CN109034892A

CN109034892A - Resource allocation methods, device, computer readable storage medium and computer equipment

Info

Publication number: CN109034892A
Application number: CN201810794502.4A
Authority: CN
Inventors: 曾嵘; 雷茂锋; 郭中明; 陈圣鹏; 徐沛
Original assignee: Tencent Technology Shenzhen Co Ltd
Current assignee: Tencent Technology Shenzhen Co Ltd
Priority date: 2018-07-19
Filing date: 2018-07-19
Publication date: 2018-12-18
Anticipated expiration: 2038-07-19
Also published as: CN109034892B

Abstract

This application involves a kind of resource allocation methods, device, computer readable storage medium and computer equipments, which comprises obtains the sharing rate of account to be chosen；The sharing attribute score value of the account to be chosen is determined according to the sharing rate；Target account is chosen according to the sharing attribute score value；Determine that the target account is directed to the allocation probability of each resource to be allocated according to the sharing attribute score value of the target account.This distribution method, it can be stronger for social attribute, share the allocation probability of the higher account setup of attribute score value resource to be allocated, it can enable this certain customers after getting resource to be allocated with this, actively activity is shared, movable propaganda strength is improved, promotion costs can also be saved by being not only able to promote movable propaganda strength.

Description

Resource allocation methods, device, computer readable storage medium and computer equipment

Technical field

This application involves field of computer technology, more particularly to a kind of resource allocation methods, device, computer-readable deposit Storage media and computer equipment.

Background technique

With the development of network technology, more and more users carry out social activity using network.User can also pass through Network gets more resources, for example user can pass through by Online Shopping commodity or when paying correlative charges Participate in sweepstake.Such as exempt from single reward.The activity side of holding can also release some activities for user by network promotion product It participates in.

In the conventional technology, the probability of the available movable prize of user is random, under this mode, for activity side Speech, movable promotion efficiency not can guarantee, and increase popularization cost, the efficiency of popularization is relatively low.

Summary of the invention

Based on this, it is necessary to which for the above technical issues, providing one kind can reduce popularization cost, improves and promote efficiency Resource allocation methods, device, computer readable storage medium and computer equipment.

A kind of resource allocation methods, comprising:

Obtain the sharing rate of account to be chosen；

The sharing attribute score value of the account to be chosen is determined according to the sharing rate；

Target account is chosen according to the sharing attribute score value；

Determine the target account for point of each resource to be allocated according to the sharing attribute score value of the target account With probability.

A kind of resource allocation device, described device include:

Sharing rate obtains module, for obtaining the sharing rate of account to be chosen；

Share attribute score value determining module, for determining the sharing attribute point of the account to be chosen according to the sharing rate Value；

Target account chooses module, for choosing target account according to the sharing attribute score value；

Allocation probability determining module, for determining the target account needle according to the sharing attribute score value of the target account To the allocation probability of each resource to be allocated.

A kind of computer equipment can be run on a memory and on a processor including memory, processor and storage Computer program, the processor perform the steps of when executing the computer program

Obtain the sharing rate of account to be chosen；

Target account is chosen according to the sharing attribute score value；

A kind of computer readable storage medium, is stored thereon with computer program, and the computer program is held by processor It is performed the steps of when row

Obtain the sharing rate of account to be chosen；

Target account is chosen according to the sharing attribute score value；

Above-mentioned resource allocation methods, device, computer readable storage medium and computer equipment, by getting wait choose After the sharing rate of account, the corresponding sharing attribute score value of account can be determined according to the sharing rate of each account, it can be according to each The sharing attribute score value of account determines allocation probability corresponding to each account for each resource to be allocated.This distribution side Method, can be stronger for social attribute, shares the allocation probability of the higher account setup of attribute score value resource to be allocated, can with this So that this certain customers actively can share activity after getting resource to be allocated, movable publicity power is improved Degree, promotion costs can also be saved by being not only able to promote movable propaganda strength.

Detailed description of the invention

Fig. 1 is the applied environment figure of resource allocation methods in one embodiment；

Fig. 2 is the flow diagram of resource allocation methods in one embodiment；

Fig. 3 is in one embodiment according to the flow diagram for sharing attribute score value selection target account step；

Fig. 4 is the flow diagram that the sharing rate step of account to be chosen is obtained in one embodiment；

Fig. 5 be in one embodiment when the sharing attribute score value of target account is higher, it is determined that target account is default Resource to be allocated the higher step of allocation probability flow diagram；

Fig. 6 is the flow diagram of resource allocation methods in another embodiment；

Fig. 7 is the interface schematic diagram that account gets resource to be allocated in one embodiment；

Fig. 8 is the structural block diagram of resource allocation device in one embodiment；

Fig. 9 is the structural block diagram of computer equipment in one embodiment.

Specific embodiment

It is with reference to the accompanying drawings and embodiments, right in order to which the objects, technical solutions and advantages of the application are more clearly understood The application is further elaborated.It should be appreciated that specific embodiment described herein is only used to explain the application, and It is not used in restriction the application.

Fig. 1 is the applied environment figure of resource allocation methods in one embodiment.Referring to Fig.1, the resource allocation methods application In resource allocation system.The resource allocation system includes terminal 110 and server 120.Terminal 110 and server 120 pass through net Network connection.Terminal 110 specifically can be terminal console or mobile terminal, and mobile terminal specifically can be with mobile phone, tablet computer, notes At least one of this computer etc..Server 120 can use the server of the either multiple server compositions of independent server Cluster is realized.Server 120 can determine that the account to be chosen that using terminal 110 logs in is directed to the distribution of each resource to be allocated Probability.Wherein, resource to be allocated refers to the resource that can distribute to user, including virtual kinds of goods and real kinds of goods, virtual kinds of goods It can be ideal money, discount coupon etc., real kinds of goods can be commodity exchange certificate, cash of return etc..

As shown in Fig. 2, in one embodiment, providing a kind of resource allocation methods.The present embodiment is mainly in this way It is illustrated applied to the server 120 in above-mentioned Fig. 1.Referring to Fig. 2, which specifically comprises the following steps:

Step 202, the sharing rate of account to be chosen is obtained；

Account to be chosen refers to the account in default application, and default application can be various social accounts, such as wechat, QQ Deng then account to be chosen is then the account of wechat user and the account of QQ user etc..Sharing rate be for measure each to Choose the numerical value that account carries out splitting glass opaque after getting resource to be allocated.When the splitting glass opaque wait choose account is more, The sharing rate for being then somebody's turn to do account to be chosen is then higher.Assuming that account to be chosen is the account of wechat user, when account to be chosen is every After once getting resource to be allocated, it is somebody's turn to do account to be chosen and has carried out splitting glass opaque, such as will acquire the page of resource to be allocated Face is shared to good friend or social platform etc., then the sharing rate for being somebody's turn to do account to be chosen then can be higher.

Step 204, the sharing attribute score value of account to be chosen is determined according to sharing rate；

What sharing rate can get the number of resource to be allocated and account to be chosen according to account to be chosen shares number meter It obtains.And share attribute score value can according to sharing rate determine, further, share attribute score value can according to sharing rate and to The Social behaviors attribute for choosing account determines, can be used for measuring account to be chosen for getting the perception of resource to be allocated and dividing It enjoys the degree of wish, i.e., is ready the probability shared wait choose account after getting resource to be allocated.Combine wait choose Each behavioural characteristic of account and wait choose account for the sharing probability after getting resource to be allocated, calculates wait choose The sharing attribute score value of account.

Step 206, target account is chosen according to sharing attribute score value；

It is each after the sharing attribute score value for choosing account calculating, target account can be chosen according to attribute score value is shared Family.Further, it can select and share to be chosen account of the attribute score value more than point threshold as target account.Set score value Threshold value can illustrate the sharing row of this part account to be chosen when the sharing attribute score value wait choose account is more than point threshold To be more, i.e., after the account when this part gets resource to be allocated, the result for getting resource also can more be happy to point It enjoys.Therefore this part account to be chosen may be selected as target account, be that target accounts that these are selected treat point The configuration of allocation probability with resource enables target account to have bigger probability to get resource to be allocated.Compared with other accounts For family, there is a possibility that bigger after getting resource to be allocated in target account, right to the resource to be allocated institute got The activity answered is shared, and is reached target account after getting resource to be allocated, is treated inside the circle of friends of target account Distribute the effect that the corresponding activity of resource carries out active propagation.

Step 208, determine target account for point of each resource to be allocated according to the sharing attribute score value of target account With probability.

It, can be according to the corresponding sharing attribute point of each target account after having selected target account according to sharing attribute score value Value determines each target account for the allocation probability of each resource to be allocated.Resource to be allocated can have multiple and different Value corresponding to resource to be allocated is also different.Determining each target account for the allocation probability of each resource to be allocated When, for each resource to be allocated, each target account has corresponding allocation probability.It is directed to each resource to be allocated, often A target account has certain probability to get.The method of salary distribution can be, and determine the target account each selected for every The allocation probability of a resource to be allocated is consistent, i.e., the probability that each target account gets each resource to be allocated is identical 's.It can not also be all that each target account is arranged according to share attribute score value according to the sharing attribute score value of target account For the different allocation probability of each resource to be allocated.I.e. each target account gets the probability meeting of each resource to be allocated It is different, target account can be determined for point of each resource to be allocated according to the sharing attribute score value of each target account With probability.Further, for sharing the higher target account of attribute score value, for the distribution for being worth higher resource to be allocated Probability can be higher.

Above-mentioned resource allocation methods, can be stronger for social attribute, shares the higher account setup of attribute score value and waits for The allocation probability for distributing resource, can enable this certain customers after getting resource to be allocated, actively to activity with this Shared, improve movable propaganda strength, promotion costs can also be saved by being not only able to promote movable propaganda strength.

In one embodiment, as shown in figure 3, according to the step of attribute score value chooses target account is shared, comprising:

Step 302, the account alternately account for sharing that attribute score value is more than point threshold is chosen.

Step 304, the corresponding influence value of alternative account is obtained.

Step 306, the alternative account of preset quantity is selected as target account according to influence value.

It is determined each after the sharing attribute score value for choosing account according to sharing rate, can select that share attribute score value super Cross the account of point threshold alternately account.Point threshold is the threshold value set by technical staff, for according to sharing attribute Score value filters out target account.It is available to influence value corresponding with each alternative account after selecting alternative account.Shadow Ringing value can be used for measuring the propagation value for making account get resource to be allocated.It therefore, can be according to the shadow of each alternative account It rings value and selects the alternative account of preset quantity as target account.Preset quantity is set by technical staff, can be according to choosing The demand taken carries out customized setting to quantity.

In one embodiment, influence value includes the first influence value and the second influence value.It is corresponding to obtain alternative account Influence value, comprising: obtain the first influence value of alternative account, the first influence value is the user for having friend relation with alternative account In, it gets resource to be allocated and there are good friend's numbers of sharing operation and alternative account to get the number of resource to be allocated With；The second influence value of alternative account is obtained, the second influence value is to have in the user of friend relation with alternative account, the first shadow Ring the number that value is default value.

Influence value is the first influence value and the second influence value respectively, getting, alternative account is corresponding there are two including When influence value, that is, get the first influence value and the second influence value of alternative account.Wherein, the first influence value is and alternative account Have the user of friend relation or be located in same relation chain with alternative account, gets resource to be allocated and there are sharing operations Good friend's number and alternative account get resource to be allocated number and.It is available when calculating the influence value of account Obtained historical data is observed to launch to account.Dispensing observation refers to be allocated to the alternative account setup selected The allocation probability of resource, so that each alternative account has certain chance to get resource to be allocated, so that observable is alternative Account is in launching the observation stage, the number of the sharing operation after getting resource to be allocated.It can be existed according to each alternative account The dispensing observation stage counts the number of the sharing operation after getting resource to be allocated, each standby so as to be calculated Select influence value corresponding to account.It gets resource to be allocated and there are sharing operations to refer to that user is getting resource to be allocated When, sharing operation can be carried out to the relevant interface of resource to be allocated is got.Sharing operation refer to user get it is to be allocated When resource, the operation checking the relevant interface of resource to be allocated and being shared, sharing operation can be user get to Screenshotss operation after distributing resource.And the second influence value is to have in the user of friend relation with alternative account, the first influence value Equal to the number of default value.Default value is by the pre-set numerical value of technical staff, it is assumed that default value is set as 0, that Second influence value is then the number that the first influence value is 0 in the user that alternative account has friend relation.

Assuming that alternative account is account A, in default application, good friend's number of account A is N, is got in this N number of people Resource to be allocated and the good friend's number operated there are screenshotss are M, and the number that account gets resource to be allocated is S.So account A The first influence value be then M+S.By this method, the first influence value of each good friend of account A is calculated, and gets the first influence Value is good friend's number of default value, as the second influence value of account A.

In one embodiment, the alternative account of preset quantity is selected as target account according to influence value, comprising: choosing Taking the first influence value is the alternative account of default influence value；The is chosen from the alternative account that the first influence value is default influence value Two influence values sort from large to small the alternative account of forward preset quantity as target account.

After getting the corresponding influence value of each alternative account, the alternative account of preset quantity can be selected according to influence value As target account.Influence value includes the first influence value and the second influence value, is influenced in calculate each alternative account first After value and the second influence value, the alternative account that the first influence value is default influence value can be selected, then therefrom choose second to influence It is worth forward alternative account as target account.It is the alternative account of default influence value by the first influence value, according to the second shadow It rings value to be ranked up from big to small, and selects the higher alternative account of the second influence value of preset quantity as target account.The What one influence value represented is account to oneself getting perception to be allocated, and the second influence value is represented and thrown the account Put the propagation value for getting resource entitlement to be allocated.Assuming that the first influence value is set as zero, according to the calculating of the first influence value Mode is not acquired the account that real estate impact to be allocated is crossed it is found that the account that this first influence value of part selected is zero refers to Family.And the second influence value is forward, represents the propagation value of account, therefore the first influence value selected be default influence value and Second influence value sorts from large to small forward alternative account, then refers to and be not acquired also that real estate impact to be allocated is crossed and have Standby certain account for propagating value.The setting that resource allocation probability to be allocated is carried out to this part account, enables these accounts Enough more high probabilities get resource to be allocated after, this part account can more have probability to getting corresponding to resource to be allocated Activity is shared and is propagated, it is easier to which correlated activation widens one's influence power.

Assuming that default influence value is set as 0, then it can select the first influence value and be 0 alternative account, then get this Second influence value of the alternative account that the first influence value of part is 0.The alternative account that the first influence value come is 0 will be selected According to the second influence value, it is ranked up from big to small.That is the highest account of the second influence value makes number one, and the second influence value is most Low account comes last bit.Then it can select the second influence value from the alternative account that the first influence value is 0 and arrange from big to small The alternative account of the forward preset quantity of sequence is as target account.

In one embodiment, as shown in figure 4, the step of obtaining the sharing rate of account to be chosen, comprising:

It step 402, is the allocation probability of account setup to be chosen resource to be allocated.

Step 404, after receiving resource to be allocated according to allocation probability wait choose account, point of account to be chosen is obtained Enjoy number.

Step 406, the sharing rate for obtaining account to be chosen is the sharing number of account to be chosen and receives money to be allocated The ratio of the number in source.

Sharing rate wait choose account can be the screenshotss probability wait choose account after getting resource to be allocated.Specifically Ground is obtaining each when choosing the sharing rate of account, can treat and choose account and carry out dispensing observation.That is, for each wait choose The account allocation probability certain for each resource distribution to be allocated, then each account to be chosen has certain probability that can obtain It is available to each wait choose sharing number of the account after getting resource to be allocated to resource to be allocated.Due to Family gets the interface of resource to be allocated and is not provided with sharing prompt, it can thus be assumed that the spontaneous screenshotss behavior of user is exactly to use The splitting glass opaque at family, therefore can be using screenshotss probability as the sharing rate of account to be chosen.In the page for getting resource to be allocated On, it is built-in with the function code of screen monitoring, operates, then may be used when user triggers screenshotss in the page for getting resource to be allocated Know that user has carried out screenshotss operation.It is equivalent to user and triggers screenshotss operation in " prize-winning " page, then can listen to user Screenshotss operation.Therefore the sharing rate of user can be time that account carries out screenshotss operation in the page for getting resource to be allocated The ratio of several numbers that resource to be allocated is got with account.

In one embodiment, before obtaining the sharing rate of account to be chosen, further includes: obtain the account in default application Family；The active account and/or influence power account chosen in the account in default application are used as account to be chosen.

Default application can be specified application, such as when the default application of setting is wechat in application, then account then refers to It is account, i.e. wechat account corresponding to the user of wechat application；When the default application of setting is QQ in application, then account refers to Account, i.e. QQ account corresponding to the user of QQ application.Account in the default application got can be screened, be selected Active account is used as account to be chosen.Active account refers to that liveness meets the account of active condition, and technical staff can be according to item Mesh demand sets one or more active conditions.When the liveness of account meets the active condition of setting, then it is believed that the account Family is active account.Influence power account refers to the account for meeting influence power account terms in default application, when the influence of account When power meets the condition of the influence power account of setting, then it is believed that the account is influence power account.

In one embodiment, the condition for meeting active account includes at least following one kind: within a preset time, account is logical Crossing and issuing the quantity of personal information in default application is more than publication amount threshold；Within a preset time, account passes through default application Middle participation group chat speech number is more than speech amount threshold；Within a preset time, that account is issued by default application with good friend The quantity that people's information is interacted is more than interaction amount threshold；Within a preset time, account checks default text by default application The number of chapter is more than to read amount threshold；Within a preset time, account is provided by default application and other interior accounts of application The quantity of source transmission is more than transmission quantity threshold value；It is more than forwarding quantity threshold that account, which forwards the quantity of default article by default application, Value.

Judge whether an account belongs to active account, then needs to judge whether account meets active condition.Specifically, living Jump condition including but not limited to:

Condition 1: within a preset time, account is more than publication quantity threshold by issuing the quantity of personal information in default application Value.Publication personal information refers to that account shares the behaviors such as personal dynamic in default application.Issue the behavior meeting of personal information According to the different and different of default application, for example when default application is wechat, then issues personal information and then refer to account Circle of friends dynamic is delivered, and when default application is QQ, publication personal information then refers to that account delivers QQ space dynamic etc..It is default Time is the setting of a period, for example the first preset time is set as 24 hours, and condition 1, which then refers to, judges that account is daily Publication amount threshold whether is had more than by the quantity of default application publication personal information.

Condition 2: within a preset time, account is more than speech amount threshold by participating in group chat speech number in default application. Group chat speech number refers to the quantity made a speech in the group created in default application.For the difference of default application, group chat it is general Thought would also vary from.Such as in QQ group chat include QQ group and discussion group both, then it is believed that in these types difference Group's mode in speech be considered group chat speech.Preset time is also the setting of a period, when default in condition 1 Preset time in time and condition 2 can be identical, can not also be identical.The preset time of assumed condition 2 is set as 7 days, that Condition 2 was judged in 7 days, and whether release news in the group quantity of number of account is more than speech amount threshold.Equally Ground, speech amount threshold can also carry out customized setting by technical staff.

Condition 3: within a preset time, the quantity that account is interacted by default application with the personal information that good friend issues More than interaction amount threshold.In default application, account can issue personal information, such as when default application is wechat, account The personal information of publication can be publication circle of friends dynamic, the article including sharing, the link of sharing, the pure words individual delivered Dynamically, or both pictures and texts are excellent personal dynamic, etc..Interact referring to the personal information of good friend's publication, as the good friend of account When having issued personal information by default application, account is commented on, thumbs up or is forwarded to the personal information that good friend issues Operation.It the operations such as can comment on, thumb up or forward to the personal information of good friend's publication within a preset time according to account Quantity judges the liveness of account.When preset time is set as 7 days, then condition 3 refers to, in 7 days, account passes through Whether the quantity that default application interacts the personal information that good friend issues has more than preset interaction amount threshold, if super It crosses, illustrates that the influence power of account meets condition 3.

Condition 4: within a preset time, account checks that the number of default article is more than to read quantity threshold by default application Value.Default article can be public platform article, be also possible to the article of good friend's sharing, available time that article is checked to account Number, detection account check the number of default article within a preset time, if pre-set reading amount threshold is had more than, if It is to be more than, then illustrates that the influence power of account meets condition 4.

Condition 5: within a preset time, account carries out the quantity of resource transmission by default application with other interior accounts of application More than transmission quantity threshold value.Resource transmission, which is carried out, with other interior accounts of application refers to that receiving other accounts in default application sends Resource, or send resource to other accounts operation.Other accounts, which can be, has good friend with the account in default application The account of relationship is also possible to the account for not having friend relation with the account, i.e. other accounts can be good friend, is also possible to Strange account.Resource transmission is carried out, can be the transmission for carrying out file, is also possible to the transmission such as virtual resource, for example give bonus, Discount coupon is represented or purchase card is represented.It can detecte account within a preset time, carried out by default application with other accounts Whether the quantity of resource transmission is more than transmission quantity threshold value, if being more than, then illustrates that the influence power of account meets condition 5.

Condition 5: it is more than forwarding amount threshold that account, which forwards the quantity of default article by default application,.The default text of forwarding Chapter, the behavior of the specified article of user's forwarding.Such as specified public platform A and B, then the article that public platform A and public platform B are delivered It is then default article.The quantity that user forwards default article can be got, judges that user forwards whether the quantity of default article surpasses Cross forwarding amount threshold.Forwarding amount threshold can also carry out customized setting by technical staff.In condition 4, it can also be added Time limits, for example is limited in 7 days, judges that account forwards whether the quantity of default article has more than forwarding by default application Amount threshold.

In one embodiment, the condition for meeting influence power account includes at least following one kind: with account in default application In have friend relation number be more than good friend's amount threshold；Within a preset time, good friend passes through default application publication to account The quantity that is interacted of personal information be more than that good friend interacts amount threshold；The identity information of account is got by default application Belong to default identity type；Belong to default address type by the certificate address information that default application gets account；With account pre- If in the account for having friend relation in, belonging to default professional class more than the occupational information of good friend's account of preset quantity Type.

Judge whether an account belongs to influence power account, then needs to judge whether account meets influence power condition.Specifically Ground, influence power condition including but not limited to:

Condition A: the number for having friend relation in default application with account is more than good friend's amount threshold.Assuming that default answer With for wechat, then referring to that wechat good friend's number of account is more than preset good friend's amount threshold.Good friend's amount threshold can be with It is adjusted by technical staff according to project demands.

Condition B: within a preset time, the quantity that good friend interacts account by the personal information of default application publication Amount threshold is interacted more than good friend.After account has issued personal information by default application, has friend relation with account Account can carry out interactive operation to the personal information that account is issued, for example thumb up, comment on, forward.It is available to arrive pre- If in the time, it is pre-set whether the number of operations that the personal information that the good friend of account issues account is interacted has more than Good friend interacts amount threshold.If being more than, then illustrate that the influence power of account meets condition B.

Condition C: default identity type is belonged to by the identity information that default application gets account.In default application, User can be with self-setting identity information, and the platform where default application can audit the identity information of user setting. When after the approval, then illustrating that the identity information of user setting is true.Identity information may include the occupation of account, age, year The information such as income.The identity type for thinking have influence power, such as institutional settings office worker, president, investment can be preset People, City Operation Manager etc..When the annual income of one of which or account that the identity information of account belongs in preparatory identity type reaches When the corresponding income condition of preset identity type, it is believed that the eligible C of the corresponding influence power of the identity information of account.

Condition D: default address type is belonged to by the certificate address information that default application gets account.The certificate address information of user Many information can be embodied, for example stays in the personnel of a certain specific region and belongs to government official's inhabitation, or meet certain body The region that the talent of part condition can live.Therefore the influence power of the account can be judged according to the certificate address information of account.When account When certificate address information belongs to pre-set address region or type, then illustrate the eligible D of the influence power of account.

Condition E: in the account for having friend relation in default application with account, more than good friend's account of preset quantity Occupational information belongs to default occupation type.Occupation type refers to the type divided to multiple occupations, such as law class, Internet, financial class etc..In default application, each account can be with multiple good friends, the available occupation to account good friend Information, judges whether the occupational information of account good friend belongs to preset occupation type.And preset occupation is belonged to occupational information Good friend's quantity of type counts, when the occupational information for the good friend's account for being more than preset quantity belongs to default occupation type, Then illustrate that the influence power of account meets condition E.

In one embodiment, above-mentioned condition 4 can also be that account propagates the number of default active link by default application Amount is more than preset propagation amount threshold.For example there are prize-winning activity A, can be propagated by replicating default active link, The then quantity of the available default active link to account forwarding about movable A.Default active link is one for specific Activity, for sharing the link of the activity or webpage, duplication creaks after password, open corresponding application can jump to it is corresponding Loose-leaf.

In one embodiment, the sharing attribute score value of account to be chosen is determined according to sharing rate, comprising: obtain wait choose Behavioural characteristic of the account in default application；The sharing rate and behavioural characteristic of account to be chosen are input in neural network, it is defeated The sharing attribute score value of account to be chosen out.

Account the behavioural characteristic in default application refer to account using default in application, by default application The reflected feature of behavior institute.Assuming that default application is wechat, then behavioural characteristic of the account in wechat then can be Speech number of the account in wechat group chat, the dynamic frequency of account publication circle of friends, the article frequency of account forwarding public platform with And share the frequency etc. of various activities between account and good friend.Getting row of each account to be chosen in default application After being characterized, can by the behavioural characteristic of each account to be chosen with share rate and be input in neural network.By neural network pair The behavioural characteristic of the account to be chosen of input is analyzed, and the common trait of account to be chosen is learnt, each so as to calculate The sharing attribute score value of account to be chosen.Specifically, the behavioural characteristic of user can be converted into corresponding behavior vector, then inputted Into neural network, convolution algorithm is carried out to the behavior vector of input by the convolutional layer of neural network, behavior vector can be obtained Feature, so as to calculate corresponding sharing attribute score value.Before with neural network, neural network can be trained. The behavioural characteristic that multiple users can be got from database, is input to after the behavioural characteristic of user is converted into behavior vector In neural network, the weight of neural network is adjusted, neural network is trained with this, so that the nerve net after training The sharing attribute score value that network is calculated is more accurate.Server can get each account to be chosen of neural network output Sharing attribute score value.That is, neural network can be according to the sharing rate of account to be chosen and the behavioural characteristic of corresponding account to be chosen Registration evaluates the sharing attribute score value of user.

In one embodiment, determine target account for each money to be allocated according to the sharing attribute score value of target account The allocation probability in source, comprising: when the sharing attribute score value of target account is higher, it is determined that target account is preset to be allocated The allocation probability of resource is higher.

Preset resource to be allocated can be in resource to be allocated, be worth highest that resource to be allocated, can also be with It is multiple higher resources to be allocated of value.It, can be by target account after getting the sharing attribute score value of each target account Family is ranked up from high to low according to the score value for sharing attribute score value.It may thereby determine that and share the higher target account of attribute score value When family is directed to preset costly resource to be allocated, corresponding allocation probability is higher.As share the higher target of attribute Account configures the costly higher allocation probability of resource to be allocated.Assuming that there are the resource to be allocated of 5 different values, point Not are as follows: [5,10], [11,30], [31,50], [51,100], [101,200], there are 5 target accounts, with this 5 targets The corresponding sharing attribute score value of account is respectively 0.4,0.5,0.8,0.6,0.7.It can be seen that, it is right for 0.8 to share attribute score value The score value for the target account answered is highest.So can be share attribute score value be 0.8 target account setting it is costly to The higher allocation probability of resource is distributed, i.e., the target account that sharing attribute score value is 0.8 is directed to [101,200] this value interval Resource to be allocated allocation probability, can be higher compared to other target accounts and other non-targeted accounts.

When configuring the allocation probability of each resource to be allocated according to the sharing attribute score value of target account, can will select Target account as a special group, such as the first group, using other non-targeted accounts as another special group, than Such as the second group.It can be target account institute to improve the target account selected for the allocation probability of resource to be allocated The first group the higher quantity for getting resource to be allocated is set, i.e., compared to second group where non-targeted account More " prize-winning quantity " is arranged for the first group in body.Target account then be can be improved by this method for resource to be allocated Allocation probability.Each target account and non-targeted account have an allocation probability for each resource to be allocated, but in order to Movable propaganda strength is improved, the higher target account of attribute score value can be selectively shared, it is higher wait divide for being worth With resource, higher allocation probability is set.

In one embodiment, as shown in figure 5, when the sharing attribute score value of target account is higher, it is determined that target account Family is higher in the allocation probability of preset resource to be allocated, comprising:

Step 502, the active budget numerical value of preset resource to be allocated is obtained.

Step 504, the payment number and payment amount of value interval, are calculated where obtaining preset resource to be allocated The payment mean value of value interval where preset resource to be allocated.

Step 506, it is determined according to the active budget numerical value of preset resource to be allocated with payment mean value preset to be allocated The quantity allotted of resource.

Active budget numerical value refers to the fund numerical value of movable active budget cost corresponding to resource to be allocated.It is determining It can be the resource distribution to be allocated of each value interval according to active budget cost after first group corresponding to target account Specific quantity configures activity " prize-winning " quantity of each value interval according to activity cost budget.Preset money to be allocated Source refers to the resource to be allocated of some specific value interval in multiple resources to be allocated.For example resource to be allocated has 5 Value interval, respectively (5,10], (10,30], (30,50], (50,100], (100,200].When it needs to be determined that in value interval (5,10] resource to be allocated quantity allotted when, preset resource to be allocated be then value interval (5,10] money to be allocated Source.

Specifically, Analysis and summary can be carried out by buying the payment amount of commodity to previous user, it is available to each The mean value of user's payment amount of a value interval.In activity operation, by the summary of operation experience, it can know, for Single user, it is larger for quantity although the current transaction order amount of money distribution of these users is in fluctuation unstable state User group default application for, for example default application is wechat, then can analyze to obtain according to data, user uses micro- Letter payment when, the transaction amount distribution proportion of the order of payment be it is relatively stable, i.e., wechat pay user in some time The probability of some specified amount of Duan Fasheng expenditure is expected.Therefore, it can be accustomed to according to the consumption expenditure of user usually, be obtained To user each value interval payment mean value.Such as it is available to payment amount (10,30] this amount of money section User's total number of persons, can calculate user (10,30] the payment mean value in this amount of money section.Assuming that (10,30] this gold Payment mean value between frontal region is 22 yuan, and for (10,30] activity cost that sets of this amount of money section as 1000 yuan, then can count Calculation obtain (10,30] " prize-winning " quantity in this amount of money section is 1000/22=45.So 45 be (10,30] this The quantity allotted in specified amount section.

Step 508, determine that target account gets preset money to be allocated according to the quantity of target account and quantity allotted The mean allocation probability in source.

After selecting target account, the quantity of target account can be got.According to the quantity of target account about step 506 The quantity allotted of middle determination, can determine each target account can get preset resource to be allocated mean allocation it is general Rate.For example target account is 200 people, quantity allotted is 100, then mean allocation probability is 100/200=0.5.

Step 510, when the sharing attribute score value of target account is higher, determine target account in preset resource to be allocated Closer to the allocation probability of mean allocation probability.

After mean allocation probability has been determined, it would know that each target account can get the probability of resource to be allocated. On the basis of this, the probability that can get resource to be allocated to target account according to attribute score value is shared is set. Higher allocation probability can be set for the higher target account of score value, it can determine and share the higher target account of attribute score value Family for resource to be allocated allocation probability closer to mean allocation probability.Closer to referring to, with mean allocation probability work For maximum value, corresponding value is successively reduced.Assuming that mean allocation probability is 0.6, then closer to mean allocation probability It is meant that, is used as maximum probability value for 0.6, the allocation probability of target account can be less than 0.6, using 0.6 as boundary, root The allocation probability of target account is determined according to the height for sharing attribute score value.For example sharing attribute score value is respectively 0.8 He 0.6 target account, corresponding allocation probability can be set to 0.55 and 0.4.Share the higher target account of attribute score value Allocation probability closer to mean allocation probability 0.6.

Further, in this embodiment distribution method, can be applied to the special group of any delineation " prize-winning " number The distribution of amount.The calculation formula that the special group that delineation comes out is directed to the allocation probability of specific resource to be allocated is as follows: specific The estimated group in the allocation probability of the resource to be allocated in amount of money section=section setting quantity allotted/section participates in number. Wherein, specified amount section refers to value interval corresponding to each resource to be allocated, if than want determine (10,30] this The allocation probability in one specified amount section, then specified amount section refer to be exactly (10,30].The quantity allotted of section setting It is the numerical value being calculated according to activity cost budget with the payment mean value in this specified amount section.The estimated group in the section Body participates in number, refers to that analysis obtains in this specified amount section according to counting to the payment behavior of previous user Belong to the payment number of special group, special group can be corresponding first group of target account, be also possible to non-targeted account Corresponding second group in family is also possible to the third group or the of the account taken out according to selected by other selection modes composition Four groups etc..

It when determining that the estimated group of specific sections participates in number, can calculate according to the following formula: the estimated group of specific sections Body participates in the participation number * of the estimated group in the number=whole section specific sections pay invoice ratio.Whole sections it is pre- The participation number of meter group refers to the payment number for the special group that delineation comes out, can be according to the total number of persons of special group and activity Participation rate is calculated.The specific sections pay invoice ratio refers to pay invoice quantity and whole areas in this specific sections Between total payoff quantity on order ratio.Therefore, the estimated group of specific sections participates in number can also be calculated by following formula Obtain: the estimated group of specific sections participates in number=(special group total number of persons * activity participation rate) *, and (specific sections payment is ordered Singular amount/whole section total payoff quantity on order).In the allocation probability for the resource to be allocated that specified amount section is calculated Afterwards, further allocation probability can be set according to the sharing attribute score value of account.

It is under normal circumstances, first group corresponding for selecting the target account come in order to improve movable propaganda strength Body can be higher than corresponding second group of non-targeted account for the quantity allotted of the higher resource to be allocated of value interval.In this way, For the first group corresponding for target account, can also it be higher than for the allocation probability of the higher resource to be allocated of value interval Corresponding second group of non-targeted account.

As shown in fig. 6, in one embodiment, providing a kind of resource allocation methods.The present embodiment is mainly in this way It is illustrated applied to the server 120 in above-mentioned Fig. 1.Referring to Fig. 6, which specifically comprises the following steps:

Step 602, the account in default application is obtained.

Step 604, the active account chosen in the account in default application is used as account to be chosen.

When being allocated to resource to be allocated, account can be screened, then determine that the target account selected is directed to The allocation probability of each resource to be allocated.Therefore, the first stage will then get the account in default application and be screened, The active account in default application is selected as account to be chosen.The condition for meeting active account includes at least following one kind: Having the number of friend relation in default application with account is more than good friend's amount threshold；In the first preset time, account is logical Crossing and issuing the quantity of personal information in default application is more than publication amount threshold；In the second preset time, account passes through default It is more than speech amount threshold that group chat speech number is participated in；It is more than to turn that account, which forwards the quantity of default article by default application, Send out amount threshold.Judge that account forwards whether the quantity of default article is more than that forwarding amount threshold can pass through by default application Model is judged that, when default application is wechat, the content of model includes to the sharing communication network of article, wechat in wechat In specific public platform article sharing communication network.It, can be in conjunction with the reading generated after user's forwarding based on above-mentioned two network The biggish crowd of number is covered after click volume and forwarding, may thereby determine which user is to belong to public platform article forwarding shadow Ring the user of power.Active condition can also be added account and share other movable frequencies, for example, user share it is to be allocated with this The activity of resource competing product each other.

In one embodiment, good friend's amount threshold can be set to 700.First preset time can be 24 hours, hair Cloth amount threshold can be 25.Second preset time can be 7 days, and speech amount threshold can be 10.

It step 606, is the allocation probability of account setup to be chosen resource to be allocated.

Step 608, after receiving resource to be allocated according to allocation probability wait choose account, cutting for account to be chosen is obtained Shield number.

Step 610, the sharing rate for obtaining account to be chosen is the screenshotss number of account to be chosen and receives money to be allocated The ratio of the number in source.

It is filtering out after choosing account, can further treat selection account and be screened, can take and treat selection account Launch the strategy of observation.For the allocation probability of each resource to be allocated of account setup to be chosen, to get when to be selected Screenshotss number after taking account to get resource to be allocated is received according to the screenshotss number of account to be chosen with account to be chosen The specific gravity of the number of resource to be allocated can determine the sharing rate to be selected for taking account.Due to getting resource to be allocated in account The page on do not have sharing prompt, it can thus be assumed that account screenshotss operation be sharing operation.

As shown in fig. 7, a in Fig. 7, b, these three figures of c are that account gets the money to be allocated that value is 0.25 yuan respectively It is that any money to be allocated has not been obtained in the page for exempting from singly to reward and user that the corresponding page in source, account, which get resource to be allocated, The page in source.On this page, there are sharings to prompt, therefore the screenshotss of active each time of user may be considered use The active sharing operation at family, i.e., it is believed that user is after screenshotss that the picture of screenshotss may be sent to good friend or be published in social flat Platform is shared with other users.

Step 612, behavioural characteristic of the account to be chosen in default application is obtained.

Step 614, it by the sharing rate and behavioural characteristic of account to be chosen, is input in neural network, exports account to be chosen The sharing attribute score value at family.

Step 616, the account alternately account for sharing that attribute score value is more than point threshold is chosen.

Further, it is getting after the sharing rate for choosing account, account to be chosen can also be got and answered default Behavioural characteristic in, therefore the sharing rate and behavioural characteristic of account to be chosen can be input in neural network.Neural network It can learn the common trait between user, to calculate the sharing attribute score value of each account to be chosen.Getting mind After the sharing attribute score value of network output, the account alternately account for sharing that attribute score value is more than point threshold can be selected Family.The range for sharing attribute score value is 0-1, can set point threshold to 0.4, i.e., sharing attribute score value be more than 0.4 it is to be selected Take account that will be selected alternately account.Point threshold can be in progress by technical staff according to the allocation activities of resource to be allocated Constantly it is adjusted.

Step 618, corresponding first influence value of alternative account and the second influence value are obtained.

Step 620, the alternative account that the first influence value is default influence value is chosen.

Step 622, the second influence value is chosen from the alternative account that the first influence value is default influence value to arrange from big to small The alternative account of the forward preset quantity of sequence is as target account.

Step 624, determine target account for point of each resource to be allocated according to the sharing attribute score value of target account With probability.

After selecting alternative account, the influence value of alternative account can be got, further so as to according to influence value Target account is selected from alternative account.Influence value is divided into the first influence value and the second influence value, the first influence value be with it is standby It selects account to have in the user of friend relation, gets resource to be allocated and there are the good friend's numbers and alternative account of screenshotss operation Get the sum of the number of resource to be allocated.Second influence value is to have in the user of friend relation with alternative account, the first shadow Ring the number that value is default value.First influence value is known as the R factor, the second influence value is known as the P factor, and the R factor represents account Family itself then represents the propagation valence that resource to be allocated is launched to account for getting the perception of resource to be allocated, the P factor Value.

When obtaining corresponding first influence value of alternative account and the second influence value, can be obtained by machine violence algorithm It takes.Due to the definition of the R factor and the P factor be it is determining, can be dependent on the high-speed computational capability of server, so that server It can traverse the alternative account selected in default application according to the definition of the R factor and the P factor and calculate each alternative account The R factor values and P factor values at family.Server can also traverse whole accounts in default application, calculate the R of whole accounts After factor values and P factor values, then filter out alternative account.To go out to meet further according to the R factor and P predictor selection of alternative account Target account.Target account refers to that the R factor be default influence value is default influence value and the forward present count of the second influence value The alternative account of amount.After selecting target account, it may be determined that each target account is directed to the allocation probability of each resource to be allocated. In the present embodiment, value corresponding to each resource to be allocated is different, than if any resource to be allocated value interval For [5,10], have plenty of [11,30].It, can be in order to improve this movable promotion efficiency of resource to be allocated and publicity coverage Determine that target account is directed to the allocation probability of each resource to be allocated according to the sharing attribute score value of the target account selected.

In one embodiment, resource to be allocated refers to that user obtains the chance of corresponding prize, and prize can be virtual goods Coin can be discount coupon, commodity exchange certificate or cash etc..Resource to be allocated can have multiple, it can there are multiple The prize of different value intervals, such as [5,10], [11,30], [31,50], [51,100], the difference such as [101,200] Valuable section Between cash bonuses etc..Account in each default application of this activity each may participate in, that is, be directed to the prize of each value interval, Each account has respective " prize-winning " probability, i.e. allocation probability.In order to improve movable propaganda strength, promotion costs are reduced, Account can be analyzed, select qualified account as target account, the target account that this part is selected Allocation probability configured.So that the probability that " prize-winning " probability of target account obtains costly prize is bigger, and target Account also can have bigger probability to share activity after getting prize compared to other users, to also can be improved Movable popularity, also reduces movable promotion costs.

Specifically, using the big several Theoretical Designs of statistics to the allocation rule of resource activity to be allocated.Such as step 506, Early period can launch resource to be allocated, it can the distribution for being the account setup to be chosen that selects to resource to be allocated These accounts to be chosen can be had an opportunity to get resource to be allocated for probability.It is observed by the dispensing of several phases, it is available to arrive Resource to be allocated is corresponding to be worth and shares the breakpoint relationship between rate.

As shown in figure 8, the sharing rate of user is minimum about when the amount of value of resource to be allocated is the section of [5,10] member It is 25%, and when the amount of money is greater than 50 yuan, sharing rate difference is simultaneously little.Therefore section similar in sharing rate can be closed And merge into a section.Separately for the ease of movable cost management, movable cost fluctuation is reduced, amount of money section span is not It is preferably excessive, therefore section similar in adjacent and sharing rate can be merged.Such as by shown in figure, similar in sharing rate (10,20] section with (20,30] section merge into a section (10,30], by similar in sharing rate (30,40] section with (40, 50] section merge into a section (30,50].Specifically, can the resource to be allocated by value interval in [5,200] be divided into 5 Value interval, respectively (5,10], (10,30], (30,50], (50,100], (100,200].

The mean value that the fund numerical value of active budget cost and user's payment amount of each value interval can be got, from And the quantity allotted of each value interval is calculated.So as to determine target account according to the quantity of target account and quantity allotted Family gets the mean allocation probability of preset resource to be allocated, that is, determine target account get each value interval to point Mean allocation probability with resource.It on this basis, can be according to the sharing attribute score value of target account, to share attribute score value Higher target account determines the allocation probability closer to mean allocation probability, " sharing wish " can either be made higher with this Target account get resource to be allocated probability it is higher, additionally it is possible to so that select come target account get it is to be allocated The probability of resource can also be higher than other non-targeted accounts.

By the above-mentioned this method of salary distribution, certain amount of resource to be allocated can be determined according to activity cost budget fund Quantity allotted, i.e., quantity of " getting the winning number in a bond ".Using Principle of Statistics, evenly dispersed to the quantity allotted after determination can be delivered to each The resource to be allocated of value interval, it can the random mimicry of quantity allotted after determination is delivered to each resource to be allocated, with This can determine the quantity allotted of each resource to be allocated.That is, " prize-winning " quantity of each value interval can be determined. In addition, this method of salary distribution has scalability, carried out in addition to the corresponding group of target account for selecting and can be directed to " in The distribution of prize " quantity can also be applied to the distribution of " prize-winning " quantity of the special group of any delineation.This method of salary distribution There are two purposes, first is that having certain similar characteristics due to selecting the target account come, has deviated from randomness, therefore pass through A certain proportion of " prize-winning " list is thrown to non-targeted account group, so that the group of non-targeted account can also have an opportunity to obtain Resource to be allocated is got, randomness is produced, the random sense of resource to be allocated can be got by increasing everybody, eliminate people The trace of work intervention.Second is that when being chosen to target account, it can be by machine learning model to the sharing attribute point of account Value is determined, if crowd's sample is all that artificial progress is selected in advance, the sharing category that machine learning model is calculated Property score value can have certain deviation, so as to cause according to share the target account group that selects of attribute score value can exist it is inclined Difference.The final effect of machine learning model is influenced whether in this way, therefore random production can be added in machine learning sample every time Raw sample can increase the comprehensive and correctness of machine learning model learning outcome in this way.

Fig. 2-Fig. 6 is the flow diagram of resource allocation methods in each embodiment.Although should be understood that each figure Flow chart in each step successively show that but these steps are not inevitable to indicate according to arrow according to the instruction of arrow Sequence successively execute.Unless expressly stating otherwise herein, there is no stringent sequences to limit for the execution of these steps, these Step can execute in other order.Moreover, at least part step in each figure may include multiple sub-steps or Multiple stages, these sub-steps or stage are not necessarily to execute completion in synchronization, but can be at different times Execute, these sub-steps perhaps the stage execution sequence be also not necessarily successively carry out but can with other steps or its The sub-step or at least part in stage of its step execute in turn or alternately.

In one embodiment, as shown in figure 8, providing a kind of resource allocation device, comprising:

Sharing rate obtains module 802, for obtaining the sharing rate of account to be chosen.

Share attribute score value determining module 804, for determining the sharing attribute score value of account to be chosen according to sharing rate.

Target account chooses module 806, for choosing target account according to sharing attribute score value.

Allocation probability determining module 808 determines target account for every for the sharing attribute score value according to target account The allocation probability of a resource to be allocated.

In one embodiment, sharing rate obtains module 802 and is also used to dividing for account setup to be chosen resource to be allocated With probability；After receiving resource to be allocated according to allocation probability wait choose account, the sharing number of account to be chosen is obtained；It obtains Take the sharing rate of account to be chosen for the ratio of the sharing number and the number for receiving resource to be allocated of account to be chosen.

In one embodiment, above-mentioned apparatus further includes that account to be chosen obtains module (not shown), for obtaining Account in default application；The active account and/or influence power account chosen in the account in default application are used as account to be chosen Family.

In one embodiment, share attribute score value determining module 804 and be also used to obtain account to be chosen in default application In behavioural characteristic；It is input to the sharing rate and behavioural characteristic of account to be chosen in neural network, exports account to be chosen Share attribute score value.

In one embodiment, target account chooses module 806 and is also used to choose distributive property score value more than point threshold Account alternately account；Obtain the corresponding influence value of alternative account；The alternative account of preset quantity is selected according to influence value Family is as target account.

In one embodiment, influence value includes the first influence value and the second influence value.Target account chooses module 806 It is also used to obtain the first influence value of alternative account, the first influence value is to have in the user of friend relation with alternative account, is obtained Get resource to be allocated and there are good friend's number of sharing operation and alternative account get the number of resource and；It obtains alternative Second influence value of account, the second influence value are to have in the user of friend relation with alternative account, and the first influence value is equal to pre- If the number of numerical value.

In one embodiment, it is default influence value that target account, which chooses module 806 and is also used to choose the first influence value, Alternative account；Chosen from the alternative account that the first influence value is default influence value the second influence value sort from large to small it is forward The alternative account of preset quantity is as target account.

In one embodiment, allocation probability determining module 808 is also used to when the sharing attribute score value of target account is higher When, it is determined that target account is higher in the allocation probability of costly resource to be allocated.

In one embodiment, allocation probability determining module 808 is also used to obtain the activity of preset resource to be allocated in advance Count value；The payment number and payment amount of value interval where obtaining preset resource to be allocated, be calculated it is preset to The payment mean value of value interval where distributing resource；It is true according to the active budget numerical value of preset resource to be allocated and payment mean value The quantity allotted of fixed preset resource to be allocated；It is pre- with quantity allotted to determine that target account is got according to the quantity of target account If resource to be allocated mean allocation probability；When the sharing attribute score value of target account is higher, determine target account pre- If resource to be allocated closer to mean allocation probability allocation probability.

Fig. 9 shows the internal structure chart of computer equipment in one embodiment.The computer equipment specifically can be Fig. 1 In server 120.As shown in figure 9, it includes the place connected by system bus that the computer equipment, which includes the computer equipment, Manage device, memory and network interface.Wherein, memory includes non-volatile memory medium and built-in storage.The computer equipment Non-volatile memory medium be stored with operating system, can also be stored with computer program, which is held by processor When row, processor may make to realize resource allocation methods.Computer program can also be stored in the built-in storage, the computer journey When sequence is executed by processor, processor execution resources distribution method may make.

It will be understood by those skilled in the art that structure shown in Fig. 9, only part relevant to application scheme is tied The block diagram of structure does not constitute the restriction for the computer equipment being applied thereon to application scheme, specific computer equipment It may include perhaps combining certain components or with different component layouts than more or fewer components as shown in the figure.

In one embodiment, resource allocation device provided by the present application can be implemented as a kind of shape of computer program Formula, computer program can be run in computer equipment as shown in Figure 9.Composition can be stored in the memory of computer equipment should Each program module of resource allocation device, for example, sharing rate shown in Fig. 8 obtain module, share attribute score value determining module, Target account chooses module and allocation probability determining module.The computer program that each program module is constituted executes processor Step in the resource allocation methods of each embodiment of the application described in this specification.

In one embodiment, a kind of computer equipment, including memory and processor are provided, is stored in memory Computer program, the processor perform the steps of the sharing rate for obtaining account to be chosen when executing computer program；According to point Enjoy the sharing attribute score value that rate determines account to be chosen；Target account is chosen according to attribute score value is shared；According to target account Share attribute score value and determines that target account is directed to the allocation probability of each resource to be allocated.

In one embodiment, target account is chosen according to sharing attribute score value, comprising: choosing sharing attribute score value is more than The account of point threshold alternately account；Obtain the corresponding influence value of alternative account；Preset quantity is selected according to influence value Alternative account as target account.

In one embodiment, influence value includes the first influence value and the second influence value；It is corresponding to obtain alternative account Influence value, comprising: obtain the first influence value of alternative account, the first influence value is the user for having friend relation with alternative account In, it gets resource to be allocated and there are good friend's numbers of sharing operation and alternative account to get the number of resource to be allocated With；The second influence value of alternative account is obtained, the second influence value is to have in the user of friend relation with alternative account, the first shadow Ring the number that value is equal to default value.

In one embodiment, the sharing rate of account to be chosen is obtained, comprising: be account setup to be chosen resource to be allocated Allocation probability；After receiving resource to be allocated according to allocation probability wait choose account, the sharing time of account to be chosen is obtained Number；The sharing rate of account to be chosen is obtained as the ratio of the sharing number and the number for receiving resource to be allocated of account to be chosen Value.

In one embodiment, before obtaining the sharing rate of account to be chosen, the computer program also makes described Processor executes following steps: obtaining the account in default application；Choose active account in the account in default application and/or Influence power account is used as account to be chosen.

In one embodiment, determine target account for each money to be allocated according to the sharing attribute score value of target account The allocation probability in source, comprising: when the sharing attribute score value of target account is higher, it is determined that target account is costly wait divide Allocation probability with resource is higher.

In one embodiment, when the sharing attribute score value of target account is higher, it is determined that target account is preset The allocation probability of resource to be allocated is higher, comprising: obtains the active budget numerical value of preset resource to be allocated；Obtain it is preset to Value interval where preset resource to be allocated is calculated in the payment number and payment amount of value interval where distributing resource Payment mean value；Preset resource to be allocated is determined according to the active budget numerical value of preset resource to be allocated and payment mean value Quantity allotted；Determine that target account gets being averaged for preset resource to be allocated with quantity allotted according to the quantity of target account Allocation probability；When the sharing attribute score value of target account is higher, determine that target account is closer in preset resource to be allocated The allocation probability of mean allocation probability.

In one embodiment, a kind of computer readable storage medium is provided, computer program is stored thereon with, is calculated Machine program performs the steps of the sharing rate for obtaining account to be chosen when being executed by processor；It is determined according to sharing rate wait choose The sharing attribute score value of account；Target account is chosen according to attribute score value is shared；It is true according to the sharing attribute score value of target account The account that sets the goal is directed to the allocation probability of each resource to be allocated.

In one embodiment, before obtaining the sharing rate of account to be chosen, when computer program is executed by processor Also perform the steps of the account obtained in default application；Choose the active account in the account in default application and/or influence Power account is used as account to be chosen.

Those of ordinary skill in the art will appreciate that realizing all or part of the process in above-described embodiment method, being can be with Relevant hardware is instructed to complete by computer program, the program can be stored in a non-volatile computer and can be read In storage medium, the program is when being executed, it may include such as the process of the embodiment of above-mentioned each method.Wherein, provided herein Each embodiment used in any reference to memory, storage, database or other media, may each comprise non-volatile And/or volatile memory.Nonvolatile memory may include that read-only memory (ROM), programming ROM (PROM), electricity can be compiled Journey ROM (EPROM), electrically erasable ROM (EEPROM) or flash memory.Volatile memory may include random access memory (RAM) or external cache.By way of illustration and not limitation, RAM is available in many forms, such as static state RAM (SRAM), dynamic ram (DRAM), synchronous dram (SDRAM), double data rate sdram (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronization link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) directly RAM (RDRAM), straight Connect memory bus dynamic ram (DRDRAM) and memory bus dynamic ram (RDRAM) etc..

Each technical characteristic of above embodiments can be combined arbitrarily, for simplicity of description, not to above-described embodiment In each technical characteristic it is all possible combination be all described, as long as however, the combination of these technical characteristics be not present lance Shield all should be considered as described in this specification.

The several embodiments of the application above described embodiment only expresses, the description thereof is more specific and detailed, but simultaneously The limitation to the application the scope of the patents therefore cannot be interpreted as.It should be pointed out that for those of ordinary skill in the art For, without departing from the concept of this application, various modifications and improvements can be made, these belong to the guarantor of the application Protect range.Therefore, the scope of protection shall be subject to the appended claims for the application patent.

Claims

1. a kind of resource allocation methods, comprising:

Obtain the sharing rate of account to be chosen；

Target account is chosen according to the sharing attribute score value；

Determine that the target account is general for the distribution of each resource to be allocated according to the sharing attribute score value of the target account Rate.

2. the method according to claim 1, wherein described choose target account according to the sharing attribute score value Family, comprising:

Choose the account alternately account that the sharing attribute score value is more than point threshold；

Obtain the corresponding influence value of the alternative account；

The alternative account of preset quantity is selected as target account according to the influence value.

3. according to the method described in claim 2, it is characterized in that, the influence value includes that the first influence value and second influence Value；It is described to obtain the corresponding influence value of the alternative account, comprising:

The first influence value of the alternative account is obtained, first influence value is to have friend relation with the alternative account In user, gets resource to be allocated and there are good friend's numbers of sharing operation and the alternative account to get resource to be allocated Number sum；

The second influence value of the alternative account is obtained, second influence value is to have friend relation with the alternative account In user, first influence value is equal to the number of default value.

4. according to the method described in claim 3, it is characterized in that, described select the standby of preset quantity according to the influence value Select account as target account, comprising:

Choose the alternative account that the first influence value is default influence value；

Chosen from the alternative account that first influence value is default influence value the second influence value sort from large to small it is forward The alternative account of preset quantity is as target account.

5. the method according to claim 1, wherein the sharing rate for obtaining account to be chosen, comprising:

For the allocation probability of the account setup resource to be allocated to be chosen；

When it is described resource to be allocated is received according to the allocation probability wait choose account after, obtain point of the account to be chosen Enjoy number；

The sharing rate for obtaining the account to be chosen is the sharing number of the account to be chosen and receives resource to be allocated The ratio of number.

6. the method according to claim 1, wherein before obtaining the sharing rate of account to be chosen, further includes:

Obtain the account in default application；

Choose active account and/or influence power the account account to be chosen as described in the account in the default application.

7. according to the method described in claim 6, it is characterized in that, described determine the account to be chosen according to the sharing rate Sharing attribute score value, comprising:

Obtain behavioural characteristic of the account to be chosen in the default application；

It is input to the sharing rate and the behavioural characteristic of the account to be chosen in neural network, exports the account to be chosen The sharing attribute score value at family.

8. according to the method described in claim 6, it is characterized in that, the condition for meeting active account is included at least with next Kind:

Within a preset time, account is more than publication amount threshold by issuing the quantity of personal information in default application；

Within a preset time, account is more than speech amount threshold by participating in group chat speech number in default application；

Within a preset time, account is more than interaction number by the quantity that default application is interacted with the personal information that good friend issues Measure threshold value；

Within a preset time, account checks that the number of default article is more than to read amount threshold by default application；

Within a preset time, account is more than transmission number by the quantity that default application carries out resource transmission with other interior accounts of application Measure threshold value；

It is more than forwarding amount threshold that account, which forwards the quantity of default article by default application,.

9. according to the method described in claim 6, it is characterized in that, the condition for meeting influence power account is including at least following It is a kind of:

Having the number of friend relation in default application with account is more than good friend's amount threshold；

Within a preset time, good friend is more than that good friend is mutual by the quantity that the personal information of default application publication is interacted to account Dynamic amount threshold；

Belong to default identity type by the identity information that default application gets account；

Belong to default address type by the certificate address information that default application gets account；

In the account for having friend relation in default application with account, more than the occupational information category of good friend's account of preset quantity In default occupation type.

10. the method according to claim 1, wherein the sharing attribute score value according to the target account Determine that the target account is directed to the allocation probability of each resource to be allocated, comprising:

When the sharing attribute score value of the target account is higher, it is determined that the target account is in preset resource to be allocated Allocation probability is higher.

11. according to the method described in claim 10, it is characterized in that, described when the sharing attribute score value of the target account is got over Gao Shi, it is determined that the target account is higher in the allocation probability of preset resource to be allocated, comprising:

Obtain the active budget numerical value of preset resource to be allocated；

The payment number and payment amount of value interval where obtaining the preset resource to be allocated, are calculated described default Resource to be allocated where value interval payment mean value；

It is determined according to the active budget numerical value of the preset resource to be allocated with the payment mean value described preset to be allocated The quantity allotted of resource；

It is described preset wait divide to determine that the target account is got according to the quantity of the target account and the quantity allotted Mean allocation probability with resource；

When the sharing attribute score value of the target account is higher, determine the target account in the preset resource to be allocated Closer to the allocation probability of the mean allocation probability.

12. a kind of resource allocation device, which is characterized in that described device includes:

Share attribute score value determining module, for determining the sharing attribute score value of the account to be chosen according to the sharing rate；

Allocation probability determining module determines the target account for every for the sharing attribute score value according to the target account The allocation probability of a resource to be allocated.

13. device according to claim 10, which is characterized in that the target account selection module is also used to choose described Distributive property score value is more than the account alternately account of point threshold；Obtain the corresponding influence value of the alternative account；According to The influence value selects the alternative account of preset quantity as target account.

14. a kind of computer readable storage medium is stored with computer program, when the computer program is executed by processor, So that the processor is executed such as the step of any one of claims 1 to 11 the method.

15. a kind of computer equipment, including memory and processor, the memory is stored with computer program, the calculating When machine program is executed by the processor, so that the processor is executed such as any one of claims 1 to 11 the method Step.