CN116452242A - Game profit prediction method, device and equipment based on fitting regression - Google Patents
Game profit prediction method, device and equipment based on fitting regression Download PDFInfo
- Publication number
- CN116452242A CN116452242A CN202310500200.2A CN202310500200A CN116452242A CN 116452242 A CN116452242 A CN 116452242A CN 202310500200 A CN202310500200 A CN 202310500200A CN 116452242 A CN116452242 A CN 116452242A
- Authority
- CN
- China
- Prior art keywords
- user
- game
- prediction model
- obtaining
- prediction
- 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
- 238000000034 method Methods 0.000 title claims abstract description 85
- 230000014759 maintenance of location Effects 0.000 claims abstract description 65
- 238000012417 linear regression Methods 0.000 claims abstract description 51
- 230000008901 benefit Effects 0.000 claims abstract description 42
- 230000006870 function Effects 0.000 claims description 52
- 238000009826 distribution Methods 0.000 claims description 26
- 238000012545 processing Methods 0.000 claims description 19
- 238000004590 computer program Methods 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000010606 normalization Methods 0.000 claims description 4
- 238000007476 Maximum Likelihood Methods 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 238000013486 operation strategy Methods 0.000 description 6
- 230000004044 response Effects 0.000 description 4
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000001514 detection method Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0202—Market predictions or forecasting for commercial activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/27—Regression, e.g. linear or logistic regression
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Abstract
The invention relates to the technical field of game profit prediction, in particular to a game profit prediction method, device and equipment based on fitting regression, wherein the method specifically comprises the following steps: obtaining a first multiple linear regression equation based on average creation benefits of game users in the first historical data, each user characteristic and relative entropy between the two, correcting errors between prediction data of the first multiple linear regression equation and the first historical data according to a difference method, and then obtaining an average creation benefit prediction model of the game users; and obtaining a second multi-element linear equation based on the advertisement putting cost and the newly added user quantity in the second historical data, correcting the error between the predicted data of the second multi-element linear regression equation and the second historical data according to a difference method, and then obtaining an advertisement putting cost and newly added user quantity prediction model. The invention combines the prediction factors such as the user retention rate, the newly added user quantity, the average user creation income and the like to form a game profit prediction model.
Description
Technical Field
The invention relates to the technical field of game profit prediction, in particular to a game profit prediction method, device and equipment based on fitting regression.
Background
Profit prediction has an important influence factor on the operation strategy, the operation strategy can be dynamically adjusted according to the daily profit in a certain time period in the future, the prediction of game profit is also important for game companies, and through the future prediction of game profit, the game companies can make different operation strategies for game products. However, the current game profit prediction method is not accurate enough, multiple predictors cannot be combined to form a game profit prediction model, and the most suitable prediction model cannot be selected according to different predictors, so that errors of game profit prediction data are not affected little.
Disclosure of Invention
The invention aims to provide a game profit prediction method, device and equipment based on fitting regression, which are used for forming a game profit prediction model according to a relative entropy method, fitting regression and a difference method and by combining prediction factors such as user retention rate, newly added user quantity, advertisement delivery cost, average user creation profit and the like, so as to solve at least one of the existing problems.
The invention provides a game profit prediction method based on fitting regression, which specifically comprises the following steps:
obtaining a first multiple linear regression equation based on average creation benefits of game users in the first historical data, each user characteristic and relative entropy between the two, correcting errors between prediction data of the first multiple linear regression equation and the first historical data according to a difference method, and then obtaining an average creation benefit prediction model of the game users;
obtaining a second multi-element linear equation based on the advertisement putting cost and the newly added user quantity in the second historical data, correcting the error between the predicted data of the second multi-element linear regression equation and the second historical data according to a difference method, and then obtaining an advertisement putting cost and the newly added user quantity predicted model;
determining a regression fit function model with the minimum error in each stage of the game life cycle according to the historical data of the retention rate of the game user, and then obtaining a prediction model of the retention rate of the game user;
and based on the average creation income prediction model of the game users, the advertisement putting cost and new user prediction model and the game user retention rate prediction model, obtaining a game profit prediction model, and determining an advertisement putting strategy according to the game profit prediction model.
Further, the obtaining a first multiple linear regression equation based on average game user creation benefits in the historical data, each user feature and relative entropy between the two, correcting errors between the prediction data of the first multiple linear regression equation and the historical data according to a difference method, and then obtaining an average game user creation benefits prediction model specifically includes:
acquiring first historical data, wherein the first historical data comprises historical data of average creation benefits of a user and historical data of each user characteristic, and the user characteristics comprise gender of the user, age of the user, region of the user, social contact of the user and income of the user;
acquiring the relative entropy between the average created benefit of each user and each user characteristic, carrying out normalization processing on the relative entropy to form a weight set, and then constructing a first multiple linear regression equation according to the weight set, the average created benefit of each user and each user characteristic;
correcting errors between the first prediction data of the first multiple linear regression equation and the first historical data according to a difference method and obtaining an average revenue generation prediction model of game users.
