CN106202904A - A kind of game amount of leading data scheduling method based on channel resource position and device - Google Patents
A kind of game amount of leading data scheduling method based on channel resource position and device Download PDFInfo
- Publication number
- CN106202904A CN106202904A CN201610526060.6A CN201610526060A CN106202904A CN 106202904 A CN106202904 A CN 106202904A CN 201610526060 A CN201610526060 A CN 201610526060A CN 106202904 A CN106202904 A CN 106202904A
- Authority
- CN
- China
- Prior art keywords
- data
- waiting
- channel
- game
- amount
- 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
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16Z—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS, NOT OTHERWISE PROVIDED FOR
- G16Z99/00—Subject matter not provided for in other main groups of this subclass
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the invention discloses a kind of game amount of leading data scheduling method based on channel resource position and device, solve current manual method more subjective, experience is difficult to precipitate, and work amount is big, the accurate and difficult to govern control of coverage, artificial method is subjective, workload is big, and regulative mode is difficult to solidification precipitation；Under shortage is to the pool of the amount of leading and configuring, the manual method very difficult balance multiclass popularization factor suitable waiting of generation, and the resource analysis of the game amount of leading caused technical problem not accurately.Embodiment of the present invention game based on the channel resource position amount of leading data scheduling method includes: acquisition channel attribute data and candidate play operation data；According to channel attribute data operation data of playing with candidate each channel is calculated corresponding channel business revenue data, and according to channel business revenue data with calculate the operation factor data of acquisition according to preset calculation and set up waiting Optimized model function；Algorithm of seeking approximate is asked to determine waiting sequence according to waiting Optimized model function by iteration.
Description
Technical field
The present invention relates to technical field of data processing, particularly relate to a kind of game amount of leading data based on channel resource position row
Phase method and device.
Background technology
Online game development is swift and violent, and substantial amounts of game constantly develops from manufacturer.These game are flat by game
Each channel promotion of platform is in corresponding subject player.In this process, different types of game is played by gaming platform
The amount of leading of family, allows the game of each money reach to operate required player's threshold values, sets up game ecology.On the other hand, the quality ginseng of game
Difference is uneven, and the game of low value consumes substantial amounts of popularization resource, causes the game amount of leading utilization rate low, affects the battalion of gaming platform
Receiving, the game amount of leading is to be through the popularization of certain channel, enters the quantity of the user that game is played.
The most how to game waiting, and the maximization amount of leading utilization rate is a business difficult problem.Tradition generally relies on manually
Empirically determined game waiting, such as promotes the game of business revenue top-20.But there is life cycle in game, the game business revenue meeting being given to
Decline, the best game of business revenue is not necessarily game with the largest potentiality.
The existence of the game that business revenue is best at present has local derviation amount can expand Matthew effect, not only affects the popularization power to new trip
Degree, and finally reduce the long-range business revenue of platform, especially, manual method is more subjective, and experience is difficult to precipitate, and work amount is big, accurate
Really and the difficult to govern control of coverage, artificial method is subjective, and workload is big, and regulative mode is difficult to solidification and precipitates；Lacking the amount of leading
Pool and configuration under, manual method is difficult to balance multiclass and promotes factor and generate suitable waiting, thus result in the game amount of leading
The accurate not technical problem of resource analysis.
Summary of the invention
A kind of based on channel resource position the game amount of the leading data scheduling method of embodiment of the present invention offer and device, solve
Current manual method is more subjective, and experience is difficult to precipitate, and work amount is big, accurately and the difficult to govern control of coverage, artificial method master
Seeing, workload is big, and regulative mode is difficult to solidification precipitation；Under lacking the pool to the amount of leading and configuration, manual method is difficult to flat
The weighing apparatus multiclass popularization factor suitable waiting of generation, and the accurate not technical problem of the resource analysis of the game amount of leading caused.
A kind of based on channel resource position the game amount of the leading data scheduling method that the embodiment of the present invention provides, including:
Acquisition channel attribute data and candidate play operation data；
According to described channel attribute data operation data of playing with described candidate, each described channel is calculated corresponding canal
Road business revenue data, and set up according to described channel business revenue data and the operation factor data obtained according to the calculating of preset calculation
Waiting Optimized model function；
Algorithm of seeking approximate is asked to determine waiting sequence according to described waiting Optimized model function by iteration.
Preferably, described channel attribute data is the amount of the leading data of the resource advertising position corresponding with each channel；
Described candidate operation data of playing includes opening and takes data, load value data, logon data, new player registers amount, newly plays
Family's registration conversion ratio, district take player and hold carrying capacity.
Preferably, according to described channel attribute data and described candidate play operation data to each described channel calculate right
The channel business revenue data answered, and set up waiting Optimized model function according to described channel business revenue data and specifically include:
According to described open take data, described district takes each player that player holds in carrying capacity, described load value data and averagely supplements with money
Amount calculates each resource advertising position business revenue data of described channel according to the first preset formula, described in open and take data by described
The amount of leading data, described new player registers conversion ratio and described district take player and hold the average carrying capacity of holding of single district clothes of carrying capacity according to second
Preset formula carries out calculating acquisition；
The all described resource advertising position business revenue data calculating acquisition are carried out summation and determines described channel business revenue data；
Waiting is set up according to described channel business revenue data and the operation factor data obtained according to the calculating of preset calculation
Optimized model function, described operation factor data is the resource sequence of the described resource advertising position in preset channel；
Wherein, described first preset formula be described resource advertising position business revenue data=described in open and take data × described district clothes
Player holds carrying capacity × each described player averagely amount of supplementing with money；
Described second preset formula be described in open take data=INT (described in the amount of leading data × described new player registers convert
Rate/described single district clothes averagely hold carrying capacity).
Preferably, build according to described channel business revenue data and the operation factor data obtained according to the calculating of preset calculation
Vertical waiting Optimized model function specifically includes:
Calculated the waiting multiformity of described resource advertising position waiting, institute according to the 3rd preset formula by similarity between game
State similarity between game and calculate acquisition by the game characteristic vector of correspondence of playing two-by-two according to the 4th preset formula；
By the fresh degree function read group total of described resource advertising position waiting is obtained waiting freshness, described freshness
Function calculates acquisition by on-line time, current time and the weight factor that game is corresponding according to the 5th preset formula；
Waiting Optimized model letter is set up according to described channel business revenue data, described waiting multiformity and described waiting freshness
Number；
Wherein, described 3rd preset formula be described waiting multiformity=1-(∑ { similarity between described game }/(0.5 ×
Described resource sequence × (described resource sequence-1)))；
Described 4th preset formula be similarity between described game=∑ 1/sqrt (described in one game characteristic vector × another
Described game characteristic vector)；
Described 5th preset formula is described fresh degree function=1/ ((the described current time-described of weight factor described in 1+
On-line time)).
Preferably, waiting sequence is specifically wrapped to ask Algorithm of seeking approximate to determine according to described waiting Optimized model function by iteration
Include:
S1: the described waiting Optimized model function of described waiting sequence is determined assessment feasible solution by Lagrange conversion
Scoring functions, described feasible solution is the most feasible waiting；
S2: according to the described current optimal solution of scoring functions record；
S3: select at least 2 candidate's feasible solutions by rule from described feasible solution, and described candidate's feasible solution is carried out multiple
System intersection operation obtains at least 2 corresponding new feasible solution results；
S4: described new feasible solution result is carried out mutation operation, generates several candidate schemes, and judge described candidate side
Whether the quantity of case reaches preset number of candidates, if it is not, then repeat step S3, until the quantity of described candidate scheme reaches preset
Number of candidates, then perform step S5；
S5: using preset number of candidates described candidate scheme as the set of new feasible solution, and substitute former feasible solution set,
Repeated execution of steps S2 to S5 carries out process iterative computation, until iterations reaches preset iteration predetermined times, determines optimum
Waiting sequence.
