CN112446764A - Game commodity recommendation method and device and electronic equipment - Google Patents
Game commodity recommendation method and device and electronic equipment Download PDFInfo
- Publication number
- CN112446764A CN112446764A CN202011384767.0A CN202011384767A CN112446764A CN 112446764 A CN112446764 A CN 112446764A CN 202011384767 A CN202011384767 A CN 202011384767A CN 112446764 A CN112446764 A CN 112446764A
- Authority
- CN
- China
- Prior art keywords
- data
- payment
- characteristic
- preset
- feature
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 61
- 238000010606 normalization Methods 0.000 claims abstract description 36
- 238000012545 processing Methods 0.000 claims abstract description 31
- 230000009466 transformation Effects 0.000 claims description 41
- 230000000875 corresponding effect Effects 0.000 claims description 29
- 238000004590 computer program Methods 0.000 claims description 8
- 230000002596 correlated effect Effects 0.000 claims description 6
- 230000001131 transforming effect Effects 0.000 claims description 5
- 238000010586 diagram Methods 0.000 description 8
- 238000004364 calculation method Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 238000009499 grossing Methods 0.000 description 3
- 235000003181 Panax pseudoginseng Nutrition 0.000 description 2
- 244000131316 Panax pseudoginseng Species 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 238000005034 decoration Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011551 log transformation method Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000001502 supplementing effect Effects 0.000 description 1
Images
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
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
-
- A—HUMAN NECESSITIES
- A63—SPORTS; GAMES; AMUSEMENTS
- A63F—CARD, BOARD, OR ROULETTE GAMES; INDOOR GAMES USING SMALL MOVING PLAYING BODIES; VIDEO GAMES; GAMES NOT OTHERWISE PROVIDED FOR
- A63F13/00—Video games, i.e. games using an electronically generated display having two or more dimensions
- A63F13/70—Game security or game management aspects
- A63F13/79—Game security or game management aspects involving player-related data, e.g. identities, accounts, preferences or play histories
- A63F13/792—Game security or game management aspects involving player-related data, e.g. identities, accounts, preferences or play histories for payment purposes, e.g. monthly subscriptions
-
- A—HUMAN NECESSITIES
- A63—SPORTS; GAMES; AMUSEMENTS
- A63F—CARD, BOARD, OR ROULETTE GAMES; INDOOR GAMES USING SMALL MOVING PLAYING BODIES; VIDEO GAMES; GAMES NOT OTHERWISE PROVIDED FOR
- A63F2300/00—Features of games using an electronically generated display having two or more dimensions, e.g. on a television screen, showing representations related to the game
- A63F2300/50—Features of games using an electronically generated display having two or more dimensions, e.g. on a television screen, showing representations related to the game characterized by details of game servers
- A63F2300/57—Features of games using an electronically generated display having two or more dimensions, e.g. on a television screen, showing representations related to the game characterized by details of game servers details of game services offered to the player
- A63F2300/575—Features of games using an electronically generated display having two or more dimensions, e.g. on a television screen, showing representations related to the game characterized by details of game servers details of game services offered to the player for trading virtual items
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Finance (AREA)
- General Business, Economics & Management (AREA)
- Multimedia (AREA)
- Accounting & Taxation (AREA)
- Computer Security & Cryptography (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application discloses a game commodity recommendation method, a game commodity recommendation device and electronic equipment, wherein the method comprises the following steps: acquiring payment characteristic data of a user; carrying out logarithmic normalization processing on the paid feature data to obtain a feature probability value of the paid feature data; and determining a preference label corresponding to the payment characteristic data according to the section of the characteristic probability value in a preset section, so as to send corresponding commodity recommendation information to the user according to the preference label.
Description
Technical Field
The present application relates to the field of computer technologies, and in particular, to a method and an apparatus for recommending game goods, and an electronic device.
Background
With the development of the internet, services that can be recommended to users are increasing. In order to meet the user requirements, in some fields, such as the game field, corresponding services are generally required to be formulated according to the commodity preferences of the user to meet the user requirements, so that commodity preference prediction needs to be performed on the user, corresponding virtual articles are recommended to the user according to different preference habits, and the user experience is improved.
In the existing preference prediction for users, the payment characteristics of the users are obtained for clustering, and the preference degree of the users to certain types of commodities is predicted according to the clustering result, so that corresponding virtual articles are recommended according to the predicted preference degree. However, in practical application, because the clustering mode is to perform clustering according to the distance between two data, and the preference degree value of the clustering mode has no upper limit in preference prediction, when the number of users is large and the payment characteristic of a certain user is high, the preference prediction after the whole clustering only presents results of two polarities, so that the bias of the preference prediction result of the user is high, and therefore, the game commodity recommended to the user is inconsistent with the current actual condition of the user, and the user experience is poor.
Disclosure of Invention
The present application aims to solve at least one of the technical problems in the prior art, and provides a game commodity recommendation method, device and electronic device, so as to improve the accuracy of game commodity recommendation and improve user experience.
The embodiment of the application provides a game commodity recommendation method, which comprises the following steps:
acquiring payment characteristic data of a user;
carrying out logarithmic normalization processing on the paid feature data to obtain a feature probability value of the paid feature data;
and determining a preference label corresponding to the payment characteristic data according to the section of the characteristic probability value in a preset section, so as to send corresponding commodity recommendation information to the user according to the preference label.
Further, the acquiring of the payment feature data of the user includes:
acquiring consumption data of the user for each commodity;
and acquiring the payment characteristic data of the user according to the preset weight of each commodity.
Further, in the embodiment of the present application, the method further includes:
and determining the preset weight according to the numerical difference between the consumption data and the preset consumption data, wherein the numerical difference is positively correlated with the preset weight.