Further, the obtaining the relative entropy between the average created benefit of each user and each user feature, normalizing the relative entropy to form a weight set, and then constructing a first multiple linear regression equation according to the weight set, the average created benefit of each user and each user feature, which specifically includes:
determining a first probability distribution according to the historical data of average generated benefits of the user, and determining a second probability distribution set according to the historical data of each user characteristic;
obtaining a set of relative entropy from each second probability distribution of the first probability distribution and the set of second probability distributions by a relative entropy formula, the relative entropy formula satisfying
Wherein D is KL (P||Q) represents the relative entropy, P represents the first probability distribution, Q represents each second probability distribution, t represents the set of time periods, and N represents the nth time period of t;
normalizing the relative entropy set to obtain a weight set, and constructing a first multiple linear regression equation according to the weight set, the average creation benefit of each user and the characteristics of each user, wherein the first multiple linear regression equation satisfies y=beta 0 +β 1 x 1 +β 2 x 2 +β 3 x 3 +β 4 x 4 +β 5 x 5 ...+β k x k Wherein y represents the average revenue generated by each user, x 1 To x k Representing each user characteristic, beta 0 To beta T Representing a set of weights.
Further, the correcting the error between the first prediction data of the first multiple linear regression equation and the first history data according to the difference method and obtaining an average created profit prediction model of the game user specifically includes:
obtaining first prediction data according to the first polynary linear equation, and obtaining a first error sequence through errors between the first prediction data and the first historical data;
checking whether the first error sequence is a stable time sequence or not, if the first error sequence is not the stable time sequence, differentiating the first error sequence according to a difference method and obtaining a second error sequence;
and obtaining an average user generated gain prediction model according to the first polynary linear equation and the second error sequence.
Further, the method for obtaining a second multiple linear equation based on the advertisement delivery cost and the newly added user quantity in the second historical data, correcting the error between the prediction data of the second multiple linear regression equation and the second historical data according to a difference method, and then obtaining an advertisement delivery cost and a newly added user prediction model specifically includes:
Determining second historical data, wherein the second historical data comprises advertisement putting cost historical data in the 1 st time period to the T time period and newly added user quantity historical data in the T+1 th time period;
constructing a second polynary linear equation according to the second historical data, wherein the second historical data comprises advertisement putting cost in the 1 st time period to the T time period and the newly-increased user quantity in the T+1 th time period;
obtaining second prediction data according to the second polynomial linear equation, and obtaining a third error sequence through errors between the second prediction data and the second historical data;
checking whether the third error sequence is a stable time sequence or not, if the first error sequence is not the stable time sequence, differentiating the third error sequence according to a difference method and obtaining a fourth error sequence;
and obtaining advertisement putting cost and a newly added user prediction model according to the second multiple linear regression equation and the fourth error sequence.
Still further, the second multiple linear equation satisfies y=β 0 +β 1 x 1 +β 2 x 2 +β 3 x 3 +β 4 x 4 +β 5 x 5 ...+β T x T Wherein y represents the newly added user quantity in the T+1th time period, and x 1 To x T Respectively representing advertisement delivery costs in the 1 st time period to the T time period, beta 0 To beta T Representing advertising cost weight per time period, and beta 0 ＜...＜β 5 ＜...＜β T 。
Further, determining a regression fit function model with the smallest error in each stage of the game life cycle according to the historical data of the retention rate of the game user, and then obtaining a prediction model of the retention rate of the game user, which specifically comprises:
determining a stage set of a game user retention history data and a game life cycle, wherein the stage set comprises an online stage, a growing stage, a maturing stage and a declining stage;
in the current stage of the game life cycle and each stage before the current stage, determining a first game user retention prediction model set according to each regression function in a regression function set, and then determining a second game user retention prediction model set through each fitting function in a fitting function set, wherein the regression function set comprises a power function, an exponential function and a logarithmic function, and the fitting function set comprises a least square method, a maximum likelihood estimation method and a maximum posterior estimation method;
screening a second game user retention prediction model with the smallest error value from the second game user retention prediction model set based on an error calculation function, wherein the error calculation function comprises a mean value error, an absolute error, a standard error and a residual square sum;
And determining the second game user retention prediction model with the minimum error value as a game user retention prediction model.
Further, the creating a profit prediction model, the advertisement putting cost and new user quantity prediction model and the game user retention rate prediction model based on the game user average, obtaining a game profit prediction model, and determining an advertisement putting strategy according to the game profit prediction model, specifically including:
determining daily average creative revenue prediction data for the users in a first future time period according to the average creative revenue prediction model of the game users;
determining daily advertisement delivery cost prediction data in the first future time period and daily new user quantity prediction data in the first future time period according to the advertisement delivery cost and the new user quantity prediction model;
determining a daily old user retention in the first future time period according to the game user retention prediction model and the old user amount before the first future time period;
determining a daily active user amount in the first future time period according to the daily newly-increased user amount prediction data in the first future time period and the daily old user remaining amount in the first future time period;
Determining daily game profit prediction data for the first future time period based on a product of the amount of active daily users in the first future time period and the average daily user creative profit prediction data for the first future time period;
and determining to stably maintain or increase or decrease the advertisement delivery cost in the first future time period according to the difference comparison of the daily advertisement delivery cost prediction data in the first future time period and the daily game profit prediction data in the first future time period.
