-mentioned user's serial number be less than first threshold in the case where,
The accumulated number that the user of user's serial number 0 to 1 obtains be 0.3417, corresponding first numerical value of the user of user's serial number 2 and
The sum of second value is 0.1623, the accumulated number 0.3417 obtained by the user of user's serial number 0 to 1 and above-mentioned first spy
Numerical value 0.7 is levied, the maximum value of the target value of the user of available user's serial number 2 can be 0.3583.Due to 0.1623
Number of targets less than 0.3583, then by the sum of above-mentioned first numerical value and above-mentioned second value 0.1623, as above-mentioned target user
Value.
In another example in user's serial number 5 of above-mentioned target user, and the case where above-mentioned user's serial number is less than first threshold
Under, the accumulated number that the user of user's serial number 0 to 4 obtains is 0.6576, corresponding first numerical value of the user of user's serial number 5
It is 0.0701 with the sum of second value, the accumulated number 0.6576 obtained by the user of user's serial number 0 to 1 and above-mentioned first
Character numerical value 0.7, the maximum value of the target value of the user of available user's serial number 5 can be 0.0424.Due to
0.0701 is greater than 0.0424, i.e., at least one user before above-mentioned first numerical value, above-mentioned second value and above-mentioned user's serial number
The sum of accumulated number be more than or equal to above-mentioned fisrt feature numerical value, then by least one user before above-mentioned user's serial number
The difference 0.0424 of accumulated number and above-mentioned fisrt feature numerical value, the target value as above-mentioned target user.
S105, Xiang Shangshu target user export above-mentioned target value.
In one possible implementation, above- mentioned information recommendation apparatus is believed according to the activity that above-mentioned target user participates in
Breath, after handling above-mentioned target value, Xiang Shangshu target user output treated result.
For example, the activity that above-mentioned target user participates in is activity of knocking down-price, the original cost for article of knocking down-price is 100 yuan, above-mentioned target
The target value that user obtains is 0.5, i.e., the prompt information of " you successfully knock down-price 50 yuan " is exported to the target user.
Further, above- mentioned information recommendation apparatus exports various forms of according to the size of above-mentioned target value to user
Prompt information.
For example, above- mentioned information recommendation apparatus exports animation to above-mentioned target user in the case where target value is more than 0.2
Prompt information.
In another example being greater than 0.1 in target value, and when less than 0.2, above- mentioned information recommendation apparatus is defeated to above-mentioned target user
It applauds out animation；In the case where target value is more than 0.2, above- mentioned information recommendation apparatus exports fireworks to above-mentioned target user and moves
It draws.
In alternatively possible implementation, above- mentioned information recommendation apparatus is in the form of active animation, Xiang Shangshu target
User exports above-mentioned target value.
For example, above- mentioned information recommendation apparatus is in the form of turntable for lucky draw animation when user waits the result of target value
Show that different numerical value, the pointer that Xiang Shangshu target user exports in turntable for lucky draw animation are just directed toward the picture of above-mentioned target value
Face.
In another example above- mentioned information recommendation apparatus, in the form of twin color ball draws a lottery animation, Xiang Shangshu target user shows target
The specific value of numerical value.
Further, before exporting above-mentioned target value to above-mentioned target user, above- mentioned information recommendation apparatus is to above-mentioned
Target user exports action message.
For example, above-mentioned target user participate in be the activity of online game store when, above- mentioned information recommendation apparatus is to above-mentioned
Target user exports virtual game stage property discount information.
In another example in the network shopping mall activity that above-mentioned target user participates in, above- mentioned information recommendation apparatus is to above-mentioned target
User exports the favor information of above-mentioned target user's collecting commodities.It should be understood that the example above is intended merely as illustrating, not to information
The way of output of recommendation apparatus makees specific limit.
User's serial number and user's portrait score in the embodiment of the present application based on user, obtain the corresponding number of targets of user
Value, can be improved the accuracy rate of recommendation information, to promote network marketing operational effect.
Referring to Fig. 4, Fig. 4 is the flow diagram of another information recommendation method provided by the embodiments of the present application, it is above-mentioned
Method includes:
S201, the user's serial number for obtaining target user, above-mentioned user's serial number are greater than or equal to 0 integer.
The specific implementation of the step can refer to the step S101 of Fig. 3 above-described embodiment, and details are not described herein.
S202, the first numerical value corresponding with above-mentioned user's serial number is obtained.
The specific implementation of the step can refer to the step S102 of Fig. 3 above-described embodiment, and details are not described herein.
S203, the user's portrait score for obtaining above-mentioned target user, acquisition and above-mentioned user portrait score corresponding second
Numerical value.
The specific implementation of the step can refer to the step S103 of Fig. 3 above-described embodiment, and details are not described herein.
S204, above-mentioned user's serial number be less than first threshold in the case where, by above-mentioned first numerical value and above-mentioned second value
The sum of, the target value as above-mentioned target user.
The specific implementation of the step can refer to the step S104 of Fig. 3 above-described embodiment, and details are not described herein.
S205, above-mentioned user's serial number be greater than or equal to above-mentioned first threshold in the case where, judge above-mentioned target user couple
Whether the random number answered is 1.
In one possible implementation, above- mentioned information recommendation apparatus is more than above-mentioned first threshold for user's serial number
Target user can provide higher target value.
For example, user's first has 70% probability to obtain above-mentioned letter when user's serial number of user's first is greater than above-mentioned first threshold
Breath recommendation apparatus is generated as 1 random number, and corresponding, user's first has 30% probability to obtain above- mentioned information recommendation apparatus to be generated as
0 random number.
Further, the activity that above- mentioned information recommendation apparatus is participated according to above-mentioned target user, it is 1 that setting, which generates random number,
Probability.
For example, above-mentioned target user's participation is rare game item discount activity, then above- mentioned information recommendation apparatus will give birth to
10% is set as at the probability that random number is 1.
In another example above-mentioned target user participate in be daily necessity popularization activity, then above- mentioned information recommendation apparatus will give birth to
80% is set as at the probability that random number is 1.
In alternatively possible implementation, above- mentioned information recommendation apparatus is according to the user of user portrait score, setting
Generate the probability that random number is 1.
For example, the user of user's serial number 0 draws a portrait, score is more than 0.6, then above- mentioned information recommendation apparatus will generate random number
Be set as 60% for 1 probability, all users that user's serial number is greater than or equal to above-mentioned first threshold obtain random number be 1 it is general
Rate is 60%.
In another example in the case where above-mentioned user's serial number is greater than or equal to above-mentioned first threshold, above- mentioned information recommendation apparatus
If user draws a portrait, score, which is more than 0.7, is judged to the user of these users score of drawing a portrait, by generate random number for 1 it is general
Rate is set as 60%；It otherwise, is at random that the user distributes the probability for generating that random number is 1 in 5%~15% range.
In another example in the case where above-mentioned user's serial number is greater than or equal to above-mentioned first threshold, above- mentioned information recommendation apparatus
According to user draw a portrait score and probability corresponding relationship, obtain some user draw a portrait score it is corresponding generate random number be 1 it is general
Rate.
If S206, confirming that the corresponding random number of above-mentioned target user is 1, by least one user before above-mentioned user's serial number
Accumulated number and above-mentioned second feature numerical value difference, the target value as above-mentioned target user.
In one possible implementation, above- mentioned information recommendation apparatus confirmation user's serial number is more than above-mentioned first threshold
Target user has the qualification for obtaining higher target value.
For example, the activity that above-mentioned target user participates in is activity of knocking down-price, higher mesh is obtained since above-mentioned target user has
The qualification of numerical value is marked, i.e. the user can obtain lower discount.For target user, if there is no the qualification, by
The accumulated number of at least one user is 0.6576 before the user corresponds to user's serial number, and above-mentioned fisrt feature numerical value is 0.7,
The target value that then user obtains is 0.0424；After obtaining the qualification, since the user corresponds at least one before user's serial number
The accumulated number of a user is 0.6576, and above-mentioned second feature numerical value is 0.9, then the target value that the user obtains is
0.2424。
In another example the activity that target user participates in is activity of knocking down-price, since above-mentioned target user can obtain lower folding
Button, then the target user can disposably cut reserve price.For the user, if there is no the qualification, due to the user
The sum of corresponding first numerical value and second value are 0.0771, then the target value that the user obtains is 0.0771；Obtain the money
After lattice, the accumulated number of at least one user is 0.5805 before corresponding to user's serial number due to the user, above-mentioned fisrt feature number
Value is 0.7, then the target value that the user obtains is 0.1195.
In alternatively possible implementation, above-mentioned target user there is no the qualification of higher target value, above-mentioned
Target value of the information recommending apparatus by the sum of the first numerical value of the target user and second value, as above-mentioned target user.
Specifically, if confirming, the corresponding random number of above-mentioned target user is not 1, is executed above-mentioned by above-mentioned first numerical value and upper
The sum of second value is stated, the step of target value as above-mentioned target user.
S207, Xiang Shangshu target user export above-mentioned target value.
The specific implementation of the step can refer to the step S105 of Fig. 3 above-described embodiment, and details are not described herein.
User's serial number and user's portrait score in the embodiment of the present application based on user, by having to each user
The marking of discrimination, to obtain the corresponding target value of user.The application can be improved the accuracy rate of recommendation information, increase simultaneously
Add the interest during recommendation information, to promote network marketing operational effect.
Referring to Fig. 5, Fig. 5 is a kind of structural schematic diagram of information recommending apparatus provided by the embodiments of the present application, such as Fig. 5 institute
The information recommending apparatus shown, it may include: acquiring unit 301, first obtains unit 302, recommendation unit 303 and output unit 304；
Optionally, above- mentioned information recommendation apparatus further include: the second obtaining unit 305；Optionally, above- mentioned information recommendation apparatus further include:
Judging unit 306.
Acquiring unit 301, for obtaining user's serial number of target user, user's serial number is whole more than or equal to 0
Number；
First obtains unit 302, for obtaining the first numerical value corresponding with user's serial number；
The first obtains unit 302 is also used to, and is obtained user's portrait score of the target user, is obtained and the use
The corresponding second value of family portrait score；
Recommendation unit 303 is used in the case where user's serial number is less than first threshold, by first numerical value and institute
The sum of second value is stated, the target value as the target user；
Output unit 304, for exporting the target value to the target user.
In one possible implementation, above-mentioned first obtains unit 302 is specifically used for, and calls the first recommended models pair
Above-mentioned user's serial number is handled, and obtains the first numerical value of above-mentioned target user.
In alternatively possible implementation, above-mentioned first obtains unit 302 is specifically also used to, and calls second to recommend mould
Type handles above-mentioned user portrait score, obtains the corresponding second value of above-mentioned user's portrait score.
In another possible implementation, above-mentioned user draws a portrait score as more than or equal to 0 and less than or equal to 1
Rational；Above-mentioned first obtains unit 302 is specifically also used to, directly proportional in above-mentioned user portrait score and above-mentioned target value
In the case where, obtain corresponding first percentile of above-mentioned user's portrait score；Call the second recommended models to the above-mentioned 100th
Quantile is handled, and second value is obtained.
In another possible implementation, above-mentioned user draws a portrait score as more than or equal to 0 and less than or equal to 1
Rational；Above-mentioned first obtains unit 302 is specifically also used to, and is inversely proportional in above-mentioned user portrait score and above-mentioned target value
In the case where, obtain above-mentioned user's portrait score the second percentile corresponding with 1 difference；Call the second recommended models to upper
It states the second percentile to be handled, obtains second value.
In another possible implementation, above-mentioned apparatus further include: the second obtaining unit 305, for obtaining first
Recommended models, above-mentioned first recommended models are constructed by control parameter, and above-mentioned control parameter includes above-mentioned first threshold, the second threshold
Value, fisrt feature numerical value, second feature numerical value and tendency information；Above-mentioned first threshold is for indicating minimum participation number；On
Second threshold is stated for indicating most participation numbers；Above-mentioned tendency information is used to determine the straight slope of above-mentioned first recommended models
For positive number, alternatively, being negative；Above-mentioned first threshold, above-mentioned second threshold and above-mentioned fisrt feature numerical value are above-mentioned for determining
The Linear intercept of first recommended models.
In another possible implementation, above-mentioned recommendation unit 303 is specifically used for, in above-mentioned first numerical value, above-mentioned
It is special to be more than or equal to above-mentioned first for the sum of accumulated number of at least one user before second value and above-mentioned user's serial number
In the case where levying numerical value, by the difference of the accumulated number of at least one user before above-mentioned user's serial number and above-mentioned fisrt feature numerical value
Value, the target value as above-mentioned target user；Otherwise, by the sum of above-mentioned first numerical value and above-mentioned second value, as above-mentioned
The target value of target user.