A kind of based on channel resource position the game amount of the leading data waiting device that the embodiment of the present invention provides, including:
Data access unit, plays operation data for acquisition channel attribute data and candidate；
Waiting modeling unit, for playing operation data to each described according to described channel attribute data and described candidate
Channel calculates corresponding channel business revenue data, and according to described channel business revenue data with according to the calculating acquisition of preset calculation
Operation factor data sets up waiting Optimized model function；
Waiting signal generating unit, for asking Algorithm of seeking approximate to determine waiting according to described waiting Optimized model function by iteration
Sequence.
Preferably, data access unit specifically includes:
Attribute information gathers subelement, for gathering the amount of the leading data of the resource advertising position corresponding with each channel；
Operation data gathers subelement, takes data, load value data, login number for gathering corresponding opening of playing with candidate
According to, new player registers amount, new player registers conversion ratio, district take player and hold carrying capacity.
Preferably, described waiting modeling unit specifically includes:
Channel business revenue data computation subunit, for according to described in open take data, described district take player hold carrying capacity, described in fill
Each player averagely amount of supplementing with money in Value Data calculates each resource advertising position battalion of described channel according to the first preset formula
Receive data, described in open take data pass through described in the amount of leading data, described new player registers conversion ratio and described district take player and hold carrying capacity
The average carrying capacity of holding of single district clothes carry out calculating according to the second preset formula and obtain；
Business revenue data summation subelement, for suing for peace to all described resource advertising position business revenue data calculating acquisition
Determine described channel business revenue data；
Waiting modeling subelement, is used for according to described channel business revenue data and calculates the fortune obtained according to preset calculation
Battalion's factor data sets up waiting Optimized model function, and described operation factor data is the described resource advertising position in preset channel
Resource sequence；
Wherein, described first preset formula be described resource advertising position business revenue data=described in open and take data × described district clothes
Player holds carrying capacity × each described player averagely amount of supplementing with money；
Described second preset formula be described in open take data=INT (described in the amount of leading data × described new player registers convert
Rate/described single district clothes averagely hold carrying capacity).
Preferably, waiting modeling subelement specifically includes:
Waiting multiformity computing module, wide for calculating described resource by similarity between game according to the 3rd preset formula
Accusing the waiting multiformity of position waiting, between described game, similarity is pre-according to the 4th by the game characteristic vector of correspondence of playing two-by-two
Put formula and calculate acquisition；
Waiting freshness computing module, for by obtaining the fresh degree function read group total of described resource advertising position waiting
Take waiting freshness, described fresh degree function by corresponding on-line time, current time and the weight factor of game according to the 5th
Preset formula calculates and obtains；
Waiting MBM, for according to described channel business revenue data, described waiting multiformity and described waiting freshness
Set up waiting Optimized model function；
Wherein, described 3rd preset formula be described waiting multiformity=1-(∑ { similarity between described game }/(0.5 ×
Described resource sequence × (described resource sequence-1)))；
Described 4th preset formula be similarity between described game=∑ 1/sqrt (described in one game characteristic vector × another
Described game characteristic vector)；
Described 5th preset formula is described fresh degree function=1/ ((the described current time-described of weight factor described in 1+
On-line time)).
Preferably, described waiting signal generating unit specifically includes:
Conversion subelement, for changing really by Lagrange the described waiting Optimized model function of described waiting sequence
Accepted opinion estimates the scoring functions of feasible solution, and described feasible solution is the most feasible waiting；
Record subelement, for according to the described current optimal solution of scoring functions record；
Replicate intersection subelement, for selecting at least 2 candidate's feasible solutions by rule from described feasible solution, and to described
Candidate's feasible solution carries out replicating intersection operation and obtains at least 2 corresponding new feasible solution results；
Mutation operation subelement, for described new feasible solution result is carried out mutation operation, generates several candidate schemes,
And judge whether the quantity of described candidate scheme reaches preset number of candidates, intersect sub single if it is not, the most again trigger described duplication
Unit, until the quantity of described candidate scheme reaches preset number of candidates, then triggers set and substitutes subelement；
Described set substitutes subelement, for using preset number of candidates described candidate scheme as the collection of new feasible solution
Close, and substitute former feasible solution set, repeat to trigger described record subelement, described duplication intersection subelement, described variation successively
Operator unit and described set substitute subelement and carry out process iterative computation, until iterations reaches preset iteration preset time
Number, determines optimum waiting sequence.
As can be seen from the above technical solutions, the embodiment of the present invention has the advantage that
A kind of based on channel resource position the game amount of the leading data scheduling method of embodiment of the present invention offer and device, its
In, the game amount of leading data scheduling method based on channel resource position includes: acquisition channel attribute data and candidate play operation number
According to；According to channel attribute data operation data of playing with candidate each channel is calculated corresponding channel business revenue data, and according to
Channel business revenue data and the operation factor data obtained according to the calculating of preset calculation set up waiting Optimized model function；According to
Waiting Optimized model function asks Algorithm of seeking approximate to determine waiting sequence by iteration.In the present embodiment, by acquisition channel attribute
Data and candidate play operation data, then according to channel attribute data and candidate play operation data each channel is calculated right
The channel business revenue data answered, and build according to channel business revenue data and the operation factor data obtained according to the calculating of preset calculation
Vertical waiting Optimized model function, asks Algorithm of seeking approximate to determine waiting sequence finally according to waiting Optimized model function by iteration,
Achieving without manually game being carried out waiting, solve current manual method more subjective, experience is difficult to precipitate, and work amount
Greatly, the accurate and difficult to govern control of coverage, artificial method is subjective, and workload is big, and regulative mode is difficult to solidification precipitation；Right lacking
Under the pool of the amount of leading and configuration, manual method is difficult to balance multiclass and promotes the factor suitable waiting of generation, and the game caused is led
The accurate not technical problem of the resource analysis of amount, and in the attribute data of each channel and the operation data of candidate's game
(operation data is supplemented with money, logs in and opens the information such as clothes) builds waiting model, when building waiting model, except take into account game
Life cycle and ripe game Decline State outside, it is also contemplated that cultivation potentiality new game to its preferably amount of leading, canal
Road and resource-niche are being led difference and the change of gauge mould and user characteristics, are being made full use of the amount of leading of resource-niche to maximize total battalion
Receive, allow result variation meet player's these factors of all kinds of demand with maximum.
Accompanying drawing explanation
In order to be illustrated more clearly that the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
In having technology to describe, the required accompanying drawing used is briefly described, it should be apparent that, the accompanying drawing in describing below is only this
Some embodiments of invention, for those of ordinary skill in the art, on the premise of not paying creative work, also may be used
To obtain other accompanying drawing according to these accompanying drawings.
One of a kind of based on channel resource position the game amount of the leading data scheduling method that Fig. 1 provides for the embodiment of the present invention
The schematic flow sheet of embodiment；
Another of a kind of based on channel resource position the game amount of the leading data scheduling method that Fig. 2 provides for the embodiment of the present invention
The schematic flow sheet of individual embodiment；
One of a kind of based on channel resource position the game amount of the leading data waiting device that Fig. 3 provides for the embodiment of the present invention
The structural representation of embodiment；
Another of a kind of based on channel resource position the game amount of the leading data waiting device that Fig. 4 provides for the embodiment of the present invention
The structural representation of individual embodiment；
Fig. 5 (a) is to the block schematic illustration that Fig. 5 (c) is Fig. 2 application examples.