The invention also provides a game profit prediction device based on fitting regression, which specifically comprises:
the first processing module is used for obtaining a first multiple linear regression equation based on average creation benefits of game users in the first historical data, each user characteristic and relative entropy between the two, correcting errors between prediction data of the first multiple linear regression equation and the first historical data according to a difference method, and then obtaining an average creation benefit prediction model of the game users;
the second processing module is used for obtaining a second multi-element linear equation based on the advertisement delivery cost and the newly added user quantity in the second historical data, correcting the error between the prediction data of the second multi-element linear regression equation and the second historical data according to a difference method, and then obtaining an advertisement delivery cost and the newly added user quantity prediction model;
The third processing module is used for determining a regression fit function model with the minimum error in each stage of the game life cycle according to the historical data of the retention rate of the game user, and then obtaining a prediction model of the retention rate of the game user;
and the fourth processing module is used for creating a profit prediction model, the advertisement putting cost, a newly-added user prediction model and the game user retention rate prediction model based on the average game user to obtain a game profit prediction model, and determining an advertisement putting strategy according to the game profit prediction model.
The present invention also provides a computer device comprising: memory and processor and computer program stored on the memory, which when executed on the processor, implements a fitting regression-based game profit prediction method according to any one of claims 1 to 8.
Compared with the prior art, the invention has at least one of the following technical effects:
1. and determining the importance degree of different factors affecting the average creation benefits of the game user according to the relative entropy formula, thereby determining the weight difference of the different influencing factors.
2. And determining game profit prediction data and recovery time based on multiple prediction factors such as average creation profit prediction data, advertisement putting cost, newly added user quantity prediction data, game user retention rate prediction data and the like of the game users.
3. According to the close relation between the game user retention and the game life cycle, determining a game user retention prediction curve with the minimum error in each stage of the game life cycle through different fitting regression functions and error calculation functions.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow chart of a game profit prediction method based on fitting regression according to an embodiment of the present invention;
FIG. 2 is a schematic structural diagram of a game profit prediction device based on fitting regression according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of a computer device according to an embodiment of the present invention.
Detailed Description
In the following description, for purposes of explanation and not limitation, specific details are set forth, such as particular system configurations, techniques, etc. in order to provide a thorough understanding of the embodiments of the present application. It will be apparent, however, to one skilled in the art that the present application may be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present application with unnecessary detail.
It should be understood that the terms "comprises" and/or "comprising," when used in this specification and the appended claims, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
It should also be understood that the term "and/or" as used in this specification and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes such combinations.
As used in this specification and the appended claims, the term "if" may be interpreted as "when..once" or "in response to a determination" or "in response to detection" depending on the context. Similarly, the phrase "if a determination" or "if a [ described condition or event ] is detected" may be interpreted in the context of meaning "upon determination" or "in response to determination" or "upon detection of a [ described condition or event ]" or "in response to detection of a [ described condition or event ]".
In addition, in the description of the present application and the appended claims, the terms "first," "second," "third," and the like are used merely to distinguish between descriptions and are not to be construed as indicating or implying relative importance.
Reference in the specification to "one embodiment" or "some embodiments" or the like means that a particular feature, structure, or characteristic described in connection with the embodiment is included in one or more embodiments of the application. Thus, appearances of the phrases "in one embodiment," "in some embodiments," "in other embodiments," and the like in the specification are not necessarily all referring to the same embodiment, but mean "one or more but not all embodiments" unless expressly specified otherwise. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless expressly specified otherwise.
Profit prediction has an important influence factor on the operation strategy, the operation strategy can be dynamically adjusted according to the daily profit in a certain time period in the future, the prediction of game profit is also important for game companies, and through the future prediction of game profit, the game companies can make different operation strategies for game products. However, the current game profit prediction method is not accurate enough, multiple predictors cannot be combined to form a game profit prediction model, and the most suitable prediction model cannot be selected according to different predictors, so that errors of game profit prediction data are not affected little.
Referring to fig. 1, an embodiment of the present invention provides a game profit prediction method based on fitting regression, where the method specifically includes:
s101: obtaining a first multiple linear regression equation based on average created benefits of game users in the first historical data, each user characteristic and relative entropy between the two, correcting errors between prediction data of the first multiple linear regression equation and the first historical data according to a difference method, and then obtaining an average created benefits prediction model of the game users.
In some embodiments, the obtaining a first multiple linear regression equation based on average game user creation benefits in the historical data, each user feature and relative entropy between the two, correcting an error between the prediction data of the first multiple linear regression equation and the historical data according to a difference method, and then obtaining an average game user creation benefit prediction model specifically includes:
acquiring first historical data, wherein the first historical data comprises historical data of average creation benefits of a user and historical data of each user characteristic, and the user characteristics comprise gender of the user, age of the user, region of the user, social contact of the user and income of the user;
Acquiring the relative entropy between the average created benefit of each user and each user characteristic, carrying out normalization processing on the relative entropy to form a weight set, and then constructing a first multiple linear regression equation according to the weight set, the average created benefit of each user and each user characteristic;
correcting errors between the first prediction data of the first multiple linear regression equation and the first historical data according to a difference method and obtaining an average revenue generation prediction model of game users.
Specifically, the obtaining the relative entropy between the average created benefit of each user and each user feature, normalizing the relative entropy to form a weight set, and then constructing a first multiple linear regression equation according to the weight set, the average created benefit of each user and each user feature, specifically including:
determining a first probability distribution according to the historical data of average generated benefits of the user, and determining a second probability distribution set according to the historical data of each user characteristic;
obtaining a set of relative entropy from each second probability distribution of the first probability distribution and the set of second probability distributions by a relative entropy formula, the relative entropy formula satisfying
Wherein D is KL (P||Q) represents the relative entropy, P represents the first probability distribution, Q represents each second probability distribution, t represents the set of time periods, and N represents the nth time period of t;
normalizing the relative entropy set to obtain a weight set, and constructing a first multiple linear regression equation according to the weight set, the average creation benefit of each user and the characteristics of each user, wherein the first multiple linear regression equation satisfies y=beta 0 +β 1 x 1 +β 2 x 2 +β 3 x 3 +β 4 x 4 +β 5 x 5 ...+β k x k Wherein y represents the average revenue generated by each user, x 1 To x k Representing each user characteristic, beta 0 To beta T Representing a set of weights.