Further, the consumption data is obtained from a plurality of applications associated with the server.
Further, the performing log normalization processing on the paid feature data to obtain a feature probability value of the paid feature data includes:
carrying out logarithmic transformation on the payment characteristic data and preset characteristic data to obtain a first characteristic value of the payment characteristic data and a second characteristic value of the preset characteristic data;
and according to the first characteristic value and the second characteristic value, carrying out normalization processing on the paid characteristic data to obtain a characteristic probability value of the paid characteristic data in the preset interval.
Further, the performing logarithmic transformation on the payment feature data and the preset feature data to obtain a first feature value of the payment feature data and a second feature value of the preset feature data includes:
according to N ═ logL(1+(Ssum) Performing logarithmic transformation on the payment characteristic data to obtain a first characteristic value of the payment characteristic data; and the number of the first and second groups,
according to N ═ logL(1+(Ssum’) And/2) carrying out logarithmic transformation on preset feature data to obtain a second feature value of the preset feature data, wherein N represents the first feature probability value, N' represents the second feature probability value, and SsumRepresenting said payment characteristic data, Ssum’And representing the preset characteristic data, wherein L is a preset base number.
Further, in the embodiment of the present application, the method further includes:
associating the payment characteristic data with the preference label.
Further, in an embodiment of the present application, there is provided a game commodity recommending apparatus, including:
the data acquisition module is used for acquiring payment characteristic data of a user;
the data processing module is used for carrying out logarithmic normalization processing on the paid feature data to obtain a feature probability value of the paid feature data;
and the commodity recommendation module is used for determining a preference label corresponding to the payment characteristic data according to the section of the characteristic probability value in a preset section, so as to send corresponding commodity recommendation information to the user according to the preference label.
Further, the data acquisition module is specifically configured to:
acquiring consumption data of the user for each commodity;
and acquiring the payment characteristic data of the user according to the preset weight of each commodity.
Further, the data acquisition module is specifically configured to:
and determining the preset weight according to the numerical difference between the consumption data and the preset consumption data, wherein the numerical difference is positively correlated with the preset weight.
Further, the data processing module is specifically configured to:
carrying out logarithmic transformation on the payment characteristic data and preset characteristic data to obtain a first characteristic value of the payment characteristic data and a second characteristic value of the preset characteristic data;
and according to the first characteristic value and the second characteristic value, carrying out normalization processing on the paid characteristic data to obtain a characteristic probability value of the paid characteristic data in the preset interval.
Further, the data processing module is specifically configured to:
according to N ═ logL(1+(Ssum) Performing logarithmic transformation on the payment characteristic data to obtain a first characteristic value of the payment characteristic data; and the number of the first and second groups,
according to N ═ logL(1+(Ssum’) And/2) carrying out logarithmic transformation on preset feature data to obtain a second feature value of the preset feature data, wherein N represents the first feature valueSign probability value, N' represents a second sign probability value, SsumRepresenting said payment characteristic data, Ssum’And representing the preset characteristic data, wherein L is a preset base number.
Further, an embodiment of the present application provides an electronic device, including: a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the game item recommendation method as described in the above embodiments when executing the program.
Further, the present application provides a computer-readable storage medium, which stores computer-executable instructions for causing a computer to execute the game commodity recommendation method according to the above embodiment.
Compared with the prior art, the embodiment obtains the payment characteristic data of the user, maps the characteristic probability value of the payment characteristic data to the interval (0,1) through normalization, and determines the preference label according to the interval where the characteristic probability value is located in the preset fetching piece, so that the obtained preference prediction result does not have higher deviation to cause extreme results, and finally, the game commodity recommendation information sent according to the preference label better conforms to the current actual situation of the user, and the user experience is effectively improved.
According to the embodiment, the payment characteristic data is obtained by obtaining the consumption data of the user for each commodity and the preset weight of each commodity, so that the obtained payment characteristic data is more accurate.
In the embodiment, the preset weight is determined according to the numerical difference between the consumption data and the preset consumption data, and the preset weight and the numerical difference form a positive correlation relationship, so that the preset weight is larger under the condition that the numerical difference is larger, and the influence of the consumption level of the user on the prediction result can be better reflected.
According to the embodiment, the consumption data are obtained from the plurality of applications associated with the server, so that the consumption data are more diversified in source and can reflect the consumption condition of the user more accurately.
According to the embodiment, the measurement of the characteristic data can be unified by carrying out logarithmic transformation on the paid characteristic data and the preset characteristic data, the numerical difference between the paid characteristic data and the preset characteristic data is reduced, and then the result after the logarithmic transformation is subjected to normalization processing, so that the situations that the deviation of the prediction result is high and the prediction result is inaccurate due to the fact that the preset characteristic data is too large are avoided more effectively.
The above embodiment avoids the calculation error by defining a specific logarithm transformation formula and setting a payment data characteristic +1 in the formula, and implements the smoothing processing of data by setting a payment technical characteristic divided by 2, i.e. averaging two commodities.
The above embodiments associate the paid feature data with the preference tag so that it will be used as reference data in the subsequent preference prediction process.
Drawings
The present application is further described with reference to the following figures and examples;
FIG. 1 is a diagram of an exemplary implementation of a method for game item recommendation;
FIG. 2 is a flow chart illustrating a method for recommending game goods according to an embodiment;
FIG. 3 is a diagram illustrating a variation of the amount of money consumed by a conventional preference prediction method;
FIG. 4 is a schematic diagram illustrating the variation of the amount of risk consumed after normalization processing is added to the existing preference prediction method;
FIG. 5 is a schematic diagram illustrating a variation of willingness to consume risks obtained by the game item recommendation method of the present invention in one embodiment;
FIG. 6 is a flowchart illustrating a method for recommending game merchandise according to another embodiment;
FIG. 7 is a block diagram showing the structure of a game article recommendation apparatus according to an embodiment;
FIG. 8 is a block diagram of a computer device in one embodiment.