Detailed description of the invention
A kind of based on channel resource position the game amount of the leading data scheduling method of embodiment of the present invention offer and device, solve
Current manual method is more subjective, and experience is difficult to precipitate, and work amount is big, accurately and the difficult to govern control of coverage, artificial method master
Seeing, workload is big, and regulative mode is difficult to solidification precipitation；Under lacking the pool to the amount of leading and configuration, manual method is difficult to flat
The weighing apparatus multiclass popularization factor suitable waiting of generation, and the accurate not technical problem of the resource analysis of the game amount of leading caused.
For making the goal of the invention of the present invention, feature, the advantage can be the most obvious and understandable, below in conjunction with the present invention
Accompanying drawing in embodiment, is clearly and completely described the technical scheme in the embodiment of the present invention, it is clear that disclosed below
Embodiment be only a part of embodiment of the present invention, and not all embodiment.Based on the embodiment in the present invention, this area
All other embodiments that those of ordinary skill is obtained under not making creative work premise, broadly fall into present invention protection
Scope.
Refer to Fig. 1, a kind of based on channel resource position the game amount of the leading data scheduling method that the embodiment of the present invention provides
An embodiment include:
101, acquisition channel attribute data and candidate play operation data；
In the present embodiment, when needs carry out waiting by the game amount of leading to game, it is necessary first to acquisition channel attribute number
Play operation data according to candidate.
102, according to channel attribute data operation data of playing with candidate, each channel is calculated corresponding channel business revenue number
According to, and set up waiting Optimized model according to channel business revenue data and the operation factor data obtained according to the calculating of preset calculation
Function；
After acquisition channel attribute data and candidate play operation data, need to swim according to channel attribute data and candidate
Play operation data calculates corresponding channel business revenue data to each channel, and according to channel business revenue data with according to preset calculating side
Formula calculates the operation factor data obtained and sets up waiting Optimized model function.
103, Algorithm of seeking approximate is asked to determine waiting sequence according to waiting Optimized model function by iteration.
When according to channel attribute data operation data of playing with candidate, each channel being calculated corresponding channel business revenue data,
And set up waiting Optimized model letter according to channel business revenue data and the operation factor data obtained according to the calculating of preset calculation
After number, need to ask Algorithm of seeking approximate to determine waiting sequence according to waiting Optimized model function by iteration.
In the present embodiment, played operation data by acquisition channel attribute data and candidate, then according to channel attribute number
According to operation data of playing with candidate each channel calculated corresponding channel business revenue data, and according to channel business revenue data and according to
Preset calculation calculates the operation factor data obtained and sets up waiting Optimized model function, finally according to waiting Optimized model letter
Number asks Algorithm of seeking approximate to determine waiting sequence by iteration, it is achieved that without manually game being carried out waiting, solve current
Manual method is more subjective, and experience is difficult to precipitate, and work amount is big, the accurate and difficult to govern control of coverage, and artificial method is subjective, workload
Greatly, and regulative mode be difficult to solidification precipitation；Under lacking the pool to the amount of leading and configuration, manual method is difficult to balance multiclass and promotes
The factor suitable waiting of generation, and the accurate not technical problem of the resource analysis of the game amount of leading caused.
The above is that the process to the game amount of leading data scheduling method based on channel resource position is described in detail, below
Detailed process will be described in detail, refer to Fig. 2, a kind of based on channel resource position the trip that the embodiment of the present invention provides
Another embodiment of the play amount of leading data scheduling method includes:
201, acquisition channel attribute data and candidate play operation data；
In the present embodiment, when needs carry out waiting by the game amount of leading to game, it is necessary first to acquisition channel attribute number
Playing operation data according to candidate, channel attribute data is the amount of the leading data of the resource advertising position corresponding with each channel, Hou Xuanyou
Play operation data include opening take data, load value data, logon data, new player registers amount, new player registers conversion ratio, district clothes play
Family holds carrying capacity.
202, according to open take data, district takes each player averagely amount of supplementing with money that player holds in carrying capacity, load value data according to the
One preset formula calculates each resource advertising position business revenue data of channel；
After acquisition channel attribute data and candidate play operation data, needs basis is opened and is taken data, district takes player and holds
Each player averagely amount of supplementing with money in carrying capacity, load value data calculates each resource advertising position of channel according to the first preset formula
Business revenue data, open and take data and take player by the amount of leading data, new player registers conversion ratio and district to hold the single district clothes of carrying capacity average
Hold carrying capacity to carry out calculating acquisition according to the second preset formula.
First preset formula is that resource advertising position business revenue data=open take data × district and take player and hold carrying capacity × each player
The averagely amount of supplementing with money；
Second preset formula takes the data=INT (amount of leading data × new player registers conversion ratio/average appearance of single district clothes for opening
Carrying capacity).
203, all resource advertising positions business revenue data calculating acquisition are carried out summation and determine channel business revenue data；
When according to open take data, district takes each player averagely amount of supplementing with money that player holds in carrying capacity, load value data according to first
After preset formula calculates each resource advertising position business revenue data of channel, need calculating all resource advertising positions obtained
Business revenue data carry out summation and determine channel business revenue data.
204, calculated the waiting multiformity of resource advertising position waiting, trip according to the 3rd preset formula by similarity between game
Between play, similarity calculates acquisition by the game characteristic vector of correspondence of playing two-by-two according to the 4th preset formula；
After all resource advertising positions business revenue data calculating acquisition being carried out summation and determines channel business revenue data, need
Calculated the waiting multiformity of resource advertising position waiting according to the 3rd preset formula by similarity between game, between game, similarity is led to
Acquisition is calculated according to the 4th preset formula after the game characteristic vector playing corresponding two-by-two.
3rd preset formula is that waiting multiformity=1-(∑ { similarity between game }/(sequence of 0.5 × resource × (arrange by resource
Sequence-1))).
4th preset formula for game between similarity=∑ 1/sqrt (game characteristic vector × another game characteristic to
Amount).
205, by the fresh degree function read group total of resource advertising position waiting is obtained waiting freshness, fresh degree function
On-line time, current time and the weight factor corresponding by game calculate acquisition according to the 5th preset formula；
, play according to the waiting multiformity of the 3rd preset formula calculating resource advertising position waiting when passing through similarity between game
Between after similarity calculated according to the 4th preset formula by the game characteristic vector of correspondence of playing two-by-two and obtains, need by right
The fresh degree function read group total of resource advertising position waiting obtains waiting freshness, and fresh degree function is by corresponding the reaching the standard grade of game
Time, current time and weight factor calculate according to the 5th preset formula and obtain.
5th preset formula is fresh degree function=1/ (1+ weight factor (current time-on-line time)).
206, waiting Optimized model function is set up according to channel business revenue data, waiting multiformity and waiting freshness；
Step 202, to after 205, needs to set up waiting according to channel business revenue data, waiting multiformity and waiting freshness
Optimized model function.
207, the waiting Optimized model function to waiting sequence passes through the Lagrangian marking changed and determine assessment feasible solution
Function, feasible solution is the most feasible waiting；
After setting up waiting Optimized model function according to channel business revenue data, waiting multiformity and waiting freshness, need
The waiting Optimized model function of waiting sequence is determined by Lagrange conversion the scoring functions of assessment feasible solution, feasible solution
For the most feasible waiting.
208, according to the current optimal solution of scoring functions record；
When the marking letter that the waiting Optimized model function of waiting sequence is determined by Lagrange conversion assessment feasible solution
Number, after feasible solution is the most feasible waiting, needs according to the current optimal solution of scoring functions record.