Specifically, the correcting the error between the first prediction data of the first multiple linear regression equation and the first historical data according to the difference method and obtaining an average created benefit prediction model of the game user specifically includes:
obtaining first prediction data according to the first polynary linear equation, and obtaining a first error sequence through errors between the first prediction data and the first historical data;
checking whether the first error sequence is a stable time sequence or not, if the first error sequence is not the stable time sequence, differentiating the first error sequence according to a difference method and obtaining a second error sequence;
And obtaining an average user generated gain prediction model according to the first polynary linear equation and the second error sequence.
In this embodiment, the relative entropy is equivalent to the difference between the two probability distribution information entropies, that is, the information loss between the user average generated gain and different factors affecting the user average generated gain can be used to obtain the tightness between each factor in different factors and the user average generated gain, and the user characteristics are the factors that can reflect the user average generated gain. The user characteristics comprise user gender, user age, user region, user social contact and user income, and after each user characteristic is further refined, a second probability distribution is established according to specific historical data of each user characteristic.
For example, the user sexes include male users and female users, the user ages include teenagers (10-20 years), young people (20-30 years), middle-aged and older (more than 30 years), the user regions include domestic and foreign regions and domestic different regions, the user income includes 1000-3000 yuan, 3000-7000 yuan, 7000-10000 yuan and 10000 yuan or more, the second probability distribution is built one by one according to each specific user characteristic, different relative entropies are obtained through a relative entropy formula with the first probability distribution, and the weight set is built according to the relative entropy. Meanwhile, the fact that the numerical value difference between the relative entropies is large is considered, so that after normalization processing is carried out on each relative entropy, a weight set is established, and the weight set is used as a coefficient of a first multi-element linear equation.
And obtaining first prediction data according to a first polynary linear equation, wherein the first prediction data creates profit prediction data for average users in a first future time period, comparing the first prediction data with historical data to obtain a first error sequence, and if the first error sequence is not a stable sequence, converting the first error sequence into the stable sequence, namely a second error sequence by a difference method. And obtaining average creation gain prediction data of the users in other future time periods according to the first multi-element linear equation, and adding the prediction data and the second error sequence to obtain more accurate average creation gain prediction data of the game users.
S102: and obtaining a second multi-element linear equation based on the advertisement putting cost and the newly added user quantity in the second historical data, correcting the error between the predicted data of the second multi-element linear regression equation and the second historical data according to a difference method, and then obtaining an advertisement putting cost and newly added user quantity prediction model.
In some embodiments, the obtaining a second multiple linear equation based on the advertisement delivery cost and the newly added user quantity in the second historical data, correcting an error between the prediction data of the second multiple linear regression equation and the second historical data according to a difference method, and then obtaining an advertisement delivery cost and a newly added user prediction model specifically includes:
Determining second historical data, wherein the second historical data comprises advertisement putting cost historical data in the 1 st time period to the T time period and newly added user quantity historical data in the T+1 th time period;
constructing a second polynary linear equation according to the second historical data, wherein the second historical data comprises advertisement putting cost in the 1 st time period to the T time period and the newly-increased user quantity in the T+1 th time period;
obtaining second prediction data according to the second polynomial linear equation, and obtaining a third error sequence through errors between the second prediction data and the second historical data;
checking whether the third error sequence is a stable time sequence or not, if the first error sequence is not the stable time sequence, differentiating the third error sequence according to a difference method and obtaining a fourth error sequence;
and obtaining advertisement putting cost and a newly added user prediction model according to the second multiple linear regression equation and the fourth error sequence.
The second multiple linear equation satisfies y=β 0 +β 1 x 1 +β 2 x 2 +β 3 x 3 +β 4 x 4 +β 5 x 5 ...+β T x T Wherein y represents the newly added user quantity in the T+1th time period, and x 1 To x T Respectively representing advertisement delivery costs in the 1 st time period to the T time period, beta 0 To beta T Representing advertising cost weight per time period, and beta 0 ＜...＜β 5 ＜...＜β T 。
In this embodiment, since the newly added user quantity is generally related to advertisement delivery, a predictive data relationship model between advertisement delivery cost and the newly added user quantity can be established, since the newly added user quantity of the current day (time period t+1) is related to all advertisement delivery costs before the current day (time period 1 to time period T), it is necessary to count advertisement delivery cost of each day before the current day, and the closer to the advertisement delivered on the current day, the more related to the newly added user quantity on the current day, so when setting advertisement delivery cost weight, the weight of time period 1 is the lowest, and then gradually increases, and the weight of time period T is the highest. When setting a specific weight ratio, consideration may be made according to specific history data.
And after the second multi-element linear equation is established, second prediction data is obtained, the second prediction data is relation prediction data between advertisement putting cost and newly increased user quantity in a first future time period, the second prediction data and historical data are also compared to obtain a third error sequence, and if the third error sequence is not a stable sequence, namely a fourth error sequence, is obtained through a difference method. And then, obtaining the relation prediction data between the advertisement putting cost and the newly increased user quantity in other future time periods according to the second multi-element linear equation, and adding the relation prediction data with the fourth error sequence to obtain more accurate relation prediction data between the advertisement putting cost and the newly increased user quantity.