Detailed Description
Reference will now be made in detail to the present embodiments of the present application, preferred embodiments of which are illustrated in the accompanying drawings, which are for the purpose of visually supplementing the description with figures and detailed description, so as to enable a person skilled in the art to visually and visually understand each and every feature and technical solution of the present application, but not to limit the scope of the present application.
In the existing preference prediction for users, the payment characteristics of the users are obtained for clustering, and the preference degree of the users to certain types of commodities is predicted according to the clustering result, so that corresponding virtual articles are recommended according to the predicted preference degree. However, in practical application, because the clustering mode is to perform clustering according to the distance between two data, and the preference degree value of the clustering mode has no upper limit in preference prediction, when the number of users is large and the payment characteristic of a certain user is high, the preference prediction after the whole clustering only presents results of two polarities, so that the bias of the preference prediction result of the user is high, and therefore, the game commodity recommended to the user is inconsistent with the current actual condition of the user, and the user experience is poor.
In order to solve the above technical problem, as shown in fig. 1, it is an application environment diagram of a game commodity recommendation method in one embodiment. Referring to fig. 1, the game item recommendation system includes a user terminal 110 and a server 120. The user terminal 110 and the server 120 are connected through a network. The user terminal 110 may be implemented as a stand-alone user terminal or as a user terminal cluster consisting of a plurality of user terminals. The user terminal 110 may be a desktop user terminal or a mobile user terminal, and the mobile user terminal may be at least one of a mobile phone, a tablet computer, a notebook computer, and the like. The server 120 may be implemented as a stand-alone server or a server cluster composed of a plurality of servers.
Hereinafter, the game commodity recommendation method provided in the embodiment of the present application will be described and explained in detail through several specific embodiments.
As shown in FIG. 2, in one embodiment, a game item recommendation method is provided. The embodiment is mainly illustrated by applying the method to computer equipment. The computer device may specifically be the server 120 in fig. 1 described above.
Referring to fig. 2, the game commodity recommendation method specifically includes the following steps:
and S11, acquiring payment characteristic data of the user.
In this embodiment, the server obtains the payment characteristic data of the user from the user terminal, where the payment characteristic data refers to the total payment situation of a certain type of goods in some fields of the user. For example, on the pseudo-ginseng game platform, the user can use the data such as average recharge amount and the like for the consumption situation of certain commodities in the games of Angel Sword and Douguo mainland. The categories of commodities can be classified into explicit commodities, rebate commodities and probabilistic commodities. The sum of the external commodities purchased by the user in the sword of Angel is 300 yuan, and the sum of the external commodities purchased by the user in the continental of fighting Rou is 500 yuan, so that 800 yuan obtained by adding the consumption sums can be used as the payment characteristic data Ssum of the user for the external commodities.
The mode that the server acquires the payment characteristic data from the user terminal can be used for acquiring the payment characteristic data within a period of time or at all times from a database of the user terminal; the server can also acquire the payment characteristic data of the user in real time when the user performs payment operation. In this embodiment, the manner in which the server obtains the payment feature data of the user is not particularly limited.
And S12, carrying out logarithmic normalization processing on the payment characteristic data to obtain the characteristic probability value of the payment characteristic data.
In this embodiment, the server performs log normalization on the payment feature data, wherein the log normalization is performed after the payment feature data is processed. The normalization process is used to map the paid feature data into the interval [0, 1], so that a conventional normalization process can be used, i.e. the proportion of each data in all data is obtained.
In one embodiment, the performing a logarithmic normalization process on the payment feature data to obtain a feature probability value of the payment feature data includes:
and carrying out logarithmic transformation on the payment characteristic data and the preset characteristic data to obtain a first characteristic value of the payment characteristic data and a second characteristic value of the preset characteristic data.
And according to the first characteristic value and the second characteristic value, carrying out normalization processing on the payment characteristic data to obtain a characteristic probability value of the payment characteristic data in a preset interval.
In this embodiment, the preset feature data is payment feature data corresponding to a maximum value in the historical payment feature data of the goods with payment feature data belonging to the same category, for example, for an explicit goods, the payment feature data is 800, the historical payment feature data is 100, 300, 500, and 1000, at this time, the maximum historical payment feature data is 1000, and therefore the preset feature data is 1000. The server acquires the payment characteristic data and the preset characteristic data from the database of the user terminal, and performs logarithmic transformation on the payment characteristic data and the preset characteristic data respectively, wherein numerical values obtained after the logarithmic transformation are a first characteristic value of the payment characteristic data and a second characteristic value of the preset characteristic data. And the server normalizes the first characteristic value through the second characteristic value, namely normalizes the payment characteristic data, and the numerical value obtained after normalization is the characteristic probability value of the payment characteristic data in a preset interval.
In this embodiment, the feature probability value may be expressed as Smean, and thus the formula for normalizing the first feature value by the second feature value may be Smean logL(Ssum1)/logL(Ssum2). max (), where Ssum1 ∈ Ssum2, Ssum1 represents pay-per-view feature data for a certain user, Ssum2 represents historical pay-per-view feature data over a period of time,. max () represents the maximum value, thus logL(Ssum2). max () represents the logarithmic transformation of the maximum value of the historical payment signature data. Wherein the preset base number L is set manually. For example, for an explicit product, the payment feature data is 800, the historical payment feature data is 100, 300, 500, and 1000, and in this case, the maximum historical payment feature data is 1000, and when L is 10, the first feature value obtained by logarithmically converting the payment feature data 800 is 2.903089987, the second feature value obtained by logarithmically converting the maximum historical feature data 1000 is 3, and the feature probability value Smean calculated according to the above-described formula of the normalization process is about 0.967. And thus 0.967 the payment characteristic data is in the preset interval [0, 1]]The characteristic probability value of (2).