209, from feasible solution, select at least 2 candidate's feasible solutions by rule, and carry out candidate's feasible solution replicating intersection
Operation obtains at least 2 corresponding new feasible solution results；
After according to the current optimal solution of scoring functions record, need to select at least 2 candidates by rule from feasible solution
Feasible solution, and carry out candidate's feasible solution replicating at least 2 new feasible solution results that intersection operation acquisition is corresponding.
210, new feasible solution result is carried out mutation operation, generate several candidate schemes, and judge the number of candidate scheme
Whether amount reaches preset number of candidates, if it is not, then repeat step 209, until the quantity of candidate scheme reaches preset number of candidates,
Then perform step 211；
When selecting at least 2 candidate's feasible solutions by rule from feasible solution, and carry out candidate's feasible solution replicating intersection behaviour
After making to obtain at least 2 corresponding new feasible solution results, need new feasible solution result is carried out mutation operation, generate several
Candidate scheme, and judge whether the quantity of candidate scheme reaches preset number of candidates, if it is not, then repeat step 209, until candidate
The quantity of scheme reaches preset number of candidates, then perform step 211.
211, using preset number of candidates candidate scheme as the set of new feasible solution, and former feasible solution set is substituted, weight
Perform step 208 again and carry out process iterative computation to 211, until iterations reaches preset iteration predetermined times, determine optimum
Waiting sequence.
When the quantity of candidate scheme reaches preset number of candidates, then using preset number of candidates candidate scheme as the most feasible
The set solved, and substitute former feasible solution set, repeated execution of steps 208 to 211 carries out process iterative computation, until iterations
Reach preset iteration predetermined times, determine optimum waiting sequence.
Being described with a concrete application scenarios below, as shown in Fig. 5 (a) to (c), application examples includes:
By the amount of the leading situation of channel, and the operation feature of game, build model and predict optimal waiting exactly, allow and lead
The utilization rate of amount is maximum.This system has incorporated and has supported YY QQGame waiting system, and power-assisted business datumization is runed, and produces huge
Big benefit.
It should be noted that as shown in Fig. 5 (a), the game amount of leading data scheduling method based on channel resource position can be
Device in conjunction with Fig. 3 and Fig. 4 embodiment is described, and device is data access unit 501 respectively, waiting modeling unit 502, and
Waiting signal generating unit 503；The waiting result generated is docked in operation system.
Data access unit 501 is responsible for accessing two class data, including the attribute data of each channel, and promotes game candidate
Operation data, as shown in Fig. 5 (b), the attribute data of channel is obtained by gaming platform, be responsible for collect channel each promote
The amount of the leading quantity of resource advertising position.Gaming platform has multiple support channels, such as QQGame channel, market channel, the inside amount of leading canal
Road etc.；Each channel has multiple advertisement position, such as at QQGame channel, has focus chart advertisement position, potentiality trip advertisement position etc., place
Reason unit 504 represents attribute information harvester.The operation data of candidate's game includes supplementing with money, logs in and opens the information such as clothes.
Processing unit 505 represents and supplements behavior catcher with money, and unit 506 represents login behavior catcher, and unit 507 represents
Open and take situation catcher.Specifically, data are included in statistics and play the newest player registers amount of given game in first 7 days day, 7 days
The registration conversion ratio of new player；Statistics plays first 7 days new player's amounts of supplementing with money in given game day and supplements number with money；Statistics rises day
First 7 days game new district clothes are average hold carry player's lower limit amount (the most each district takes the minimum player's amount holding to carry, if less than this lower limit,
Player causes district's clothes operation ratio the lowest very little, cannot operate)；
These data are basis and the information sources of game waiting.
Channel attribute data and candidate based on accessing are played operation data, and this module is responsible for building waiting model.Most preferably
Waiting need to consider following operation demand:
Make full use of the amount of leading of resource-niche, maximize total business revenue；
Allowing result variation, maximum meet all kinds of demand of player, such as preference MOBA, ORPG plays；
Consider the life cycle of game, and the Decline State of ripe game, cultivate potentiality and newly swim and to its more preferable amount of leading；
Consider that channel and resource-niche are leading difference and the change of gauge mould and user characteristics；
Based on above operation demand, this module is converted into a permutation and combination problem this waiting problem；For given
The resource-niche of channel, from n candidate's game, sequentially resource-niche is put in extraction m；Allow ranking results can maximize target letter
Number, and the resource-niche promoted according to result.In view of the domain feature difference of each channel, channel is modeled by this module respectively,
A model set up by the most each channel；Waiting result allows and improves multiformity, Income Maximum between channel in channel.
Waiting object function is to maximize total revenue；In given time interval, the business revenue of channel is equal to each resource
The business revenue summation of position.
Business revenue=the ∑ of channel Rr∈R{ Revenue (r) }, wherein r refers to resource-niche；----formula 1
The business revenue of resource-niche be the player by this resource-niche amount of leading in gaming supplement summation with money.Specifically, resource-niche
By giving the game delivery user specified, these users are converted into the registration player of game according to a certain percentage；These player's quilts
Arrange in each district clothes all played；Player does electricity skill activity (level like a white silk) in Game Zone takes, and produces and supplements consumption with money；This consumption
Income for weigh resource-niche the amount of leading be worth.Briefly, can be write as equation below:
Revenue (r)=#Server (r) × Server_size (g) × ARPS (g)----formula 2
Wherein #Server (r) represents opening of game corresponding to the resource-niche amount of leading and takes number, Server_size (g) game representation
The player of district's clothes holds carrying capacity；ARPS (g) represents single district and takes the averagely amount of supplementing with money of each player.Such as resource-niche r leads to game g
Sending player to measure, this tittle enough opens 4 district's clothes, and each district clothes hold load 3000 people, supplement 1.2 yuan per capita with money, then the resource-niche r amount of leading
It is worth as 4*3000*1.2 unit.
On the other hand, there is the lowest number lower limit requirement opening clothes in each district clothes；If the amount of leading is less than this lower limit, this amount of leading
Cannot form district's clothes, i.e. this amount of leading may be wasted, or is guided in the old liberated area clothes that some value are on the low side.Open take number and
A kind of stage relation is there is between the amount of leading quantity；
#Server (r)=INT (Users (r) × %P (g)/Server_size (g))----formula 3
Wherein Users (r) represent resource-niche r give game the amount of leading quantity, the registration conversion ratio of %P (g) game representation g,
Server_size (g) represents the average of single district clothes and holds manned number；INT (.) example represents bracket function；Such as resource-niche r gives trip
Play g delivery 100,000 customer volume, these users have the 20% player's amount that can be converted into registration player, i.e. 2w；Each district clothes are opened under clothes
It is limited to 3000 people；So these amounts of leading can open 6 new district clothes；It is relatively low that remaining 2000 players may be guided to some value
Old liberated area clothes in；Game in view of new district clothes takes far above old liberated area, and this scattered amount of leading of 2000 cannot form district's clothes
The amount of leading, produced value is less and can ignore；
Aggregative formula 1, formula 2, formula 3, we can obtain the object function that waiting is to be optimized.On the other hand, game
Waiting needs to consider following two operation factor, the multiformity of the type of play user to be met that each resource-niche is promoted in channel
Demand and player's pursuit to INVENTIONThe novel game.We are to the two operation factor modeling below；
Note game of each resource-niche in channel is ordered as π, and the multiformity of resource-niche waiting can be weighed by formula 4
Amount:
Diversity (π)=1-(∑i,j∈π,i≠j{Similarity(i,j)}/(0.5×|π|×(|π|-1)))----
Formula 4
Wherein Similarity (. .) refer to game between similarity；
Specifically, using multiple feature to portray game i, the similarity between two game can be used between characteristic of correspondence
Similarity is weighed, i.e.