S103: and determining a regression fit function model with the minimum error in each stage of the game life cycle according to the historical data of the retention rate of the game user, and then obtaining a prediction model of the retention rate of the game user.
In some embodiments, the determining a regression fit function model with minimum error in each stage of the game life cycle according to the historical data of the retention rate of the game user, and then obtaining a prediction model of the retention rate of the game user specifically includes:
determining a stage set of a game user retention history data and a game life cycle, wherein the stage set comprises an online stage, a growing stage, a maturing stage and a declining stage;
in the current stage of the game life cycle and each stage before the current stage, determining a first game user retention prediction model set according to each regression function in a regression function set, and then determining a second game user retention prediction model set through each fitting function in a fitting function set, wherein the regression function set comprises a power function, an exponential function and a logarithmic function, and the fitting function set comprises a least square method, a maximum likelihood estimation method and a maximum posterior estimation method;
Screening a second game user retention prediction model with the smallest error value from the second game user retention prediction model set based on an error calculation function, wherein the error calculation function comprises a mean value error, an absolute error, a standard error and a residual square sum;
and determining the second game user retention prediction model with the minimum error value as a game user retention prediction model.
In this embodiment, a game product typically has a life cycle, and the life cycle of the game is related to the number of users 'reservations, so that it can be determined which stage of the life cycle a game is currently in according to the number of users' reservations, and the life cycle of the game typically includes four stages, namely, an online stage, a growing stage, a maturing stage, and a declining stage. In different stages, the accuracy of different functions for carrying out predictive curve regression fitting based on historical data is also different, so that each initial predictive model is obtained according to each regression function and fitting function in each stage, the error value of each initial predictive model is comprehensively evaluated according to the error calculation function, and the initial predictive model with the minimum error value is screened out. If the current game life cycle is in the online stage, only the initial prediction model with the minimum error value in the online stage is needed to be screened out and determined as the game user retention prediction model, otherwise, the initial prediction model with the minimum error value in each stage is needed to be spliced, and then the game user retention prediction model is formed.
S104: and based on the average creation income prediction model of the game users, the advertisement putting cost and new user prediction model and the game user retention rate prediction model, obtaining a game profit prediction model, and determining an advertisement putting strategy according to the game profit prediction model.
In some embodiments, the creating a profit prediction model, the advertisement putting cost and new user quantity prediction model and the game user retention rate prediction model based on the game user average, obtaining a game profit prediction model, and determining an advertisement putting strategy according to the game profit prediction model specifically includes:
determining daily average creative revenue prediction data for the users in a first future time period according to the average creative revenue prediction model of the game users;
determining daily advertisement delivery cost prediction data in the first future time period and daily new user quantity prediction data in the first future time period according to the advertisement delivery cost and the new user quantity prediction model;
determining a daily old user retention in the first future time period according to the game user retention prediction model and the old user amount before the first future time period;
Determining a daily active user amount in the first future time period according to the daily newly-increased user amount prediction data in the first future time period and the daily old user remaining amount in the first future time period;
determining daily game profit prediction data for the first future time period based on a product of the amount of active daily users in the first future time period and the average daily user creative profit prediction data for the first future time period;
and determining to stably maintain or increase or decrease the advertisement delivery cost in the first future time period according to the difference comparison of the daily advertisement delivery cost prediction data in the first future time period and the daily game profit prediction data in the first future time period.
In this embodiment, respective prediction data in the first future period is obtained from each of the determined prediction models, and game profit prediction data is obtained from the respective prediction data. The daily game profit is equal to the product of the average daily user creation profit and the daily active user quantity, wherein the average daily user creation profit prediction data can be obtained through a game user average creation profit prediction model, the daily active user quantity is equal to the sum of the daily old user retention and the daily newly-increased user quantity, the daily old user retention can be obtained according to the current day game user retention obtained through a game user retention prediction model and the old user quantity before the current day, and the daily newly-increased user quantity can be obtained according to the advertisement putting cost and the newly-increased user quantity prediction model.
After determining the daily game profit prediction data in the first future time period through the respective prediction data, the difference between the daily advertisement putting cost prediction data in the first future time period and the daily game profit prediction data in the first future time period can be compared, and the advertisement putting cost is determined to be increased, or the advertisement putting cost is reduced, or the advertisement putting cost is stably kept for the game in the first future time period.
Referring to fig. 2, the embodiment of the present invention further provides a game profit prediction apparatus 2 based on fitting regression, where the apparatus 2 specifically includes:
a first processing module 201, configured to obtain a first multiple linear regression equation based on average revenue generation of the game user and each user feature in the first historical data and relative entropy therebetween, correct an error between the prediction data of the first multiple linear regression equation and the first historical data according to a difference method, and then obtain an average revenue generation prediction model of the game user;
a second processing module 202, configured to obtain a second multiple linear equation based on the advertisement delivery cost and the newly added user quantity in the second historical data, correct an error between the predicted data of the second multiple linear regression equation and the second historical data according to a difference method, and then obtain an advertisement delivery cost and a newly added user quantity prediction model;
A third processing module 203, configured to determine a regression fit function model with the smallest error in each stage of the game life cycle according to the historical data of the retention rate of the game user, and then obtain a prediction model of the retention rate of the game user;
and a fourth processing module 204, configured to create a profit prediction model, the advertisement delivery cost and new user prediction model, and the game user retention prediction model based on the average game user, obtain a game profit prediction model, and determine an advertisement delivery strategy according to the game profit prediction model.