In this embodiment, since there is a special activity in the game field for a period of time, which results in a rapid increase of consumption for the period of time, 1000 may be consumed during the activity with a preset weight of 0.6, and 200 may be consumed during the non-activity with a preset weight of 0.6, so that consumption values corresponding to the same preset weight in different periods of time may be different, and thus, the historical payment feature data may be historical payment feature data in a period of time, such as during the activity, rather than all historical payment feature data.
In this embodiment, when the payment feature data is stored in the database of the user terminal, all the payment feature data of the product of the category to which the payment feature data belongs are updated, and the maximum value of the historical payment feature data is determined. For example, when the payment characteristic data is 800 and the historical payment characteristic data is 100, 300, 500, 1000, then the maximum historical payment characteristic data is 1000; when the payment characteristic data is 800 and the historical payment characteristic data is 100, 300, 500, 700, then the maximum historical payment characteristic data is 800.
In this embodiment, the metric of the feature data can be unified by performing logarithmic transformation on the paid feature data and the preset feature data, and meanwhile, the numerical difference between the paid feature data and the preset feature data is reduced, and then the result after the logarithmic transformation is normalized, so that the situations that the offset of the prediction result is high and the prediction result is inaccurate due to the overlarge preset feature data are more effectively avoided.
In one embodiment, logarithmically transforming the payment characteristic data and the preset characteristic data to obtain a first characteristic value of the payment characteristic data and a second characteristic value of the preset characteristic data includes:
according to N ═ logL(1+(Ssum) Performing logarithmic transformation on the payment characteristic data to obtain a first characteristic value of the payment characteristic data; and the number of the first and second groups,
according to N ═ logL(1+(Ssum’) /2) carrying out logarithmic transformation on the preset characteristic data to obtain the preset characteristic dataWherein N represents a first feature probability value, N' represents a second feature probability value, SsumRepresenting pay characteristic data, Ssum’And representing preset characteristic data, wherein L is a preset base number.
In the present embodiment, for the logarithmic transformation, the formula N ═ log may also be adoptedL(1+(Ssum) Log transformation of pay feature data and using the formula N' logL(1+(Ssum’) Log-transforming the preset feature data.
In this embodiment, the server follows the formula N ═ logL(1+(Ssum) /2) logarithmically transforming the pay characteristic data and according to the formula N ═ logL(1+(Ssum’) And/2) carrying out logarithmic transformation on the preset characteristic data.
In the embodiment, a specific logarithmic transformation formula is defined, and a payment data characteristic +1 is set in the formula to avoid calculation errors, and the data smoothing processing is realized by setting a payment technical characteristic to be divided by 2, namely, calculating the average of two commodities.
And S13, determining a preference label corresponding to the payment characteristic data according to the section where the characteristic probability value is located in the preset section, and sending corresponding commodity recommendation information to the user according to the preference label.
In this embodiment, the preset interval is an interval in a range of [0, 1], the interval is divided into sub-intervals according to a certain condition, and each interval is provided with a corresponding preference tag, where the preference tags refer to different degrees of consumption intentions of a user on a certain type of goods. For example, in the range of the preset interval [0, 1], the preset division is divided into four intervals [0, 0.25), [0.25, 0.5), [0.5, 0.75) and [0.75, 1] according to the consumption intention of the user on the explicit goods, and the corresponding preference tags are sequentially no preference, general preference, better preference and preference.
In this embodiment, the server determines the preference tag corresponding to the payment feature data according to the section in which the feature probability value obtained by the payment feature data Ssum is located, and sends the recommendation information of the related product according to the preference tag of the user for the certain type of product. For example, the characteristic probability value corresponding to the payment characteristic data 800 of the explicit product by the user is 0.967, and it can be known that the characteristic probability value is in the section of [0.75, 1], and the corresponding preference label is preference, so that the user has a higher consumption intention for the explicit product, and further sends the product recommendation information of the explicit product to the user.
In the above embodiment, as shown in fig. 3, when the existing preference prediction method is adopted, the data clustering operation therein may cause the risk consumption amount to have a power law distribution, which affects the accuracy of the preference prediction. When the risk consumption amount is normalized to replace clustering operation, as shown in fig. 4, the risk consumption amount is still distributed in the power law within the range of [0, 1], and the preference prediction result is still different from the actual situation, thereby affecting the accuracy of the commodity recommendation information. After the operation of logarithmic transformation is added before normalization processing, as shown in fig. 5, the risk consumption will be within the range of [0, 1] without deviation, and better conforms to the current actual situation of the user.
In the embodiment, the paying feature data of the user is obtained, the feature probability value of the paying feature data is mapped to the interval [0, 1] through normalization, and the preference label is determined according to the interval of the feature probability value in the preset fetching piece, so that the obtained preference prediction result is not extremely high in deviation amount, the result is not extremely high, the game commodity recommendation information sent according to the preference label finally better accords with the current actual condition of the user, and the user experience is effectively improved.
In one embodiment, the practical application can be performed in a reverse direction, for example, a gift package is pushed to 6000 users with explicit preferences, 6000 users can be set, the explicit characteristic probability value is calculated in a reverse direction to be 0.75, and then, in combination with the high-consumption users (assuming 20000 users), 4800 high-value users with characteristic probability values >0.75 can be obtained, and even more kinds of combinations can be obtained, so that the application is more convenient.
In another embodiment, as shown in FIG. 6, a game item recommendation method is provided. The embodiment is mainly illustrated by applying the method to computer equipment. The computer device may specifically be the server 120 in fig. 1 described above.