In another possible implementation, above-mentioned apparatus further include:
Judging unit 306, for judging above-mentioned in the case where above-mentioned user's serial number is greater than or equal to above-mentioned first threshold
Whether the corresponding random number of target user is 1；
Above-mentioned recommendation unit 303 is specifically also used to, if confirming, the corresponding random number of above-mentioned target user is 1, by above-mentioned use
The difference of the accumulated number of at least one user and above-mentioned second feature numerical value before the serial number of family, the mesh as above-mentioned target user
Mark numerical value；
Above-mentioned recommendation unit 303 is specifically also used to, otherwise, execute it is above-mentioned by above-mentioned first numerical value and above-mentioned second value it
The step of with, target value as above-mentioned target user.
Related above-mentioned acquiring unit 301, first obtains unit 302, recommendation unit 303, output unit 304, second obtain
Unit 305 and the more detailed description of judging unit 306 can be referred to directly to be believed in embodiment of the method described in above-mentioned Fig. 3 and Fig. 4
The associated description of breath recommended method directly obtains, and is not added repeats here.
According to a kind of information recommending apparatus provided by the embodiments of the present application, user's serial number and user's portrait based on user
Score, by the marking for carrying out having discrimination to each user, to obtain the corresponding target value of user.The application can mention
The accuracy rate of high recommendation information increases the interest during recommendation information, to promote network marketing operational effect.
Referring to Fig. 6, Fig. 6 is a kind of entity apparatus structural representation of information recommending apparatus provided by the embodiments of the present application
Figure.Information recommending apparatus in the present embodiment as shown in FIG. 6 may include: processor 401, input unit 402, output device
403 and memory 404.It can be by total between above-mentioned processor 401, input unit 402, output device 403 and memory 404
Line is connected with each other.
Memory includes but is not limited to be read-only memory (read-only memory, ROM) or can store static information
With the other kinds of static storage device of instruction, random access memory (random access memory, RAM) or can
The other kinds of dynamic memory for storing information and instruction, is also possible to Electrically Erasable Programmable Read-Only Memory
(Electrically Erasable Programmable Read-Only Memory, EEPROM), CD-ROM (Compact
Disc Read-Only Memory, CD-ROM) or other optical disc storages, optical disc storage (including compression optical disc, laser disc, light
Dish, Digital Versatile Disc, Blu-ray Disc etc.), magnetic disk storage medium or other magnetic storage apparatus or can be used in carry or
Store have instruction or data structure form desired program code and can by any other medium of computer access, but
It is without being limited thereto.Memory, which can be, to be individually present, and is connected by bus with processor.Memory can also be integrated with processor
Together.
Processor is referred to as processing component, and processing unit handles veneer, processing module, processing unit etc..Processor
It can be central processing unit (central processing unit, CPU), network processing unit (network processor,
) or the combination of CPU and NP NP.Processor may include be one or more processors, for example including one or more centers
Processor, in the case where processor is a CPU, which can be monokaryon CPU, be also possible to multi-core CPU.
Memory is used for the program code and data of storage networking device.
Input unit is used for output data and/or signal for input data and/or signal and output device.Output
Device and input unit can be independent device, be also possible to the device of an entirety.
Processor is used to call the program code and data in the memory, executes following steps: obtaining target user's
User's serial number, above-mentioned user's serial number are greater than or equal to 0 integer；Obtain the first numerical value corresponding with above-mentioned user's serial number；It obtains
The user's portrait score for taking above-mentioned target user, obtains second value corresponding with above-mentioned user portrait score；In above-mentioned user
In the case that serial number is less than first threshold, by the sum of above-mentioned first numerical value and above-mentioned second value, as above-mentioned target user's
Target value；It controls output device and exports above-mentioned target value to above-mentioned target user.
In one possible implementation, above-mentioned processor executes above-mentioned acquisition and above-mentioned user's serial number corresponding first
The step of numerical value, comprising: call the first recommended models to handle above-mentioned user's serial number, obtain the first of above-mentioned target user
Numerical value.
In alternatively possible implementation, it is corresponding with above-mentioned user portrait score that above-mentioned processor executes above-mentioned acquisition
Second value the step of, comprising: call the second recommended models to above-mentioned user draw a portrait score handle, obtain above-mentioned user
The corresponding second value of portrait score.
In another possible implementation, above-mentioned user draws a portrait score as more than or equal to 0 and less than or equal to 1
Rational；Above-mentioned processor executes above-mentioned the second recommended models of calling and handles above-mentioned user portrait score, obtains
State user draw a portrait score corresponding second value the step of, comprising: draw a portrait score and above-mentioned target value at just in above-mentioned user
Than in the case where, corresponding first percentile of above-mentioned user's portrait score is obtained；Call the second recommended models to above-mentioned first
Percentile is handled, and second value is obtained.