It will be appreciated that the content of the embodiment of the method for predicting game profit based on fitting regression shown in fig. 1 is applicable to the embodiment of the device for predicting game profit based on fitting regression, and the functions of the embodiment of the device for predicting game profit based on fitting regression are the same as those of the embodiment of the method for predicting game profit based on fitting regression shown in fig. 1, and the advantages achieved are the same as those achieved by the embodiment of the method for predicting game profit based on fitting regression shown in fig. 1.
It should be noted that, because the content of information interaction and execution process between the above devices is based on the same concept as the method embodiment of the present invention, specific functions and technical effects thereof may be referred to in the method embodiment section, and will not be described herein.
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-described division of the functional units and modules is illustrated, and in practical application, the above-described functional distribution may be performed by different functional units and modules according to needs, i.e. the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-described functions. The functional units and modules in the embodiment may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit, where the integrated units may be implemented in a form of hardware or a form of a software functional unit. In addition, specific names of the functional units and modules are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present application. The specific working process of the units and modules in the above system may refer to the corresponding process in the foregoing method embodiment, which is not described herein again.
Referring to fig. 3, an embodiment of the present invention further provides a computer device 3, including: memory 302 and processor 301 and a computer program 303 stored on memory 302, which computer program 303, when executed on processor 301, implements a fitting regression based game profit prediction method as described in any of the above methods.
The computer device 3 may be a desktop computer, a notebook computer, a palm computer, a cloud server, or the like. The computer device 3 may include, but is not limited to, a processor 301, a memory 302. It will be appreciated by those skilled in the art that fig. 3 is merely an example of the computer device 3 and is not meant to be limiting as the computer device 3, and may include more or fewer components than shown, or may combine certain components, or different components, such as may also include input-output devices, network access devices, etc.
The processor 301 may be a central processing unit (Central Processing Unit, CPU), the processor 301 may also be other general purpose processors, digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), off-the-shelf programmable gate arrays (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The memory 302 may in some embodiments be an internal storage unit of the computer device 3, such as a hard disk or a memory of the computer device 3. The memory 302 may in other embodiments also be an external storage device of the computer device 3, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash Card (Flash Card) or the like, which are provided on the computer device 3. Further, the memory 302 may also include both an internal storage unit and an external storage device of the computer device 3. The memory 302 is used to store an operating system, application programs, boot loader (BootLoader), data, and other programs, such as program code for the computer program. The memory 302 may also be used to temporarily store data that has been output or is to be output.
The embodiment of the invention also provides a computer readable storage medium, on which a computer program is stored, which when being run by a processor, implements the game profit prediction method based on fitting regression as described in any one of the above methods.
In this embodiment, the integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the present application implements all or part of the flow of the method of the above embodiments, and may be implemented by a computer program to instruct related hardware, where the computer program may be stored in a computer readable storage medium, where the computer program, when executed by a processor, may implement the steps of each of the method embodiments described above. Wherein the computer program comprises computer program code which may be in source code form, object code form, executable file or some intermediate form etc. The computer readable medium may include at least: any entity or device capable of carrying computer program code to a photographing device/terminal apparatus, recording medium, computer Memory, read-Only Memory (ROM), random access Memory (RAM, random Access Memory), electrical carrier signals, telecommunications signals, and software distribution media. Such as a U-disk, removable hard disk, magnetic or optical disk, etc. In some jurisdictions, computer readable media may not be electrical carrier signals and telecommunications signals in accordance with legislation and patent practice.
In the foregoing embodiments, the descriptions of the embodiments are emphasized, and in part, not described or illustrated in any particular embodiment, reference is made to the related descriptions of other embodiments.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
In the embodiments disclosed in the present application, it should be understood that the disclosed apparatus/terminal device and method may be implemented in other manners. For example, the apparatus/terminal device embodiments described above are merely illustrative, e.g., the division of the modules or units is merely a logical function division, and there may be additional divisions in actual implementation, e.g., multiple units or components may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection via interfaces, devices or units, which may be in electrical, mechanical or other forms.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
Claims (10)
1. A game profit prediction method based on fitting regression is characterized by comprising the following steps:
obtaining a first multiple linear regression equation based on average creation benefits of game users in the first historical data, each user characteristic and relative entropy between the two, correcting errors between prediction data of the first multiple linear regression equation and the first historical data according to a difference method, and then obtaining an average creation benefit prediction model of the game users;
obtaining a second multi-element linear equation based on the advertisement putting cost and the newly added user quantity in the second historical data, correcting the error between the predicted data of the second multi-element linear regression equation and the second historical data according to a difference method, and then obtaining an advertisement putting cost and the newly added user quantity predicted model;
Determining a regression fit function model with the minimum error in each stage of the game life cycle according to the historical data of the retention rate of the game user, and then obtaining a prediction model of the retention rate of the game user;
and based on the average creation income prediction model of the game users, the advertisement putting cost and new user prediction model and the game user retention rate prediction model, obtaining a game profit prediction model, and determining an advertisement putting strategy according to the game profit prediction model.