Referring to fig. 6, the game commodity recommendation method specifically includes the following steps:
and S21, acquiring consumption data of the user for each commodity.
In this embodiment, the server obtains consumption data of the user for each commodity from a database of the user terminal, where the consumption data refers to a consumption amount of a certain commodity of the user in some fields. The commodity categories can be classified into explicit commodities, rebate commodities and probabilistic commodities, and the consumption data can be data such as average recharge amount, total recharge amount, first recharge amount and the like. For example, the server obtains the total recharge amount of the explicit commodities in the games of Angel Sword and fighting Luo mainland on the pseudo-ginseng game platform, wherein the amount of the explicit commodities purchased by the user in the Angel Sword is 300 yuan, and the amount of the explicit commodities purchased by the user in the fighting Luo mainland is 500 yuan, so that 800 yuan obtained by adding the consumption amounts of the explicit commodities can be used as the consumption data of the user.
In one embodiment, the consumption data is obtained from a plurality of applications associated with the server.
In this embodiment, the server may obtain the consumption data from a plurality of applications associated therewith, where the plurality of applications may be a plurality of applications installed in different terminals respectively connected to the server network. In addition, the server acquires consumption data from a plurality of applications, and the plurality of applications can be a plurality of applications logged in by adopting the same account number, namely the consumption data of the same user in different applications is acquired; the plurality of applications may be a plurality of applications logged in by using an account, that is, consumption data of different users in different applications is acquired. In this embodiment, the association manner between the application and the server is not specifically limited, and the relationship between the plurality of applications is also not specifically limited.
In the embodiment, the consumption data is acquired from a plurality of applications associated with the server, so that the consumption data has more various sources and can reflect the consumption condition of the user more accurately.
And S22, acquiring payment characteristic data of the user according to the preset weight of each commodity.
Wherein the preset weight of each commodity is set and adjusted by people. In this embodiment, the commodities are risk commodities including risk commodities a and B, the commodity a is a commodity which is always offered at ordinary times, the commodity a is offered with a discount at ordinary times, the discount is reduced as the number of purchases increases, the commodity B is offered at holidays, and the risk is higher and the profit is higher. And calculating the average consumption SA of the commodity A as the total consumption/the number of days for purchasing the risk commodities, and calculating the average consumption SB of the commodity B as the total consumption/the number of days for purchasing the risk commodities. The payment feature data Ssum is 0.36SB +0.64 SA. After many calculation analyses, the two values of 0.36 and 0.64 are the two best weights identified by practice.
In this embodiment, the payment feature data is obtained by obtaining the consumption data of the user for each commodity and the preset weight of each commodity, so that the obtained payment feature data is more accurate.
In one embodiment, the game item recommendation method further includes:
and determining a preset weight according to the numerical difference between the consumption data and the preset consumption data, wherein the numerical difference is positively correlated with the preset weight.
In this embodiment, when the consumption data is larger, it can be known that the consumption will of the type of currently consumed product is stronger, so that the influence of the payment feature data corresponding to the consumption data on the preference tag of the type of product is determined to be closer to the preference, therefore, the consumption data can be classified by setting a preset consumption data, the preset weight is adjusted according to the difference value between the consumption data and the preset consumption data, and the larger the difference value between the consumption data and the preset consumption data is, the larger the preset weight is. The preset consumption data can be determined according to historical payment characteristic data corresponding to an endpoint value of the section where the preference tag is located, or can be set manually. For example, for the above-mentioned article a, if SA > J, SA ═ SA × 1.2, J is a value exceeding a certain amount of the non-discount quota, it means that the user is willing to purchase a risky article for the article at the original price, i.e. the risky article is more willing to be consumed. Wherein SA is SA 1.2, which corresponds to the preset weight 1.2, that is, the preset weight of the article a is adjusted.
In the embodiment, the preset weight is determined according to the numerical difference between the consumption data and the preset consumption data, and the preset weight and the numerical difference form a positive correlation relationship, so that the preset weight is larger under the condition that the numerical difference is larger, and the influence of the consumption level of the user on the prediction result can be better reflected.
And S23, carrying out logarithmic normalization processing on the payment characteristic data to obtain the characteristic probability value of the payment characteristic data.
In this embodiment, the server performs log normalization on the payment feature data, wherein the log normalization is performed after the payment feature data is processed. The normalization process is used to map the paid feature data into the interval [0, 1], so that a conventional normalization process can be used, i.e. the proportion of each data in all data is obtained.
In one embodiment, the performing a logarithmic normalization process on the payment feature data to obtain a feature probability value of the payment feature data includes:
and carrying out logarithmic transformation on the payment characteristic data and the preset characteristic data to obtain a first characteristic value of the payment characteristic data and a second characteristic value of the preset characteristic data.
And according to the first characteristic value and the second characteristic value, carrying out normalization processing on the payment characteristic data to obtain a characteristic probability value of the payment characteristic data in a preset interval.
In this embodiment, the preset feature data is payment feature data corresponding to a maximum value in the historical payment feature data of the goods with payment feature data belonging to the same category, for example, for an explicit goods, the payment feature data is 800, the historical payment feature data is 100, 300, 500, and 1000, at this time, the maximum historical payment feature data is 1000, and therefore the preset feature data is 1000. The server acquires the payment characteristic data and the preset characteristic data from the database of the user terminal, and performs logarithmic transformation on the payment characteristic data and the preset characteristic data respectively, wherein numerical values obtained after the logarithmic transformation are a first characteristic value of the payment characteristic data and a second characteristic value of the preset characteristic data. And the server normalizes the first characteristic value through the second characteristic value, namely normalizes the payment characteristic data, and the numerical value obtained after normalization is the characteristic probability value of the payment characteristic data in a preset interval.
In this embodiment, the feature probability value may be expressed as Smean, and thus the formula for normalizing the first feature value by the second feature value may be Smean logL(Ssum1)/logL(Ssum2). max (), where Ssum1 represents the pay-per-view characteristic data for a certain user, Ssum2 represents the historical pay-per-view characteristic data over a period of time,. max () represents the maximum value, and thus logL(Ssum2). max () represents the logarithmic transformation of the maximum value of the historical payment signature data. Wherein the preset base number L is set manually. For example, for an explicit product, the payment feature data is 800, the historical payment feature data is 100, 300, 500, and 1000, and in this case, the maximum historical payment feature data is 1000, and when L is 10, the first feature value obtained by logarithmically converting the payment feature data 800 is 2.903089987, the second feature value obtained by logarithmically converting the maximum historical feature data 1000 is 3, and the feature probability value Smean calculated according to the above-described formula of the normalization process is about 0.967. Therefore, 0.967 is the payment characteristic data in the preset interval [0, 1]]The characteristic probability value of (2).
In this embodiment, when the payment feature data is stored in the database of the user terminal, all the payment feature data of the product of the category to which the payment feature data belongs are updated, and the maximum value of the historical payment feature data is determined. For example, when the payment characteristic data is 800 and the historical payment characteristic data is 100, 300, 500, 1000, then the maximum historical payment characteristic data is 1000; when the payment characteristic data is 800 and the historical payment characteristic data is 100, 300, 500, 700, then the maximum historical payment characteristic data is 800.
In this embodiment, the metric of the feature data can be unified by performing logarithmic transformation on the paid feature data and the preset feature data, and meanwhile, the numerical difference between the paid feature data and the preset feature data is reduced, and then the result after the logarithmic transformation is normalized, so that the situations that the offset of the prediction result is high and the prediction result is inaccurate due to the overlarge preset feature data are more effectively avoided.
In one embodiment, logarithmically transforming the payment characteristic data and the preset characteristic data to obtain a first characteristic value of the payment characteristic data and a second characteristic value of the preset characteristic data includes:
according to N ═ logL(1+(Ssum) Performing logarithmic transformation on the payment characteristic data to obtain a first characteristic value of the payment characteristic data; and the number of the first and second groups,
according to N ═ logL(1+(Ssum’) And/2) carrying out logarithmic transformation on the preset feature data to obtain a second feature value of the preset feature data, wherein N represents a first feature probability value, N' represents a second feature probability value, and SsumRepresenting pay characteristic data, Ssum’And representing preset characteristic data, wherein L is a preset base number.
In this embodiment, the server follows the formula N ═ logL(1+(Ssum) /2) logarithmically transforming the pay characteristic data and according to the formula N ═ logL(1+(Ssum’) And/2) carrying out logarithmic transformation on the preset characteristic data.
In the embodiment, a specific logarithmic transformation formula is defined, and a payment data characteristic +1 is set in the formula to avoid calculation errors, and the data smoothing processing is realized by setting a payment technical characteristic to be divided by 2, namely, calculating the average of two commodities.
And S24, determining a preference label corresponding to the payment characteristic data according to the section where the characteristic probability value is located in the preset section, and sending corresponding commodity recommendation information to the user according to the preference label.
This step is the same as the above embodiment, and the detailed analysis may refer to the above embodiment, and is not repeated herein to avoid repetition.
In one embodiment, the game item recommendation method further includes:
the payment characteristic data is associated with the preference tag.
In this embodiment, the server associates the payment feature data with the obtained corresponding preference tag, so that when the server acquires the same payment feature data later, the server can directly output the preference tag, and the step of determining the preference tag is omitted. For example, the payment characteristic data 800 of the explicit merchandise is associated with the preference tag that is a preference, so that when the payment characteristic data of the explicit merchandise is subsequently acquired as 800, the preference tag is directly output as a preference. The payment feature data and the preference tag may be associated in a data splicing manner or in a manner of being stored in the same path, and in this embodiment, the association manner of the payment feature data and the preference tag is not specifically limited.
In the present embodiment, the payment feature data is associated with the preference label, so that the payment feature data is used as reference data in the subsequent preference prediction process.
In one embodiment, as shown in fig. 7, there is provided a game article recommendation device including:
a data obtaining module 101, configured to obtain payment feature data of a user.
And the data processing module 102 is configured to perform logarithmic normalization processing on the payment feature data to obtain a feature probability value of the payment feature data.
And the commodity recommendation module 103 is configured to determine a preference tag corresponding to the payment feature data according to the section in which the feature probability value is located in the preset section, so as to send corresponding commodity recommendation information to the user according to the preference tag.
In one embodiment, the data obtaining module 101 is further configured to:
and acquiring consumption data of the user aiming at each commodity.
And acquiring payment characteristic data of the user according to the preset weight of each commodity.
In one embodiment, the data obtaining module 101 is further configured to:
and determining a preset weight according to the numerical difference between the consumption data and the preset consumption data, wherein the numerical difference is positively correlated with the preset weight.
In one embodiment, the data processing module 102 is further configured to:
and carrying out logarithmic transformation on the payment characteristic data and the preset characteristic data to obtain a first characteristic value of the payment characteristic data and a second characteristic value of the preset characteristic data.
And according to the first characteristic value and the second characteristic value, carrying out normalization processing on the payment characteristic data to obtain a characteristic probability value of the payment characteristic data in a preset interval.