Similarity (i, j)=∑u∈N(i)∩N(j)1/sqrt (| N (i) | × | N (j) |)----formula 5
The wherein characteristic vector of N (i) game representation i；
Such as game i, feature (such as title, theme, type etc.) can be used to weigh its characteristic, characteristic of correspondence to
Amount such as [1,2,3]；Similarly, for game j, as belonged to theme 2, characteristic vector [3,2,1] can be expressed as；So this two
The similarity of money game is 1/ (3 × 3)；
From formula it can be seen that the similar features of two game correspondence is the most, (i, j) value is the biggest for Similarity；
On the other hand, the freshness of waiting is defined as each money game freshness sum；I.e.
Novelty (π)=∑g∈πDamping (g)----formula 6
Wherein, g is game to be promoted in waiting, and Damping (.) is fresh degree function；
Considering that game exists life cycle, on-line time is the longest, and the freshness of game is the lowest；Definition fresh degree function is such as
Under；
Damping (g)=1/ (1+ α (T (g)-t (g)))----formula 7
The wherein on-line time of t (g) game representation, T (g) represents the current time, and α is weight factor；
Aggregative formula 1-7, the majorized function of game waiting is as follows:
max{Revenue(π)}
s.t.max{Diversity(π)}
max{Novelty(π)}；π～(0,1)----formula 8
Wherein π is the waiting sequence needing to ask；
As shown in Fig. 5 (c), the process that waiting sequence generates is as follows:
This module is responsible for solving waiting Optimized model, generates corresponding waiting result.Majorized function in view of waiting
(i.e. formula 8) is non-differentiability, can not lead, and the function of non-convex does not has the explicit solution of mathematic(al) representation；And the asking of non-explicit solution
Solution is also NP-hard on computation complexity.
In order to solve this problem, this module proposes a kind of iteration and seeks the scheme of approximate solution, and iterative process can be asymptotic
Convergence, and to a certain degree avoid local optimum trap.Specifically, iteration thinking is as follows, concrete operating process.
Based on constraints, heuristic rules is utilized to generate initial feasible solution；
Utilize sequence pattern operation to generate new feasible solution, and determine whether to accept this solution according to risk function；
Guide with object function, find optimal solution iteratively；
Specifically, changing according to Lagrange, object function is as follows:
F (Π)=RevenueT(Π)+θ1·Diversity(Π)+θ2·Novelty(Π)
----formula 9
The scoring functions of assessment feasible solution is:
Wherein η is predetermined risk threshold values；
Above iterative process can be restrained, and convergence gradient is as follows:
Wherein π represents waiting sequence；(π t) represents the number of string in sequence π present in t iteration to m；^f (π) table
Show the average fitness of sequence π string in t iteration colony；λ represents the length of string；pcRepresent emergence pattern exchange behaviour between string
The probability made；pmRefer to the probability of a string emergence pattern mutation operation；δ (π) represents that the definition of sequence π is long；O (π) represents sequence
The rank of row π.Can be proved that this iterative process restrains by formula 11, and maximum of probability approaches global optimum.
Solved by iterative algorithm, the waiting sequence of optimum can be obtained；The sequence results generated is docked in operation system,
Instruct the game waiting of operation.
The amount of leading attribute according to channel and the operation feature of game, automatically generate the most preferably amount of the leading waiting sequence of game.Should
Waiting result considers multiclass operation factor, including multiformity and the novelty of waiting game；Leading of energy maximum resource position
Amount utilization rate.
Using AB to test the evaluating system performance of multiple months, wherein A group is the waiting strategy of model, and B group is artificial rule
The waiting strategy of (according to nearest 7 days business revenues of game), the tactics of the game quantity of two groups of waitings is consistent, and contemporaneity is promoted.Example
Such as certain channel given, having two to promote resource-niches, wherein the waiting strategy of first resource-niche uses model A group, second
The waiting strategy of resource-niche uses artificial B group；These two groups of resource-niches have different renewal frequencies and strategy；By statistics by this
The income that channel resource position is corresponding, for weighing two kinds of tactful qualities.Evaluation metrics is each resource-niche new registration in 15 days
The cost per capita of user；Wherein spending per capita in October, 2015 and risen to 89 yuan by 78 yuan, November is risen to by 91 yuan
105 yuan；This project has incorporated in YY waiting system, produces huge value.
In the present embodiment, played operation data by acquisition channel attribute data and candidate, then according to channel attribute number
According to operation data of playing with candidate each channel calculated corresponding channel business revenue data, and according to channel business revenue data and according to
Preset calculation calculates the operation factor data obtained and sets up waiting Optimized model function, finally according to waiting Optimized model letter
Number asks Algorithm of seeking approximate to determine waiting sequence by iteration, it is achieved that without manually game being carried out waiting, solve current
Manual method is more subjective, and experience is difficult to precipitate, and work amount is big, the accurate and difficult to govern control of coverage, and artificial method is subjective, workload
Greatly, and regulative mode be difficult to solidification precipitation；Under lacking the pool to the amount of leading and configuration, manual method is difficult to balance multiclass and promotes
The factor suitable waiting of generation, and the accurate not technical problem of the resource analysis of the game amount of leading caused, and in each canal
The operation data (operation data is supplemented with money, logs in and opens the information such as clothes) of the attribute data in road and candidate's game builds waiting model,
When building waiting model, in addition to the Decline State of the life cycle and ripe game that take into account game, it is also contemplated that
Cultivate potentiality new game and to its preferably the amount of leading, channel and resource-niche lead the difference of gauge mould and user characteristics and change,
Make full use of the amount of leading of resource-niche with maximize total business revenue, allow result variation with maximum meet all kinds of demand of player these because of
Element.
Refer to Fig. 3, a kind of based on channel resource position the game amount of the leading data waiting dress provided in the embodiment of the present invention
The embodiment put includes:
Data access unit 301, plays operation data for acquisition channel attribute data and candidate；
Waiting modeling unit 302, for calculating each channel according to channel attribute data and candidate operation data of playing
Corresponding channel business revenue data, and according to channel business revenue data with calculate the operation factor data obtained according to preset calculation
Set up waiting Optimized model function；
Waiting signal generating unit 303, for asking Algorithm of seeking approximate to determine waiting according to waiting Optimized model function by iteration
Sequence.
In the present embodiment, played operation data by data access unit 301 acquisition channel attribute data and candidate, then
Waiting modeling unit 302 calculates corresponding channel business revenue according to channel attribute data operation data of playing with candidate to each channel
Data, and set up waiting optimization mould according to channel business revenue data and the operation factor data obtained according to the calculating of preset calculation
Type function, last waiting signal generating unit 303 asks Algorithm of seeking approximate to determine waiting sequence according to waiting Optimized model function by iteration
Row, it is achieved that without manually game being carried out waiting, solving current manual method more subjective, experience is difficult to precipitate, and work
Amount is big, the accurate and difficult to govern control of coverage, and artificial method is subjective, and workload is big, and regulative mode is difficult to solidification precipitation；Lacking
To under the pool of the amount of leading and configuration, manual method is difficult to balance multiclass and promotes factor and generate suitable waiting, and the game caused
The accurate not technical problem of the resource analysis of the amount of leading.
The above is that each unit to the game amount of leading data waiting device based on channel resource position is described in detail, under
Sub-unit is described in detail by face, refers to Fig. 4, and provide in the embodiment of the present invention is a kind of based on channel resource position
Another embodiment of the game amount of leading data waiting device includes:
Data access unit 401, plays operation data for acquisition channel attribute data and candidate；
Data access unit 401 specifically includes:
Attribute information gathers subelement 4011, for gathering the amount of the leading data of the resource advertising position corresponding with each channel；
Operation data gathers subelement 4012, takes data, load value data, login for gathering corresponding opening of playing with candidate
Data, new player registers amount, new player registers conversion ratio, district take player and hold carrying capacity.
Waiting modeling unit 402, for calculating each channel according to channel attribute data and candidate operation data of playing
Corresponding channel business revenue data, and according to channel business revenue data with calculate the operation factor data obtained according to preset calculation
Set up waiting Optimized model function；
Waiting modeling unit 402 specifically includes:
Channel business revenue data computation subunit 4021, for according to open take data, district takes player and holds in carrying capacity, load value data
Each player averagely amount of supplementing with money calculate each resource advertising position business revenue data of channel according to the first preset formula, open and take number
Hold the single district clothes of carrying capacity according to taking player by the amount of leading data, new player registers conversion ratio and district average to hold carrying capacity pre-according to second
Put formula to carry out calculating acquisition；
Business revenue data summation subelement 4022, for suing for peace to all resource advertising positions business revenue data calculating acquisition
Determine channel business revenue data；
Waiting modeling subelement 4023, is used for according to channel business revenue data and calculates the fortune obtained according to preset calculation
Battalion's factor data sets up waiting Optimized model function, and operation factor data is the resource row of the resource advertising position in preset channel
Sequence；
Wherein, the first preset formula is that resource advertising position business revenue data=open take data × district and take player and hold carrying capacity × each
Player's averagely amount of supplementing with money；
Second preset formula takes the data=INT (amount of leading data × new player registers conversion ratio/average appearance of single district clothes for opening
Carrying capacity).
Waiting modeling subelement 4023 specifically includes:
Waiting multiformity computing module 4023a, for calculating resource by similarity between game according to the 3rd preset formula
The waiting multiformity of advertisement position waiting, between game, similarity is preset according to the 4th by the game characteristic vector of correspondence of playing two-by-two
Formula calculates and obtains；
Waiting freshness computing module 4023b, for by the fresh degree function read group total to resource advertising position waiting
Obtaining waiting freshness, fresh degree function is pre-according to the 5th by on-line time, current time and the weight factor that game is corresponding
Put formula and calculate acquisition；
Waiting MBM 4023c, for setting up waiting according to channel business revenue data, waiting multiformity and waiting freshness
Optimized model function；
Wherein, the 3rd preset formula be waiting multiformity=1-(∑ { similarity between game }/(0.5 × resource sequence ×
(resource sequence-1)))；
4th preset formula for game between similarity=∑ 1/sqrt (game characteristic vector × another game characteristic to
Amount)；
5th preset formula is fresh degree function=1/ (1+ weight factor (current time-on-line time)).
Waiting signal generating unit 403, for asking Algorithm of seeking approximate to determine waiting according to waiting Optimized model function by iteration
Sequence.
Waiting signal generating unit 403 specifically includes:
Conversion subelement 4031, is commented for being determined the waiting Optimized model function of waiting sequence by Lagrange conversion
Estimating the scoring functions of feasible solution, feasible solution is the most feasible waiting；
Record subelement 4032, for according to the current optimal solution of scoring functions record；
Replicate intersection subelement 4033, for selecting at least 2 candidate's feasible solutions by rule from feasible solution, and to candidate
Feasible solution carries out replicating intersection operation and obtains at least 2 corresponding new feasible solution results；
Mutation operation subelement 4034, for new feasible solution result is carried out mutation operation, generates several candidate schemes,
And judge whether the quantity of candidate scheme reaches preset number of candidates, replicate intersection subelement 4033 if it is not, the most again trigger, directly
Quantity to candidate scheme reaches preset number of candidates, then trigger set and substitute subelement 4035；
Set substitutes subelement 4035, for using preset number of candidates candidate scheme as the set of new feasible solution, and
Substitute former feasible solution set, repeat trigger recording subelement 4032, duplication intersection subelement 4033, mutation operation subelement successively
4034 substitute subelement 4035 with set carries out process iterative computation, until iterations reaches preset iteration predetermined times, really
Fixed optimum waiting sequence.
In the present embodiment, played operation data by data access unit 401 acquisition channel attribute data and candidate, then
Waiting modeling unit 402 calculates corresponding channel business revenue according to channel attribute data operation data of playing with candidate to each channel
Data, and set up waiting optimization mould according to channel business revenue data and the operation factor data obtained according to the calculating of preset calculation
Type function, last waiting signal generating unit 403 asks Algorithm of seeking approximate to determine waiting sequence according to waiting Optimized model function by iteration
Row, it is achieved that without manually game being carried out waiting, solving current manual method more subjective, experience is difficult to precipitate, and work
Amount is big, the accurate and difficult to govern control of coverage, and artificial method is subjective, and workload is big, and regulative mode is difficult to solidification precipitation；Lacking
To under the pool of the amount of leading and configuration, manual method is difficult to balance multiclass and promotes factor and generate suitable waiting, and the game caused
The accurate not technical problem of the resource analysis of the amount of leading, and in the attribute data of each channel and the operation data of candidate's game
(operation data is supplemented with money, logs in and opens the information such as clothes) builds waiting model, when building waiting model, except take into account game
Life cycle and ripe game Decline State outside, it is also contemplated that cultivation potentiality new game to its preferably amount of leading, canal
Road and resource-niche are being led difference and the change of gauge mould and user characteristics, are being made full use of the amount of leading of resource-niche to maximize total battalion
Receive, allow result variation meet player's these factors of all kinds of demand with maximum.
Those skilled in the art is it can be understood that arrive, for convenience and simplicity of description, and the system of foregoing description,
The specific works process of device and unit, is referred to the corresponding process in preceding method embodiment, does not repeats them here.
In several embodiments provided herein, it should be understood that disclosed system, apparatus and method are permissible
Realize by another way.Such as, device embodiment described above is only schematically, such as, and described unit
Dividing, be only a kind of logic function and divide, actual can have other dividing mode, the most multiple unit or assembly when realizing
Can in conjunction with or be desirably integrated into another system, or some features can be ignored, or does not performs.Another point, shown or
The coupling each other discussed or direct-coupling or communication connection can be the indirect couplings by some interfaces, device or unit
Close or communication connection, can be electrical, machinery or other form.
The described unit illustrated as separating component can be or may not be physically separate, shows as unit
The parts shown can be or may not be physical location, i.e. may be located at a place, or can also be distributed to multiple
On NE.Some or all of unit therein can be selected according to the actual needs to realize the mesh of the present embodiment scheme
's.
It addition, each functional unit in each embodiment of the present invention can be integrated in a processing unit, it is also possible to
It is that unit is individually physically present, it is also possible to two or more unit are integrated in a unit.Above-mentioned integrated list
Unit both can realize to use the form of hardware, it would however also be possible to employ the form of SFU software functional unit realizes.
If described integrated unit realizes and as independent production marketing or use using the form of SFU software functional unit
Time, can be stored in a computer read/write memory medium.Based on such understanding, technical scheme is substantially
The part that in other words prior art contributed or this technical scheme completely or partially can be with the form of software product
Embodying, this computer software product is stored in a storage medium, including some instructions with so that a computer
Equipment (can be personal computer, server, or the network equipment etc.) performs the complete of method described in each embodiment of the present invention
Portion or part steps.And aforesaid storage medium includes: USB flash disk, portable hard drive, read only memory (ROM, Read-Only
Memory), random access memory (RAM, Random Access Memory), magnetic disc or CD etc. are various can store journey
The medium of sequence code.