In another possible implementation, above-mentioned user draws a portrait score as more than or equal to 0 and less than or equal to 1
Rational；Above-mentioned processor executes above-mentioned the second recommended models of calling and handles above-mentioned user portrait score, obtains
State user draw a portrait score corresponding second value the step of, comprising: draw a portrait score and above-mentioned target value at anti-in above-mentioned user
Than in the case where, above-mentioned user's portrait score the second percentile corresponding with 1 difference is obtained；Call the second recommended models pair
Above-mentioned second percentile is handled, and second value is obtained.
In another possible implementation, above-mentioned processor is executing above-mentioned the first recommended models of calling to above-mentioned use
Before the step of family serial number is handled, above-mentioned processor is also used to execute following steps: obtain the first recommended models, above-mentioned the
One recommended models are constructed by control parameter, and above-mentioned control parameter includes above-mentioned first threshold, second threshold, fisrt feature number
Value, second feature numerical value and tendency information；Above-mentioned first threshold is for indicating minimum participation number；Above-mentioned second threshold is used for
It indicates at most to participate in number；Above-mentioned tendency information is used to determine that the straight slope of above-mentioned first recommended models to be positive number, alternatively, being
Negative；Above-mentioned first threshold, above-mentioned second threshold and above-mentioned fisrt feature numerical value are for determining above-mentioned first recommended models
Linear intercept.
In another possible implementation, above-mentioned processor execution is above-mentioned to count above-mentioned first numerical value with above-mentioned second
The step of the sum of value, target value as above-mentioned target user, comprising: above-mentioned first numerical value, above-mentioned second value and
The sum of accumulated number of at least one user is more than or equal to the case where above-mentioned fisrt feature numerical value before above-mentioned user's serial number
Under, by the difference of the accumulated number of at least one user before above-mentioned user's serial number and above-mentioned fisrt feature numerical value, as above-mentioned
The target value of target user；Otherwise, the mesh by the sum of above-mentioned first numerical value and above-mentioned second value, as above-mentioned target user
Mark numerical value.
In another possible implementation, above-mentioned processor is also used to execute following steps, comprising: in above-mentioned user
In the case that serial number is greater than or equal to above-mentioned first threshold, judge whether the corresponding random number of above-mentioned target user is 1；If confirmation
The corresponding random number of above-mentioned target user is 1, by the accumulated number of at least one user before above-mentioned user's serial number and above-mentioned the
The difference of two character numerical values, the target value as above-mentioned target user；Otherwise, it executes above-mentioned by above-mentioned first numerical value and above-mentioned
The step of the sum of second value, target value as above-mentioned target user.
It is designed it is understood that Fig. 6 illustrate only simplifying for information recommending apparatus.In practical applications, information pushes away
Necessary other elements can also be separately included by recommending device, including but not limited to any number of network interface, input unit, defeated
Device, processor, memory etc. out, and all computing platforms that the embodiment of the present application may be implemented are all in the protection model of the application
Within enclosing.
In this application, the unit as illustrated by the separation member may or may not be physically separate
, component shown as a unit may or may not be physical unit, it can and it is in one place, or can also
To be distributed over a plurality of network elements.Some or all of unit therein can be selected to realize this Shen according to the actual needs
Please example scheme purpose.
In addition, each functional unit in each embodiment of the application, which can integrate, is also possible to each group in a component
Part physically exists alone, and is also possible to two or more components and is integrated in a component.Above-mentioned integrated component both may be used
To use formal implementation of hardware, can also realize in the form of software functional units.
If the integrated component is realized in the form of SFU software functional unit and sells or use as independent product
When, it can store in a computer readable storage medium.Based on this understanding, technical solution of the present invention is substantially
The all or part of the part that contributes to existing technology or the technical solution can be in the form of software products in other words
It embodies, which is stored in a storage medium, including some instructions are used so that a computer
Equipment (can be personal computer, server or the network equipment etc.) executes the complete of each the embodiment of the present application the method
Portion or part steps.And storage medium above-mentioned include: USB flash disk, mobile hard disk, read-only memory (read-only memory,
ROM), random access memory (random access memory, RAM), magnetic or disk etc. are various can store program
The medium of code.