In one embodiment, the data processing module 102 is further configured to:
according to N ═ logL(1+(Ssum) Performing logarithmic transformation on the payment characteristic data to obtain a first characteristic value of the payment characteristic data; and the number of the first and second groups,
according to N ═ logL(1+(Ssum’) And/2) carrying out logarithmic transformation on the preset feature data to obtain a second feature value of the preset feature data, wherein N represents a first feature probability value, N' represents a second feature probability value, and SsumRepresenting pay characteristic data, Ssum’And representing preset characteristic data, wherein L is a preset base number.
In one embodiment, a computer apparatus is provided, as shown in fig. 8, which includes a processor, a memory, a network interface, an input device, and a display screen connected by a system bus. Wherein the memory includes a non-volatile storage medium and an internal memory. The non-volatile storage medium of the computer device stores an operating system and may also store a computer program that, when executed by the processor, causes the processor to implement a game good recommendation method. The internal memory may also have a computer program stored therein, which when executed by the processor, causes the processor to execute a game item recommendation method. Those skilled in the art will appreciate that the architecture shown in fig. 8 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
In one embodiment, the game item recommendation device provided by the present application may be implemented in the form of a computer program that is executable on a computer device such as that shown in fig. 8. The memory of the computer device may store therein the respective program modules constituting the game article recommending apparatus. The computer program constituted by the respective program modules causes the processor to execute the steps in the game article recommendation method of the respective embodiments of the present application described in the present specification.
In one embodiment, there is provided an electronic device including: the game commodity recommendation method comprises a memory, a processor and a computer program which is stored on the memory and can run on the processor, wherein the processor executes the program to execute the steps of the game commodity recommendation method. Here, the steps of the game commodity recommendation method may be the steps in the game commodity recommendation methods of the above-described respective embodiments.
In one embodiment, a computer-readable storage medium is provided, which stores computer-executable instructions for causing a computer to perform the steps of the game item recommendation method described above. Here, the steps of the game commodity recommendation method may be the steps in the game commodity recommendation methods of the above-described respective embodiments.
The foregoing is a preferred embodiment of the present application, and it should be noted that, for those skilled in the art, various modifications and decorations can be made without departing from the principle of the present application, and these modifications and decorations are also regarded as the protection scope of the present application.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), or the like.
Claims (13)
1. A game article recommendation method, comprising:
acquiring payment characteristic data of a user;
carrying out logarithmic normalization processing on the paid feature data to obtain a feature probability value of the paid feature data;
and determining a preference label corresponding to the payment characteristic data according to the section of the characteristic probability value in a preset section, so as to send corresponding commodity recommendation information to the user according to the preference label.
2. The game item recommendation method according to claim 1, wherein the acquiring of the payment feature data of the user comprises:
acquiring consumption data of the user for each commodity;
and acquiring the payment characteristic data of the user according to the preset weight of each commodity.
3. The game item recommendation method according to claim 2, further comprising:
and determining the preset weight according to the numerical difference between the consumption data and the preset consumption data, wherein the numerical difference is positively correlated with the preset weight.
4. A game item recommendation method according to claim 2, wherein said consumption data is obtained from a plurality of server-associated applications.
5. The game commodity recommendation method according to claim 1 or 3, wherein the performing a logarithmic normalization process on the payment feature data to obtain the feature probability value of the payment feature data includes:
carrying out logarithmic transformation on the payment characteristic data and preset characteristic data to obtain a first characteristic value of the payment characteristic data and a second characteristic value of the preset characteristic data;
and according to the first characteristic value and the second characteristic value, carrying out normalization processing on the paid characteristic data to obtain a characteristic probability value of the paid characteristic data in the preset interval.
6. The game commodity recommendation method according to claim 5, wherein the logarithmically transforming the payment feature data and the preset feature data to obtain a first feature value of the payment feature data and a second feature value of the preset feature data includes:
according to N ═ logL(1+(Ssum) Performing logarithmic transformation on the payment characteristic data to obtain a first characteristic value of the payment characteristic data; and the number of the first and second groups,
according to N ═ logL(1+(Ssum’) And/2) carrying out logarithmic transformation on preset feature data to obtain a second feature value of the preset feature data, wherein N represents the first feature probability value, N' represents the second feature probability value, and SsumRepresenting said payment characteristic data, Ssum’And representing the preset characteristic data, wherein L is a preset base number.
7. The game item recommendation method according to claim 1, further comprising:
associating the payment characteristic data with the preference label.
8. A game article recommendation device, comprising:
the data acquisition module is used for acquiring payment characteristic data of a user;
the data processing module is used for carrying out logarithmic normalization processing on the paid feature data to obtain a feature probability value of the paid feature data;
and the commodity recommendation module is used for determining a preference label corresponding to the payment characteristic data according to the section of the characteristic probability value in a preset section, so as to send corresponding commodity recommendation information to the user according to the preference label.
9. The game item recommendation device of claim 8, wherein the data acquisition module is specifically configured to:
acquiring consumption data of the user for each commodity;
and acquiring the payment characteristic data of the user according to the preset weight of each commodity.
10. The game item recommendation device of claim 9, wherein the data acquisition module is specifically configured to:
and determining the preset weight according to the numerical difference between the consumption data and the preset consumption data, wherein the numerical difference is positively correlated with the preset weight.
11. A game item recommendation device according to claim 8 or 10, wherein the data processing module is specifically configured to:
carrying out logarithmic transformation on the payment characteristic data and preset characteristic data to obtain a first characteristic value of the payment characteristic data and a second characteristic value of the preset characteristic data;
and according to the first characteristic value and the second characteristic value, carrying out normalization processing on the paid characteristic data to obtain a characteristic probability value of the paid characteristic data in the preset interval.