2. The method according to claim 1, wherein the obtaining a first multiple linear regression equation based on average game user creation benefits and each user feature in the history data and relative entropy therebetween, correcting an error between the prediction data of the first multiple linear regression equation and the history data according to a difference method, and then obtaining an average game user creation benefit prediction model, specifically comprises:
acquiring first historical data, wherein the first historical data comprises historical data of average creation benefits of a user and historical data of each user characteristic, and the user characteristics comprise gender of the user, age of the user, region of the user, social contact of the user and income of the user;
Acquiring the relative entropy between the average created benefit of each user and each user characteristic, carrying out normalization processing on the relative entropy to form a weight set, and then constructing a first multiple linear regression equation according to the weight set, the average created benefit of each user and each user characteristic;
correcting errors between the first prediction data of the first multiple linear regression equation and the first historical data according to a difference method and obtaining an average revenue generation prediction model of game users.
3. The method according to claim 2, wherein the obtaining the relative entropy between the average created benefit of each user and each user feature, normalizing the relative entropy to form a weight set, and then constructing a first multiple linear regression equation according to the weight set, the average created benefit of each user and each user feature, specifically includes:
determining a first probability distribution according to the historical data of average generated benefits of the user, and determining a second probability distribution set according to the historical data of each user characteristic;
obtaining a set of relative entropy from each second probability distribution of the first probability distribution and the set of second probability distributions by a relative entropy formula, the relative entropy formula satisfying
Wherein D is KL (P||Q) represents the relative entropy, P represents the first probability distribution, Q represents each second probability distribution, t represents the set of time periods, and N represents the nth time period of t;
normalizing the relative entropy set to obtain a weight set, and constructing a first multiple linear regression equation according to the weight set, the average creation benefit of each user and the characteristics of each user, wherein the first multiple linear regression equation satisfies y=beta 0 +β 1 x 1 +β 2 x 2 +β 3 x 3 +β 4 x 4 +β 5 x 5 ...+β k x k Wherein y represents the average revenue generated by each user, x 1 To x k Representing each user characteristic, beta 0 To beta T Representing a set of weights.
4. The method according to claim 2, wherein said correcting the error between the first prediction data of the first multiple linear regression equation and the first history data according to the difference method and obtaining the average created revenue prediction model for the game user specifically comprises:
obtaining first prediction data according to the first polynary linear equation, and obtaining a first error sequence through errors between the first prediction data and the first historical data;
checking whether the first error sequence is a stable time sequence or not, if the first error sequence is not the stable time sequence, differentiating the first error sequence according to a difference method and obtaining a second error sequence;
And obtaining an average user generated gain prediction model according to the first polynary linear equation and the second error sequence.
5. The method according to claim 1, wherein the obtaining a second multiple linear equation based on the advertisement delivery cost and the newly added user quantity in the second historical data, correcting an error between the predicted data of the second multiple linear regression equation and the second historical data according to a difference method, and then obtaining an advertisement delivery cost and a newly added user prediction model, specifically includes:
determining second historical data, wherein the second historical data comprises advertisement putting cost historical data in the 1 st time period to the T time period and newly added user quantity historical data in the T+1 th time period;
constructing a second polynary linear equation according to the second historical data, wherein the second historical data comprises advertisement putting cost in the 1 st time period to the T time period and the newly-increased user quantity in the T+1 th time period;
obtaining second prediction data according to the second polynomial linear equation, and obtaining a third error sequence through errors between the second prediction data and the second historical data;
checking whether the third error sequence is a stable time sequence or not, if the first error sequence is not the stable time sequence, differentiating the third error sequence according to a difference method and obtaining a fourth error sequence;
And obtaining advertisement putting cost and a newly added user prediction model according to the second multiple linear regression equation and the fourth error sequence.
6. According to claim 5Wherein said second plurality of linear equations satisfies y=β 0 +β 1 x 1 +β 2 x 2 +β 3 x 3 +β 4 x 4 +β 5 x 5 ...+β T x T Wherein y represents the newly added user quantity in the T+1th time period, and x 1 To x T Respectively representing advertisement delivery costs in the 1 st time period to the T time period, beta 0 To beta T Representing advertising cost weight per time period, and beta 0 ＜...＜β 5 ＜...＜β T 。
7. The method according to claim 1, wherein determining a regression fit function model with the least error at each stage of the game life cycle based on the historical data of the retention rate of the game user, and then obtaining a prediction model of the retention rate of the game user, comprises:
determining a stage set of a game user retention history data and a game life cycle, wherein the stage set comprises an online stage, a growing stage, a maturing stage and a declining stage;
in the current stage of the game life cycle and each stage before the current stage, determining a first game user retention prediction model set according to each regression function in a regression function set, and then determining a second game user retention prediction model set through each fitting function in a fitting function set, wherein the regression function set comprises a power function, an exponential function and a logarithmic function, and the fitting function set comprises a least square method, a maximum likelihood estimation method and a maximum posterior estimation method;
Screening a second game user retention prediction model with the smallest error value from the second game user retention prediction model set based on an error calculation function, wherein the error calculation function comprises a mean value error, an absolute error, a standard error and a residual square sum;
and determining the second game user retention prediction model with the minimum error value as a game user retention prediction model.