The above, above example only in order to technical scheme to be described, is not intended to limit；Although with reference to front
State embodiment the present invention has been described in detail, it will be understood by those within the art that: it still can be to front
State the technical scheme described in each embodiment to modify, or wherein portion of techniques feature is carried out equivalent；And these
Amendment or replacement, do not make the essence of appropriate technical solution depart from the spirit and scope of various embodiments of the present invention technical scheme.
Claims (10)
1. the game amount of leading a data scheduling method based on channel resource position, it is characterised in that including:
Acquisition channel attribute data and candidate play operation data；
Each described channel is calculated corresponding channel seek according to described channel attribute data operation data of playing with described candidate
Receive data, and set up waiting according to described channel business revenue data and the operation factor data obtained according to the calculating of preset calculation
Optimized model function；
Algorithm of seeking approximate is asked to determine waiting sequence according to described waiting Optimized model function by iteration.
The game amount of leading data scheduling method based on channel resource position the most according to claim 1, it is characterised in that described
Channel attribute data is the amount of the leading data of the resource advertising position corresponding with each channel；
Described candidate play operation data include opening take data, load value data, logon data, new player registers amount, new player note
Volume conversion ratio, district take player and hold carrying capacity.
The game amount of leading data scheduling method based on channel resource position the most according to claim 2, it is characterised in that according to
Described channel attribute data operation data of playing with described candidate calculates corresponding channel business revenue data to each described channel, and
Set up waiting Optimized model function according to described channel business revenue data to specifically include:
According to described open take data, described district take each player averagely amount of supplementing with money that player holds in carrying capacity, described load value data by
Calculate each resource advertising position business revenue data of described channel according to the first preset formula, described in open take data pass through described in the amount of leading
Data, described new player registers conversion ratio and described district take player and hold the single district clothes of carrying capacity average to hold carrying capacity preset according to second
Formula carries out calculating acquisition；
The all described resource advertising position business revenue data calculating acquisition are carried out summation and determines described channel business revenue data；
Waiting optimization is set up according to described channel business revenue data and the operation factor data obtained according to the calculating of preset calculation
Pattern function, described operation factor data is the resource sequence of the described resource advertising position in preset channel；
Wherein, described first preset formula be described resource advertising position business revenue data=described in open and take data × described district and take player
Hold carrying capacity × each described player averagely amount of supplementing with money；
Described second preset formula be described in open take data=INT (described in the amount of leading data × described new player registers conversion ratio/institute
State single district clothes and averagely hold carrying capacity).
The game amount of leading data scheduling method based on channel resource position the most according to claim 3, it is characterised in that according to
Described channel business revenue data and the operation factor data obtained according to the calculating of preset calculation set up waiting Optimized model function
Specifically include:
Calculated the waiting multiformity of described resource advertising position waiting, described trip according to the 3rd preset formula by similarity between game
Between play, similarity calculates acquisition by the game characteristic vector of correspondence of playing two-by-two according to the 4th preset formula；
By the fresh degree function read group total of described resource advertising position waiting is obtained waiting freshness, described fresh degree function
On-line time, current time and the weight factor corresponding by game calculate acquisition according to the 5th preset formula；
Waiting Optimized model function is set up according to described channel business revenue data, described waiting multiformity and described waiting freshness；
Wherein, described 3rd preset formula be described waiting multiformity=1-(∑ { similarity between described game }/(0.5 × described
Resource sequence × (described resource sequence-1)))；
Described 4th preset formula be similarity between described game=∑ 1/sqrt (described in one game characteristic vector × another described in
Game characteristic vector)；
Described 5th preset formula be described fresh degree function=1/ (weight factor described in 1+ (described current time-described in reach the standard grade
Time)).
The game amount of leading data scheduling method based on channel resource position the most according to claim 3, it is characterised in that according to
Described waiting Optimized model function asks Algorithm of seeking approximate to determine by iteration, and waiting sequence specifically includes:
S1: the described waiting Optimized model function of described waiting sequence is determined beating of assessment feasible solution by Lagrange conversion
Dividing function, described feasible solution is the most feasible waiting；
S2: according to the described current optimal solution of scoring functions record；
S3: select at least 2 candidate's feasible solutions by rule from described feasible solution, and carry out described candidate's feasible solution replicating friendship
Fork operation obtains at least 2 corresponding new feasible solution results；
S4: described new feasible solution result is carried out mutation operation, generates several candidate schemes, and judge described candidate scheme
Whether quantity reaches preset number of candidates, if it is not, then repeat step S3, until the quantity of described candidate scheme reaches preset candidate
Quantity, then perform step S5；
S5: using preset number of candidates described candidate scheme as the set of new feasible solution, and substitute former feasible solution set, repeat
Perform step S2 to S5 and carry out process iterative computation, until iterations reaches preset iteration predetermined times, determine optimum waiting
Sequence.
6. the game amount of leading a data waiting device based on channel resource position, it is characterised in that including:
Data access unit, plays operation data for acquisition channel attribute data and candidate；
Waiting modeling unit, for playing operation data to each described channel according to described channel attribute data and described candidate
Calculate corresponding channel business revenue data, and according to described channel business revenue data with calculate the operation obtained according to preset calculation
Factor data sets up waiting Optimized model function；
Waiting signal generating unit, for asking Algorithm of seeking approximate to determine waiting sequence according to described waiting Optimized model function by iteration
Row.
The game amount of leading data waiting device based on channel resource position the most according to claim 6, it is characterised in that data
Access unit specifically includes:
Attribute information gathers subelement, for gathering the amount of the leading data of the resource advertising position corresponding with each channel；
Operation data gathers subelement, takes data, load value data, logon data, newly for gathering corresponding opening of playing with candidate
Player registers amount, new player registers conversion ratio, district take player and hold carrying capacity.
The game amount of leading data waiting device based on channel resource position the most according to claim 7, it is characterised in that described
Waiting modeling unit specifically includes:
Channel business revenue data computation subunit, for according to described in open take data, described district take player hold carrying capacity, described in supplement number with money
Each player averagely amount of supplementing with money according to calculates each resource advertising position business revenue number of described channel according to the first preset formula
According to, described in open take data pass through described in the amount of leading data, described new player registers conversion ratio and described district take player and hold the list of carrying capacity
Clothes average carrying capacity of holding in individual district carries out calculating acquisition according to the second preset formula；
Business revenue data summation subelement, determines for all described resource advertising position business revenue data calculating acquisition are carried out summation
Described channel business revenue data；
Waiting modeling subelement, for according to described channel business revenue data and according to preset calculation calculate obtain operation because of
Prime number is according to setting up waiting Optimized model function, and described operation factor data is the money of the described resource advertising position in preset channel
Sort in source；
Wherein, described first preset formula be described resource advertising position business revenue data=described in open and take data × described district and take player
Hold carrying capacity × each described player averagely amount of supplementing with money；
Described second preset formula be described in open take data=INT (described in the amount of leading data × described new player registers conversion ratio/institute
State single district clothes and averagely hold carrying capacity).
The game amount of leading data waiting device based on channel resource position the most according to claim 8, it is characterised in that waiting
Modeling subelement specifically includes:
Waiting multiformity computing module, for calculating described resource advertising position by similarity between game according to the 3rd preset formula
The waiting multiformity of waiting, between described game, similarity is vectorial according to the 4th preset public affairs by the game characteristic of correspondence of playing two-by-two
Formula calculates and obtains；
Waiting freshness computing module, for by obtaining row to the fresh degree function read group total of described resource advertising position waiting
Phase freshness, described fresh degree function is preset according to the 5th by on-line time, current time and the weight factor that game is corresponding
Formula calculates and obtains；
Waiting MBM, for setting up according to described channel business revenue data, described waiting multiformity and described waiting freshness
Waiting Optimized model function；
Wherein, described 3rd preset formula be described waiting multiformity=1-(∑ { similarity between described game }/(0.5 × described
Resource sequence × (described resource sequence-1)))；
Described 4th preset formula be similarity between described game=∑ 1/sqrt (described in one game characteristic vector × another described in
Game characteristic vector)；
Described 5th preset formula be described fresh degree function=1/ (weight factor described in 1+ (described current time-described in reach the standard grade
Time)).
The game amount of leading data waiting device based on channel resource position the most according to claim 9, it is characterised in that institute
State waiting signal generating unit to specifically include:
Conversion subelement, is commented for being determined the described waiting Optimized model function of described waiting sequence by Lagrange conversion
Estimating the scoring functions of feasible solution, described feasible solution is the most feasible waiting；
Record subelement, for according to the described current optimal solution of scoring functions record；
Replicate intersection subelement, for selecting at least 2 candidate's feasible solutions by rule from described feasible solution, and to described candidate
Feasible solution carries out replicating intersection operation and obtains at least 2 corresponding new feasible solution results；
Mutation operation subelement, for described new feasible solution result is carried out mutation operation, generates several candidate schemes, and sentences
Whether the quantity of disconnected described candidate scheme reaches preset number of candidates, if it is not, the most again trigger the described intersection subelement that replicates, directly
Quantity to described candidate scheme reaches preset number of candidates, then trigger set and substitute subelement；
Described set substitutes subelement, for using preset number of candidates described candidate scheme as the set of new feasible solution, and
Substitute former feasible solution set, repeat to trigger described record subelement, described duplication intersection subelement, described mutation operation successively
Unit and described set substitute subelement and carry out process iterative computation, until iterations reaches preset iteration predetermined times, really
Fixed optimum waiting sequence.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610526060.6A CN106202904A (en) | 2016-07-05 | 2016-07-05 | A kind of game amount of leading data scheduling method based on channel resource position and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610526060.6A CN106202904A (en) | 2016-07-05 | 2016-07-05 | A kind of game amount of leading data scheduling method based on channel resource position and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106202904A true CN106202904A (en) | 2016-12-07 |
Family
ID=57465294
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610526060.6A Pending CN106202904A (en) | 2016-07-05 | 2016-07-05 | A kind of game amount of leading data scheduling method based on channel resource position and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106202904A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108632134A (en) * | 2017-03-23 | 2018-10-09 | 北京小唱科技有限公司 | Statistical method, system and the project setting method and system of channel evaluation parameter |
| CN110538458A (en) * | 2019-07-31 | 2019-12-06 | 中至数据集团股份有限公司 | Game recommendation method and device, readable storage medium and terminal |
| CN110737830A (en) * | 2019-09-17 | 2020-01-31 | Oppo广东移动通信有限公司 | Information processing method, device and storage medium |
| CN112348556A (en) * | 2020-09-27 | 2021-02-09 | 北京淇瑀信息科技有限公司 | Channel resource consumption optimization method and device and electronic equipment |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090248375A1 (en) * | 2000-05-11 | 2009-10-01 | Jean-Marie Billiotte | Centralised stochastic simulation process |
| CN104636954A (en) * | 2014-12-08 | 2015-05-20 | 北京掌阔技术有限公司 | Data mining method and data mining device for advertising media putting quantity |
| CN105574025A (en) * | 2014-10-15 | 2016-05-11 | 阿里巴巴集团控股有限公司 | Methods and devices for sorting score calculation and model building, and commodity recommendation system |
-
2016
- 2016-07-05 CN CN201610526060.6A patent/CN106202904A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090248375A1 (en) * | 2000-05-11 | 2009-10-01 | Jean-Marie Billiotte | Centralised stochastic simulation process |
| CN105574025A (en) * | 2014-10-15 | 2016-05-11 | 阿里巴巴集团控股有限公司 | Methods and devices for sorting score calculation and model building, and commodity recommendation system |
| CN104636954A (en) * | 2014-12-08 | 2015-05-20 | 北京掌阔技术有限公司 | Data mining method and data mining device for advertising media putting quantity |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108632134A (en) * | 2017-03-23 | 2018-10-09 | 北京小唱科技有限公司 | Statistical method, system and the project setting method and system of channel evaluation parameter |
| CN110538458A (en) * | 2019-07-31 | 2019-12-06 | 中至数据集团股份有限公司 | Game recommendation method and device, readable storage medium and terminal |
| CN110737830A (en) * | 2019-09-17 | 2020-01-31 | Oppo广东移动通信有限公司 | Information processing method, device and storage medium |
| CN112348556A (en) * | 2020-09-27 | 2021-02-09 | 北京淇瑀信息科技有限公司 | Channel resource consumption optimization method and device and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109472596B (en) | Alliance's chain common recognition method and system based on transaction assessment | |
| CN110427311A (en) | Disk failure prediction technique and system based on temporal aspect processing and model optimization | |
| US10737179B2 (en) | Predictive recommendations for skills development | |
| CN106202904A (en) | A kind of game amount of leading data scheduling method based on channel resource position and device | |
| CN105354725A (en) | Prediction method and system of promotion effect of application | |
| CN108171280A (en) | A kind of grader construction method and the method for prediction classification | |
| CN105069534A (en) | Customer loss prediction method and device | |
| CN109993428A (en) | Resource allocation methods and device | |
| CN112465626A (en) | Joint risk assessment method based on client classification aggregation and related equipment | |
| CN107741986A (en) | User's behavior prediction and corresponding information recommend method and apparatus | |
| CN108304853A (en) | Acquisition methods, device, storage medium and the electronic device for the degree of correlation of playing | |
| CN108830478B (en) | Crowd-sourced task processing-oriented team recommendation method | |
| Rhim et al. | Assessing potential threats to incumbent brands: New product positioning under price competition in a multisegmented market | |
| CN112712383A (en) | Potential user prediction method, device, equipment and storage medium of application program | |
| Kim et al. | Technological change, human capital structure, and multiple growth paths | |
| Sampath et al. | A generalized decision support framework for large‐scale project portfolio decisions | |
| KR20230154647A (en) | Influencer discovery method and device in social networking services | |
| CN117291655A (en) | Consumer life cycle operation analysis method based on entity and network collaborative mapping | |
| Baumann et al. | Bias On Demand: Investigating Bias with a Synthetic Data Generator. | |
| WO2009015069A1 (en) | Methods and systems of evaluating forest management and harvesting schemes | |
| Deb et al. | India’s primary energy evolution: Past trends and future prospects | |
| CN110135626A (en) | Credit management method and device, electronic equipment, storage medium | |
| Dong et al. | Research on academic early warning model based on improved SVM algorithm | |
| CN115738285A (en) | Game quality evaluation feedback method and system | |
| Henke et al. | Exploring European decarbonisation pathways in the Power Decisions Game |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| EE01 | Entry into force of recordation of patent licensing contract | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20161207 Assignee: GUANGZHOU HUYA INFORMATION TECHNOLOGY Co.,Ltd. Assignor: Guangzhou Huaduo Network Technology Co., Ltd. Contract record no.: 2018990000088 Denomination of invention: Channel resource niche based game guidance amount data scheduling method and device License type: Common License Record date: 20180413 |
|
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20161207 |