12. The game item recommendation device of claim 11, wherein the data processing module is specifically configured to:
according to N ═ logL(1+(Ssum) Performing logarithmic transformation on the payment characteristic data to obtain a first characteristic value of the payment characteristic data; and the number of the first and second groups,
according to N ═ logL(1+(Ssum’) /2) carrying out logarithmic transformation on preset characteristic data to obtain the preset characteristicCharacterizing a second feature value of the data, wherein N represents the first feature probability value, N' represents a second feature probability value, SsumRepresenting said payment characteristic data, Ssum’And representing the preset characteristic data, wherein L is a preset base number.
13. An electronic device, comprising: memory, processor and computer program stored on the memory and executable on the processor, characterized in that the processor implements the game item recommendation method according to any one of claims 1 to 7 when executing the program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011384767.0A CN112446764A (en) | 2020-11-30 | 2020-11-30 | Game commodity recommendation method and device and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011384767.0A CN112446764A (en) | 2020-11-30 | 2020-11-30 | Game commodity recommendation method and device and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112446764A true CN112446764A (en) | 2021-03-05 |
Family
ID=74740108
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011384767.0A Pending CN112446764A (en) | 2020-11-30 | 2020-11-30 | Game commodity recommendation method and device and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112446764A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113368496A (en) * | 2021-05-14 | 2021-09-10 | 广州三七互娱科技有限公司 | Weather rendering method and device for game scene and electronic equipment |
| CN113384894A (en) * | 2021-06-01 | 2021-09-14 | 广州三七极耀网络科技有限公司 | Virtual commodity recommendation method, device, equipment and storage medium |
| CN113407826A (en) * | 2021-06-09 | 2021-09-17 | 广州三七极创网络科技有限公司 | Virtual commodity recommendation method, device, equipment and storage medium |
| CN113805995A (en) * | 2021-09-09 | 2021-12-17 | 维沃移动通信有限公司 | Display method and device and electronic equipment |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105719164A (en) * | 2016-01-21 | 2016-06-29 | 海信集团有限公司 | Paid multimedia resource recommending method and paid multimedia resource recommending device |
| CN109493123A (en) * | 2018-10-23 | 2019-03-19 | 佛山欧神诺云商科技有限公司 | A kind of Method of Commodity Recommendation and device based on big data |
| CN111127152A (en) * | 2019-12-23 | 2020-05-08 | 深圳市赛维网络科技有限公司 | Commodity recommendation method, device and equipment based on user preference prediction and readable medium |
| CN111408143A (en) * | 2020-03-13 | 2020-07-14 | 网易（杭州）网络有限公司 | Game payment prediction method, model training method and device |
-
2020
- 2020-11-30 CN CN202011384767.0A patent/CN112446764A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105719164A (en) * | 2016-01-21 | 2016-06-29 | 海信集团有限公司 | Paid multimedia resource recommending method and paid multimedia resource recommending device |
| CN109493123A (en) * | 2018-10-23 | 2019-03-19 | 佛山欧神诺云商科技有限公司 | A kind of Method of Commodity Recommendation and device based on big data |
| CN111127152A (en) * | 2019-12-23 | 2020-05-08 | 深圳市赛维网络科技有限公司 | Commodity recommendation method, device and equipment based on user preference prediction and readable medium |
| CN111408143A (en) * | 2020-03-13 | 2020-07-14 | 网易（杭州）网络有限公司 | Game payment prediction method, model training method and device |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113368496A (en) * | 2021-05-14 | 2021-09-10 | 广州三七互娱科技有限公司 | Weather rendering method and device for game scene and electronic equipment |
| CN113384894A (en) * | 2021-06-01 | 2021-09-14 | 广州三七极耀网络科技有限公司 | Virtual commodity recommendation method, device, equipment and storage medium |
| CN113407826A (en) * | 2021-06-09 | 2021-09-17 | 广州三七极创网络科技有限公司 | Virtual commodity recommendation method, device, equipment and storage medium |
| CN113805995A (en) * | 2021-09-09 | 2021-12-17 | 维沃移动通信有限公司 | Display method and device and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112446764A (en) | Game commodity recommendation method and device and electronic equipment | |
| US20210035163A1 (en) | Predictive platform for determining incremental lift | |
| CN106897905B (en) | Method and device for pushing information and electronic equipment | |
| US20160275570A1 (en) | Methods and systems for dynamic auction floors | |
| US20160267519A1 (en) | Allocating online advertising budget based on return on investment (roi) | |
| CN109213936B (en) | Commodity searching method and device | |
| CN113095893A (en) | Method and device for determining sales of articles | |
| CN111292149A (en) | Method and device for generating return processing information | |
| CN112598472A (en) | Product recommendation method, device, system, medium and program product | |
| CN110598183A (en) | Flow distribution method, device, equipment and storage medium | |
| CN107679916A (en) | For obtaining the method and device of user interest degree | |
| CN113763019A (en) | User information management method and device | |
| CN114170010A (en) | Matching transaction method and device, electronic equipment and storage medium | |
| WO2022156589A1 (en) | Method and device for determining live broadcast click rate | |
| CN110827102A (en) | Method and device for adjusting goods price ratio | |
| CN110555747A (en) | method and device for determining target user | |
| CN108665312A (en) | Method and apparatus for generating information | |
| US20120265588A1 (en) | System and method for recommending new connections in an advertising exchange | |
| CN112287208A (en) | User portrait generation method and device, electronic equipment and storage medium | |
| CN115936764A (en) | Product promotion method and device | |
| CN116128135A (en) | Data processing method and device, electronic equipment and storage medium | |
| CN110880119A (en) | Data processing method and device | |
| CN114549125A (en) | Item recommendation method and device, electronic equipment and computer-readable storage medium | |
| CN114943582A (en) | Information recommendation method and system and recommendation server | |
| KR102102788B1 (en) | Artificial intelligence based financial product ordering system and method |
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 |