The above, the only specific embodiment of the application, but the protection scope of the application is not limited thereto, it is any
Those familiar with the art in the technical scope disclosed by the present invention, can readily occur in various equivalent modifications or replace
It changes, these modifications or substitutions should be covered by the protection scope of the present invention.Therefore, the protection scope of the application should be with right
It is required that protection scope subject to.
It should be understood that magnitude of the sequence numbers of the above procedures are not meant to execute suitable in the various embodiments of the application
Sequence it is successive, the execution of each process sequence should be determined by its function and internal logic, the implementation without coping with the embodiment of the present application
Process constitutes any restriction.Although the application is described in conjunction with each embodiment herein, however, being protected required by embodiment
During the application of shield, those skilled in the art are appreciated that and realize other variations of open embodiment.
Claims (10)
1. a kind of information recommendation method characterized by comprising
User's serial number of target user is obtained, user's serial number is greater than or equal to 0 integer；
Obtain the first numerical value corresponding with user's serial number；
The user's portrait score for obtaining the target user, obtains second value corresponding with user portrait score；
In the case where user's serial number is less than first threshold, by the sum of first numerical value and described second value, as
The target value of the target user；
The target value is exported to the target user.
2. the method according to claim 1, wherein described obtain the first number corresponding with user's serial number
Value, comprising:
It calls the first recommended models to handle user's serial number, obtains the first numerical value of the target user；
It is described to obtain second value corresponding with user portrait score, comprising:
It calls the second recommended models to handle user portrait score, obtains user's portrait score corresponding second
Numerical value.
3. according to the method described in claim 2, score is more than or equal to 0 and to be less than it is characterized in that, the user draws a portrait
Or the rational equal to 1；
The second recommended models of the calling handle user portrait score, and it is corresponding to obtain user's portrait score
Second value, comprising:
In the case where the user draws a portrait score and the directly proportional target value, it is corresponding to obtain user's portrait score
First percentile；
It calls the second recommended models to handle first percentile, obtains second value.
4. according to the method described in claim 2, score is more than or equal to 0 and to be less than it is characterized in that, the user draws a portrait
Or the rational equal to 1；
The second recommended models of the calling handle user portrait score, and it is corresponding to obtain user's portrait score
Second value, comprising:
The user draw a portrait score and the target value be inversely proportional in the case where, obtain the user and draw a portrait score and 1
Corresponding second percentile of difference；
It calls the second recommended models to handle second percentile, obtains second value.
5. according to the described in any item methods of claim 2 to 4, which is characterized in that in the first recommended models of the calling to institute
It states before user's serial number handles, the method also includes:
Obtain the first recommended models, first recommended models are constructed by control parameter, and the control parameter includes described the
One threshold value, second threshold, fisrt feature numerical value, second feature numerical value and tendency information；
The first threshold is for indicating minimum participation number；
The second threshold is for indicating most participation numbers；
The tendency information is used to determine that the straight slope of first recommended models to be positive number, alternatively, being negative；
The first threshold, the second threshold and the fisrt feature numerical value are for determining the straight of first recommended models
Line intercept.
6. according to the method described in claim 5, it is characterized in that, it is described by first numerical value and the second value it
With target value as the target user, comprising:
The sum of the accumulated number of at least one user before first numerical value, the second value and user's serial number
In the case where more than or equal to the fisrt feature numerical value, by the accumulated number of at least one user before user's serial number
With the difference of the fisrt feature numerical value, target value as the target user；
Otherwise, the target value by the sum of first numerical value and described second value, as the target user.
7. according to the method described in claim 5, it is characterized in that, the method also includes:
In the case where user's serial number is greater than or equal to the first threshold, the corresponding random number of the target user is judged
It whether is 1；
If confirming, the corresponding random number of the target user is 1, by the accumulative total of at least one user before user's serial number
The difference of value and the second feature numerical value, the target value as the target user；
Otherwise, the target value by the sum of first numerical value and described second value, as the target user is executed
The step of.
8. a kind of information recommending apparatus characterized by comprising
Acquiring unit, for obtaining user's serial number of target user, user's serial number is greater than or equal to 0 integer；
First obtains unit, for obtaining the first numerical value corresponding with user's serial number；
The first obtains unit is also used to, and is obtained user's portrait score of the target user, is obtained and draw a portrait with the user
The corresponding second value of score；
Recommendation unit is used in the case where user's serial number is less than first threshold, by first numerical value and described second
The sum of numerical value, the target value as the target user；
Output unit, for exporting the target value to the target user.
9. a kind of information recommending apparatus characterized by comprising processor, input unit, output device and memory, wherein
The memory is for storing computer program, and the computer program includes program instruction, and the processor is configured for
Described program instruction is called, method as described in any one of claim 1 to 7 is executed.
10. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the computer program
The method as described in any one of claims 1 to 7 is realized when being executed by processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910826249.0A CN110532477A (en) | 2019-08-30 | 2019-08-30 | A kind of method and device of information recommendation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910826249.0A CN110532477A (en) | 2019-08-30 | 2019-08-30 | A kind of method and device of information recommendation |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110532477A true CN110532477A (en) | 2019-12-03 |
Family
ID=68666299
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910826249.0A Pending CN110532477A (en) | 2019-08-30 | 2019-08-30 | A kind of method and device of information recommendation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110532477A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112348587A (en) * | 2020-11-16 | 2021-02-09 | 脸萌有限公司 | Information pushing method and device and electronic equipment |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20160143111A (en) * | 2015-06-04 | 2016-12-14 | 장은미 | System and method for reward advertisement based on competition of watching advertisement |
| CN107277178A (en) * | 2017-08-07 | 2017-10-20 | 百度在线网络技术（北京）有限公司 | Method and apparatus for pushed information |
| CN107818479A (en) * | 2017-10-27 | 2018-03-20 | 携程旅游网络技术（上海）有限公司 | Favor information method for pushing, system, storage medium and electronic equipment |
-
2019
- 2019-08-30 CN CN201910826249.0A patent/CN110532477A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20160143111A (en) * | 2015-06-04 | 2016-12-14 | 장은미 | System and method for reward advertisement based on competition of watching advertisement |
| CN107277178A (en) * | 2017-08-07 | 2017-10-20 | 百度在线网络技术（北京）有限公司 | Method and apparatus for pushed information |
| CN107818479A (en) * | 2017-10-27 | 2018-03-20 | 携程旅游网络技术（上海）有限公司 | Favor information method for pushing, system, storage medium and electronic equipment |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112348587A (en) * | 2020-11-16 | 2021-02-09 | 脸萌有限公司 | Information pushing method and device and electronic equipment |
| CN112348587B (en) * | 2020-11-16 | 2024-04-23 | 脸萌有限公司 | Information pushing method and device and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| TWI690884B (en) | Abnormal transfer detection method, device, storage medium, electronic equipment and products | |
| CN107547759B (en) | Quality inspection method and device for customer service staff call | |
| CN111553779A (en) | Method and device for sorting bid winning candidates, bid inviting terminal and storage medium | |
| CN112035717A (en) | Information processing method, device, equipment and storage medium | |
| CN112434188A (en) | Data integration method and device for heterogeneous database and storage medium | |
| CN109472615A (en) | Service processing method, device and equipment and readable medium | |
| CN110298508A (en) | Behavior prediction method, device and equipment | |
| CN110532477A (en) | A kind of method and device of information recommendation | |
| CN103309885A (en) | Method and device for identifying feature user in electronic trading platform, search method and device | |
| CN112650921A (en) | Object recommendation method, device, equipment and storage medium | |
| CN110929285B (en) | Method and device for processing private data | |
| CN111311276B (en) | Identification method and device for abnormal user group and readable storage medium | |
| CN109450963B (en) | Message pushing method and terminal equipment | |
| CN107679980A (en) | The determination method and apparatus of user credit | |
| CN111612616A (en) | Block chain account evaluation method and device, terminal device and computer readable medium | |
| CN111415197A (en) | Method for acquiring article exchange voucher, method for providing article exchange voucher, device and storage medium | |
| CN110688582B (en) | Application recommendation method, application recommendation device and terminal equipment | |
| CN112836971A (en) | Quota resource determination method and device, electronic equipment and storage medium | |
| CN112765449A (en) | Application program display method and device and storage medium | |
| CN111881148A (en) | Object combination attribute determination method and device and electronic equipment | |
| CN111124918A (en) | Test data prediction method and device and processing equipment | |
| CN111311372A (en) | User identification method and device | |
| CN106487836B (en) | Information processing method and server | |
| CN111242656A (en) | User credit evaluation method and system based on telecommunication big data | |
| CN110209690B (en) | Data acquisition method and device and electronic 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 |