8. The method according to claim 1, wherein the creating a profit prediction model based on the average of the game users, the advertisement delivery cost and new user quantity prediction model, and the game user retention rate prediction model, obtaining a game profit prediction model, and determining an advertisement delivery strategy according to the game profit prediction model, specifically comprises:
determining daily average creative revenue prediction data for the users in a first future time period according to the average creative revenue prediction model of the game users;
determining daily advertisement delivery cost prediction data in the first future time period and daily new user quantity prediction data in the first future time period according to the advertisement delivery cost and the new user quantity prediction model;
Determining a daily old user retention in the first future time period according to the game user retention prediction model and the old user amount before the first future time period;
determining a daily active user amount in the first future time period according to the daily newly-increased user amount prediction data in the first future time period and the daily old user remaining amount in the first future time period;
determining daily game profit prediction data for the first future time period based on a product of the amount of active daily users in the first future time period and the average daily user creative profit prediction data for the first future time period;
and determining to stably maintain or increase or decrease the advertisement delivery cost in the first future time period according to the difference comparison of the daily advertisement delivery cost prediction data in the first future time period and the daily game profit prediction data in the first future time period.
9. A game profit prediction device based on fitting regression, characterized in that the device specifically comprises:
the first processing module is used for obtaining a first multiple linear regression equation based on average creation benefits of game users in the first historical data, each user characteristic and relative entropy between the two, correcting errors between prediction data of the first multiple linear regression equation and the first historical data according to a difference method, and then obtaining an average creation benefit prediction model of the game users;
The second processing module is used for obtaining a second multi-element linear equation based on the advertisement delivery cost and the newly added user quantity in the second historical data, correcting the error between the prediction data of the second multi-element linear regression equation and the second historical data according to a difference method, and then obtaining an advertisement delivery cost and the newly added user quantity prediction model;
the third processing module is used for determining a regression fit function model with the minimum error in each stage of the game life cycle according to the historical data of the retention rate of the game user, and then obtaining a prediction model of the retention rate of the game user;
and the fourth processing module is used for creating a profit prediction model, the advertisement putting cost, a newly-added user prediction model and the game user retention rate prediction model based on the average game user to obtain a game profit prediction model, and determining an advertisement putting strategy according to the game profit prediction model.
10. A computer device, comprising: memory and processor and computer program stored on the memory, which when executed on the processor, implements a fitting regression-based game profit prediction method according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310500200.2A CN116452242A (en) | 2023-05-06 | 2023-05-06 | Game profit prediction method, device and equipment based on fitting regression |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310500200.2A CN116452242A (en) | 2023-05-06 | 2023-05-06 | Game profit prediction method, device and equipment based on fitting regression |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116452242A true CN116452242A (en) | 2023-07-18 |
Family
ID=87125616
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310500200.2A Pending CN116452242A (en) | 2023-05-06 | 2023-05-06 | Game profit prediction method, device and equipment based on fitting regression |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116452242A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116956594A (en) * | 2023-07-25 | 2023-10-27 | 广州锐兴科技有限公司 | Rural power grid optimization method, device and equipment based on topological structure |
-
2023
- 2023-05-06 CN CN202310500200.2A patent/CN116452242A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116956594A (en) * | 2023-07-25 | 2023-10-27 | 广州锐兴科技有限公司 | Rural power grid optimization method, device and equipment based on topological structure |
| CN116956594B (en) * | 2023-07-25 | 2024-02-09 | 广州锐兴科技有限公司 | Rural power grid optimization method, device and equipment based on topological structure |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109299356B (en) | Activity recommendation method and device based on big data, electronic equipment and storage medium | |
| US20110047058A1 (en) | Apparatus and method for modeling loan attributes | |
| CN108363788B (en) | Post intelligent ranking method and device and computer readable storage medium | |
| CN108965951B (en) | Advertisement playing method and device | |
| EP3961384A1 (en) | Automatic derivation of software engineering artifact attributes from product or service development concepts | |
| CN116452242A (en) | Game profit prediction method, device and equipment based on fitting regression | |
| CN111091287A (en) | Risk object identification method and device and computer equipment | |
| CN111797320A (en) | Data processing method, device, equipment and storage medium | |
| CN108197795B (en) | Malicious group account identification method, device, terminal and storage medium | |
| US20210350427A1 (en) | Method and device for cost early warning of prefabricated building | |
| CN109597745B (en) | Abnormal data processing method and device | |
| CN111160614B (en) | Training method and device of resource transfer prediction model and computing equipment | |
| CN112148880A (en) | Customer service dialogue corpus clustering method, system, equipment and storage medium | |
| CN110717653B (en) | Risk identification method and apparatus, and electronic device | |
| CN109523296B (en) | User behavior probability analysis method and device, electronic equipment and storage medium | |
| CN112287208A (en) | User portrait generation method and device, electronic equipment and storage medium | |
| CN113095939B (en) | Block chain intelligent contract recommendation method and device | |
| CN113486933B (en) | Model training method, user identity information prediction method and device | |
| CN115587310A (en) | Baseline load estimation method and device, electronic equipment and storage medium | |
| CN110246026B (en) | Data transfer output combination setting method and device and terminal equipment | |
| CN111523083A (en) | Method and device for determining power load declaration data | |
| CN113095392A (en) | Abnormal service identification method and device | |
| CN109472455B (en) | Activity evaluation method, activity evaluation device, electronic equipment and storage medium | |
| CN115221663A (en) | Data processing method, device, equipment and computer readable storage medium | |
| CN117132177B (en) | Runoff forecasting model construction and runoff forecasting method based on multiple hypothesis test |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |