KR102641885B1 - Method for providing payment analysis information in game providing apparatus and game providing apparatus - Google Patents

Abstract

By applying log data containing behavior information of a plurality of users for the first game to a payment probability model that predicts the possibility of a user performing a payment action in the first game, prediction of the possibility of payment of the plurality of users is made. Obtaining information; Classifying the plurality of users into a plurality of groups according to prediction information about the possibility of payment of the plurality of users; Obtaining characteristic information of the first group based on log data of a user belonging to the first group among the plurality of groups; and providing characteristic information of the first group. A method of operating a game providing device is provided.

Description

Method for providing payment analysis information in game provision device and game provision device {METHOD FOR PROVIDING PAYMENT ANALYSIS INFORMATION IN GAME PROVIDING APPARATUS AND GAME PROVIDING APPARATUS}

It relates to a method of providing payment analysis information in a game providing device and a game providing device.

The types and amounts of data that can be obtained from a given game are vast. For example, various game-related data, such as the number of simultaneous game users for a given game, in-game item purchase information, and user activity information in the game, can be used to analyze the user's payment behavior in the game.

Therefore, there is a need for technology that can more quickly and accurately process data used to analyze users' payment behavior within games.

We want to accurately predict the payment possibility of multiple users for a given game and determine the characteristics of groups classified according to payment possibility.

We want to accurately predict the purchasing power of multiple users for a given game, which represents the expected payment amount, and provide a payment conversion rate that represents the actual payment amount compared to the expected payment amount.

According to one side, log data containing behavior information of a plurality of users for the first game is applied to a payment probability model that predicts the possibility of a user performing a payment action in the first game, and the plurality of users make payments. Obtaining predictive information about possibilities; Classifying the plurality of users into a plurality of groups according to prediction information about the possibility of payment of the plurality of users; Obtaining characteristic information of the first group based on log data of a user belonging to the first group among the plurality of groups; and providing characteristic information of the first group. A method of operating a game providing device is provided.

According to another aspect, based on payment data of a plurality of users for a predetermined period for one game, obtaining purchasing power information indicating an expected payment amount of the plurality of users; classifying the plurality of users into a plurality of groups according to the purchasing power information; Obtaining payment conversion rate information representing the actual payment amount compared to the expected payment amount for each of the plurality of groups; A method of operating a game providing device is provided, including providing the payment conversion rate information.

According to another aspect, a computer program stored in a medium is provided to execute the operation method of the game providing device in combination with hardware.

According to another aspect, a computer-readable recording medium is provided that records a program for executing a method of operating the game providing device on a computer.

According to one side, a communication device; user interface device; storage; processor; and a memory storing instructions executable by the processor, wherein the processor generates log data including behavior information of a plurality of users for the first game by executing the instructions. By applying a payment probability model that predicts the possibility of performing a payment action, prediction information about the payment possibility of the plurality of users is obtained, and according to the prediction information about the payment possibility of the plurality of users, the plurality of users Classify into a plurality of groups, obtain characteristic information of the first group based on log data of users belonging to a first group among the plurality of groups, and obtain characteristic information of the first group through the user interface device. A game providing device is provided.

According to another side, a communication device; user interface device; storage; processor; and a memory storing instructions executable by the processor, wherein the processor, by executing the instructions, determines, based on payment data of the plurality of users for a predetermined period of time, the plurality of users for the first game. Obtain purchasing power information indicating the expected payment amount, classify the plurality of users into a plurality of groups according to the purchasing power information, and provide payment conversion rate information indicating the actual payment amount compared to the expected payment amount for each of the plurality of groups. A game providing device is provided, which acquires and provides information on the payment conversion rate through the user interface device.

A game providing device can accurately predict the payment possibility of a plurality of users for a given game and provide characteristic information of groups classified according to payment possibility. Characteristic information of groups classified according to payment possibility can be used in game development and marketing strategies to increase the payment possibility of multiple users.

The game providing device can accurately predict the purchasing power representing the expected payment amount of a plurality of users for a predetermined game and provide payment conversion rate information representing the actual payment amount compared to the expected payment amount. Payment conversion rate information can be used in game development and marketing strategies to increase payment conversion rates.

The present disclosure may be readily understood by combination of the following detailed description and accompanying drawings, where reference numerals refer to structural elements.
1 shows an online game providing system including a server and a client, according to one embodiment.
FIG. 2 is a flowchart illustrating a method of operating a game providing device that provides characteristic information of a plurality of groups classified according to payment possibilities of a plurality of users, according to an embodiment.
FIG. 3 is a diagram illustrating a process of predicting payment possibilities of a plurality of users in a game providing device and classifying the plurality of users into a plurality of groups according to the prediction results, according to an embodiment.
FIG. 4 is a diagram illustrating a method of displaying information of a plurality of groups in a game providing device, according to an embodiment.
Figure 5 is a flowchart for explaining a method of operating a game providing device that generates a payment probability model, according to an embodiment.
FIG. 6 is a diagram schematically showing an artificial neural network with a plurality of hierarchical structures used by a game providing device to learn a payment probability model, according to an embodiment.
FIG. 7 is a diagram illustrating a process in which a game providing device learns a payment probability model, according to an embodiment.
FIG. 8 is a diagram illustrating characteristic information of a plurality of groups provided by a game providing device, according to an embodiment.
FIG. 9 is a diagram illustrating characteristic information of a plurality of groups provided by a game providing device, according to another embodiment.
FIG. 10 is a flowchart illustrating a method of operating a game providing device that obtains information on factors that influenced the payment behavior of a paying user, according to an embodiment.
FIG. 11 is a diagram illustrating a process of obtaining information on factors that influenced the payment behavior of a payment user in a game providing device, according to an embodiment.
FIG. 12 is a flowchart illustrating a method of operating a game providing device that provides purchasing power information and payment conversion rate information of a plurality of users, according to an embodiment.
FIG. 13 is a diagram illustrating a process of providing payment conversion rate information in a game providing device, according to an embodiment.
Figure 14 is a block diagram showing the configuration of a game providing device, according to one embodiment.

Hereinafter, various embodiments will be described in detail with reference to the drawings. The embodiments described below may be modified and implemented in various different forms. In order to more clearly explain the characteristics of the embodiments, detailed descriptions of matters widely known to those skilled in the art to which the following embodiments belong will be omitted.

Meanwhile, in this specification, when a configuration is said to be “connected” to another configuration, this includes not only the case of being “directly connected,” but also the case of being “connected with another configuration in between.” In addition, when a configuration “includes” another configuration, this means that other configurations may be further included rather than excluding other configurations, unless specifically stated to the contrary.

Additionally, terms including ordinal numbers such as 'first' or 'second' used in this specification may be used to describe various components, but the components should not be limited by the terms. The above terms are used only for the purpose of distinguishing one component from another.

Throughout the specification, “game providing device” may refer to a device used to provide a game service to a user's client and manage the game service. The “game providing device” may be a server. In addition, the “game providing device” may be implemented as a special-purpose computer, general-purpose computer, supercomputer, mainframe computer, etc., but is not limited thereto. .

Throughout the specification, a user's “log data” for a given game may be data that includes a record of the user's activities within the given game. For example, “log data” includes a record of the user logging into a given game, a record of the user purchasing items within a certain game, a record of the user making payments within a certain game, and a record of the user playing within a certain game. , information or data about the user logging out of a certain game, etc. may be included.

Throughout the specification, "machine learning" may refer to the art of studying and building systems or algorithms for systems that learn, make predictions, and improve their own performance based on data about a game.

Throughout the specification, a “payment probability model” for a given game may mean a model that predicts the likelihood that a user will perform a payment action in a given game. In addition, the "payment probability model" for a given game is based on the similarity between the behavior information of a plurality of users for a certain game and the behavior information of users who performed payment actions in a certain game during a certain period of time. It may refer to a model that predicts the payment possibility of multiple users.

Throughout the specification, “characteristic information” of a given group may mean characteristic or attribute information of a given group that can be distinguished from other groups.

Throughout the specification, the "payment behavior prediction model" for a given game predicts the payment behavior of a plurality of users based on log data of a plurality of users and determines information on factors that influenced the payment behavior of users in a given game. It may be a model used for.

Throughout the specification, “purchasing power information” may refer to information on the user's expected payment amount in a given game over a certain period of time.

Throughout the specification, “payment conversion rate information” may refer to ratio information indicating the actual payment amount compared to the expected payment amount in a given game.

1 shows an online game providing system including a server and a client, according to one embodiment.

An online game providing system according to an embodiment may include a server 4000 and a plurality of clients 2000a, 2000b, 2000c, 2000d, 2000e, and 2000f, as shown in FIG. 1. The server 4000 can provide the game to a plurality of clients 2000a, 2000b, 2000c, 2000d, 2000e, and 2000f by servicing the game over a network.

1 shows a desktop computer 2000a, a tablet 2000b, a mobile phone 2000c, a laptop 2000d, a smartphone 2000e, and a television 2000f as examples of clients used by a user; is not limited to this and may include various electronic devices such as PDAs (Personal Digital Assistants), PMPs (Portable Multimedia Players), navigation, MP3 players, digital cameras, refrigerators, washing machines, and vacuum cleaners.

A game (eg, an online game) may be played on multiple clients (2000a, 2000b, 2000c, 2000d, 2000e, and 2000f). To run a game, a program, such as, but not limited to, an application or launcher, may be installed on the clients 2000a, 2000b, 2000c, 2000d, 2000e, and 2000f. Users can enjoy games provided by the server 4000 by running programs installed on the clients 2000a, 2000b, 2000c, 2000d, 2000e, and 2000f. Various embodiments of the present disclosure described later may be implemented within a program for executing a game, or may be implemented through a program separate from the program.

Server 4000 may be operated by a game provider, and server 4000 may include, but is not limited to, a single server, an aggregate of servers, and a cloud server. The server 4000 provides games to users and may include a database that stores users' data. Additionally, the server 4000 may further include a payment server or payment server that generates and processes payment events. The payment server or payment server may be a separate server from the server 4000 and may mediate payments or payments between the server 4000 and the clients 2000a, 2000b, 2000c, 2000d, 2000e, and 2000f. The server 4000 may further include a database (DB) server that stores and manages user data, and the DB server may be a separate server from the server 4000.

A program for servicing games may be run in the server 4000. Various embodiments of the present disclosure described later may be implemented within a program for servicing a game, or may be implemented through a program separate from the program.

Network means a connection established (or formed) using a predetermined communication method, between clients (2000a, 2000b, 2000c, 2000d, 2000e, and 2000f) or between clients (2000a, 2000b, 2000c, 2000d, 2000e) , and 2000f) and the server 4000 may refer to a communication network connected through a predetermined communication means that transmits data.

Communication means include communication through a predetermined communication standard, a predetermined frequency band, a predetermined protocol or a predetermined channel, short-distance communication, long-distance communication, wireless communication, and wired communication. For example, communication means may include, but are not limited to, Bluetooth, BLE, Wi-Fi, Zigbee, 3G, LTE, and ultrasonic communication means.

Short-distance communication means may refer to communication means that enable communication with each other only when devices performing communication are within a predetermined range, and may include, for example, Bluetooth and NFC, but are not limited thereto.

A long-distance communication means may refer to a communication means that allows devices performing communication to communicate with each other regardless of distance. For example, a long-distance communication means is a means of communicating even when two devices performing communication through a repeater such as an AP are over a predetermined distance, and a means of communicating using a cellular network (3G, LTE) used for text transmission and phone calls. It may include, but is not limited to, communication means.

In the process of the server 4000 serving an online game using a network, communication between the server 4000 and the clients 2000a, 2000b, 2000c, 2000d, 2000e, and 2000f can be performed through any communication means. there is.

Online games include RPG (Role Playing Game), TRPG (Tabletop RPG), MMORPG (Massively Multiplayer Online RPG), MOBA (Multiplayer Online Battle Arena), AOS (Aeon of Strife), RTS (Real Time Strategy), and FPS (First-Time Strategy). It may be a game of various genres such as person shooter), TCG (Trading Card Game), CCG (Collectible Card Game), sports, fighting, etc., but is not limited thereto. Online games may be played between users or between users and a computer (eg, artificial intelligence), but are not limited to this and can be implemented in various forms.

The online game providing system shown in FIG. 1 can also function as an online platform providing system. Meanwhile, the server 4000 shown in FIG. 1 may correspond to the game providing device 10 described in FIGS. 2 to 14.

FIG. 2 is a flowchart illustrating a method of operating a game providing device that provides characteristic information of a plurality of groups classified according to payment possibilities of a plurality of users, according to an embodiment.

In step S210 of FIG. 2, the game providing device 10 may apply log data of a plurality of users for the first game to a payment probability model to obtain prediction information about the payment probability of a plurality of users. Log data of multiple users may include behavior information of multiple users.

Here, the payment probability model may be a model that predicts the possibility that a user will perform a payment action in the first game. For example, the payment probability model is based on the similarity between the behavior information of multiple users and the behavior information of users who performed payment actions during a predetermined period when log data of multiple users is input. It may be a model that predicts.

In step S220, the game providing device 10 may classify a plurality of users into a plurality of groups according to prediction information about the possibility of payment of the plurality of users. For example, the possibility of payment can be expressed as a probability. The game providing device 10 divides users with a payment probability of 0 or more and less than 0.3 into a first group, users with a payment probability of 0.3 or more but less than 0.6 as a second group, and users with a payment probability of 0.6 or more and 1 or less. It can be classified into the third group.

In step S230, the game providing device 10 may obtain characteristic information of the first group based on log data of a user belonging to the first group among the plurality of groups.

For example, the game providing device 10 compares the log data of the first group and the log data of the second group to determine whether the first group includes at least one characteristic that can distinguish the first group from the second group. Characteristic information can be obtained.

Specifically, the game providing device 10 compares the log data of the first group and the log data of the second group to determine the user's level of immersion in the first game for at least one category. Characteristic information of the first group can be obtained.

For another example, the game providing device 10 provides information on factors that influenced the payment behavior of the paying user based on the behavior information of the paying user and the behavior information of the non-paying user obtained from the actual payment data of the first group of users. It can be obtained. The process of obtaining information on factors that influenced the payment behavior of the payment user in the game providing device 10 is explained in FIGS. 10 and 11.

In step S240, the game providing device 10 may provide characteristic information of the first group. For example, the game providing device 10 may display the result of comparing the characteristic information of the first group with the characteristic information of the second group.

FIG. 3 is a diagram illustrating a process of predicting payment possibilities of a plurality of users in a game providing device and classifying the plurality of users into a plurality of groups according to the prediction results, according to an embodiment.

Referring to the block diagram 310 of FIG. 3, the game providing device 10 can acquire log data of a plurality of users. For example, the game providing device 10 may acquire log data of the first to tenth users who played the first game. The game providing device 10 may predict the payment probability of each user by applying log data of the first to tenth users to a payment probability model. For example, the game providing device 10 applies last month's log data of the first game for the first to tenth users to the payment probability model, and determines the current month's payment for the first game for the first to tenth users. Probabilities can be predicted.

Here, the payment probability model may be a model that predicts the payment probability of a given user based on the similarity between the behavior information of the given user obtained from log data and the behavior information of the user who performed the payment action.

For example, the similarity between the behavior information of the first user and the behavior information of the paying user who performed the payment action can be determined by comparing the play pattern of the first game of the first user and the play pattern of the first game of the paying user. there is. The play pattern can be determined from all of the user's behavior information within the first game.

Referring to the table 320 of FIG. 3, the game providing device 10 can obtain prediction information about the payment possibilities of the first to tenth users. Prediction information can be expressed as payment probability. The payment probability can be expressed as a number between 0 and 1, or as a number between 0% and 100%.

The game providing device 10 may classify the first to tenth users into first to third groups based on prediction information about payment possibility. For example, the game providing device 10 divides users with a probability of payment between 0% and less than 30% into the first group 331, which is a group with a low probability of payment, and users with a probability of payment between 30% and less than 60%. Users with a probability of payment between 60% and 100% can be classified into the second group 332, a group with a medium probability of payment, and the third group 333, a group with a high probability of payment.

Specifically, the game providing device 10 may classify the first user, third user, fifth user, seventh user, and ninth user into the first group 331. Additionally, the game providing device 10 may classify the sixth user, eighth user, and tenth user into the second group 332. Additionally, the game providing device 10 may classify the second user and the fourth user into the third group 333.

FIG. 4 is a diagram illustrating a method of displaying information of a plurality of groups in a game providing device, according to an embodiment.

Referring to the graph 410 of FIG. 4, the game providing device 10 may display numerical information of a plurality of groups classified according to payment probability in the first game. Specifically, the game providing device 10 may display the number of users belonging to each group.

Additionally, the game providing device 10 may display information on a plurality of groups by reflecting actual payment data of a plurality of users. For example, the game providing device 10 may display the number of paying users and the number of non-paying users for each plurality of groups.

Specifically, the game providing device 10 uses log data of a plurality of users of the first game for a predetermined period (e.g., the past 6 months, 12 months, etc.) to log data of the plurality of users in the first period (e.g., For example, the possibility of payment (next month) can be predicted. The game providing device 10 may classify a plurality of users into first to third groups according to payment probability indicating payment possibility. Then, when the first period arrives, the game providing device 10 may check the number of paying users and the number of non-paying users for each plurality of groups based on actual payment data. The game providing device 10 can check the accuracy of the payment probability model by checking the number of paying users relative to the total number of users for each plurality of groups. The game providing device 10 may update the payment probability model based on actual result data.

As shown in the graph 410 of FIG. 4, if the numerical information of a plurality of groups classified according to payment probability is expressed in terms of the number of users, the manager of the first game can more easily check the volatility for each group. For example, by comparing the previous month's payment probability graph with this month's payment probability graph, the administrator of the first game can easily check the volatility of each group and prepare measures to increase the payment probability of multiple users. You can.

The three groups shown in FIG. 4 are examples, and the game providing device 10 can classify a plurality of users into more than three groups and display information on the plurality of groups.

Referring to the graph 420 of FIG. 4, the game providing device 10 may display numerical information of a plurality of groups classified according to payment probability in the first game. Specifically, the game providing device 10 may display ratio information indicating the number of paying users relative to the number of users belonging to each group.

As shown in the graph 420 of FIG. 4, if the numerical information of a plurality of groups classified according to payment probability is expressed as payment user ratio information, the administrator of the first game can more easily check the payment conversion rate for each group. there is. For example, by comparing last month's payment probability graph with this month's payment probability graph, the manager of the first game can easily identify the group with an increased payment conversion rate and the group with a decreased payment conversion rate. The administrator of the first game may prepare measures to increase the payment conversion rate for groups with reduced payment conversion rates.

Figure 5 is a flowchart for explaining a method of operating a game providing device that generates a payment probability model, according to an embodiment.

Referring to step S510 of FIG. 5, the game providing device 10 may obtain log data of a paying user who performed a payment action in the first game during a predetermined period. Additionally, the game providing device 10 may obtain log data of non-paying users who have not performed payment actions in the first game for a predetermined period of time. The game providing device 10 may perform calculations through machine learning to learn behavior information of paying users from log data of paying users and learn behavior information of non-paying users from log data of non-paying users.

In step S520, the game providing device 10 may learn factors that influenced the payment behavior of the paying user based on the behavior information of the paying user and the behavior information of the non-paying user. The game providing device 10 may generate a payment probability model by using the results of learning factors that influenced the payment behavior of the payment user.

Additionally, the payment probability model may be a data recognition model used to determine the user's payment probability for the first game. The data recognition model may be a model obtained through deep learning built on an artificial neural network (ANN). For example, learning models such as Deep Neural Network (DNN), Recurrent Neural Network (RNN), and Bidirectional Recurrent Deep Neural Network (BRDNN) may be used as data recognition models, but are not limited thereto.

In step S530, the game providing device 10 may update the payment probability model for the first game based on actual payment data of a plurality of users and information on factors that affect the payment behavior of the paying user. For example, the game providing device 10 may relearn the payment probability model by weighting information on factors that affect the payment behavior of the payment user.

FIG. 6 is a diagram schematically showing an artificial neural network with a plurality of hierarchical structures used by a game providing device to learn a payment probability model, according to an embodiment.

Referring to FIG. 6, the artificial neural network may include an input layer 610, at least one hidden layer 620, 630, and an output layer 640. Additionally, calculations using an artificial neural network may be performed in a processor within the game providing device 10.

Additionally, weights between each layer and nodes can be learned through learning and training performed in the hidden layers 620 and 630. For example, the processor in the game providing device 10 learns the behavior information of the paying user and the behavior information of the non-paying user through repetitive learning, and produces the learned results of the behavior information of the paying user and the behavior information of the non-paying user. You can use this to learn the factors that influenced the payment behavior of payment users. The processor within the game providing device 10 may obtain weight values of factors that influenced the payment behavior of the payment user.

Additionally, the game providing device 10 may retrain the payment probability model from a trained artificial neural network so that the obtained weight value is reflected in the payment probability model.

FIG. 7 is a diagram illustrating a process in which a game providing device learns a payment probability model, according to an embodiment.

The game providing device 10 may acquire log data including behavior information of a plurality of users for the first game. For example, as shown in FIG. 7, the game providing device 10 includes the first user's log data 701, the second user's log data 702, and the third user's log data 703. It can be obtained. The game providing device 10 may acquire log data of users other than the first user, second user, and third user. For convenience of explanation, FIG. 7 illustrates the first user's log data 701, the second user's log data 702, and the third user's log data 703 as examples.

The game providing device 10 may perform calculations using an artificial neural network and learn each of the actions of a plurality of users based on log data of the plurality of users.

Referring to FIG. 7 , the game providing device 10 may learn the first user's behavior information from the first user's log data 701. Specifically, the game providing device 10 may extract information on actions performed by the first user in the first game from the first user's log data 701. For example, the game providing device 10 may extract information about the first action, second action, and third action of the first user. Based on the behavior information of the first user, the first user logs in to the first game, logs out, the duration of access to the first game, and the first game within a predetermined time interval. Number of accesses to the game, specific activity details of the first user within the first game, amount paid by the first user within the first game, items purchased by the first user within the first game, within the first game Each action regarding the cache used by the first user can be distinguished and learned.

Likewise, the game providing device 10 may learn the behavior information of the second user from the log data 702 of the second user and the behavior information of the third user from the log data 703 of the third user. .

In addition, the game providing device 10 may learn by distinguishing between the behavior information of a paying user who performed a payment behavior during a predetermined period and the behavior information of a non-paying user who did not perform a payment behavior among a plurality of users of the first game. The game providing device 10 may learn factors that influenced the payment behavior of the paying user based on the results of learning the behavior information of the paying user and the behavior information of the non-paying user.

For example, the game providing device 10 may learn the behavior of “the first user purchasing the first item to achieve the third round mission in the first game” from the first user’s log data 701. You can. The game providing device 10 may learn from the second user's log data 702 the behavior of “the second user achieving the third round mission using the first item in the first game.” The game providing device 10 detects the action of "a third user purchases the first item in the first game and completes the mission of the third round using the first item" from the third user's log data 703. You can learn. Referring to block 710 of FIG. 7, the game providing device 10 selects the “third round of the first game” as the user's payment behavior based on the results learned from the behavior information of the first to third users. You can make a decision based on the factors that influenced it and learn the determined results.

FIG. 8 is a diagram illustrating characteristic information of a plurality of groups provided by a game providing device, according to an embodiment.

Referring to the graph 810 of FIG. 8, the game providing device 10 may display numerical information of a plurality of groups classified according to payment probability in the first game. Specifically, the game providing device 10 may display the number of users belonging to each group. In the graph 810 of FIG. 8, parts that overlap with the description of the graph 410 of FIG. 4 are omitted.

The game providing device 10 may compare log data of at least two groups among a plurality of groups to obtain characteristic information including at least one characteristic that can distinguish each group. Specifically, the game providing device 10 compares log data of at least two groups to obtain group characteristic information for at least one category that serves as a standard for determining the user's immersion in the first game. can do. Here, the at least one category may be one of achievements for the first game, economic activity, used cache, growth, community, in-game, access, and out-game, and is not limited to the above example.

Specifically, the content of the “achievement” category for the first game may include user behavior information related to the first game. For example, the contents of the “achievement” category may include information such as experience points of the user's character in the first game and participation in events held in the first game. The contents of the “economic activity” category for the first game may include information such as various store activities of users and transactions between users in the first game. “Cash” category content for the first game may include information on cash spent by the user in the first game. The “growth” category content for the first game may include information such as the experience level of the character held by the user in the first game and the user’s level. The “community” category content for the first game may include information such as chatting and making plans between users within the first game. The “in-game” category content for the first game may include event information occurring within the first game. The “access” category content for the first game may include information such as the number of times, time, and duration of access to the first game. The “out game” category content for the first game may include event information that occurs outside the first game.

For example, in order to identify the characteristics of the “third group”, which is a group with a high probability of payment, the game providing device 10 combines log data of the “first group”, which is a group with a low probability of payment, and a group with a high probability of payment. The log data of the “third group” can be compared. Based on the comparison result, the game providing device 10 may obtain characteristic information of the third group that can distinguish the third group from the first group.

For a specific example, the game providing device 10 compares the log data of the third group and the log data of the first group to determine the "growth" category as a standard for determining the user's level of immersion in the first game. Characteristic information of the third group can be obtained. The graph 820 in FIG. 8 is a graph showing the user ratio according to the engagement index for the "growth" category by the "third group", a group with a high payment probability, and the "first group", a group with a low payment probability. am. The engagement index may represent an index indicating the user's level of immersion in a certain game. The higher the engagement index, the higher the user's immersion in a certain game.

Referring to the graph 820 in FIG. 8, the “third group”, which is a group with a high probability of payment, tends to have a high user ratio for growth the higher the engagement index, and the lower the engagement index, the higher the proportion of users for growth. There is a tendency for the user ratio to be low. On the other hand, the "first group", which is the group with a low probability of payment, tends to have a low user ratio to growth as the engagement index is high, and has a tendency to have a high user ratio to growth as the engagement index is low. . The game providing device 10 may determine that a group with a higher probability of payment for the first game has a higher propensity for activities related to the growth of the first game. The manager of the first game may induce the development of the first game to further increase the propensity for activities related to the growth of the first game.

FIG. 9 is a diagram illustrating characteristic information of a plurality of groups provided by a game providing device, according to another embodiment.

The game providing device 10 may compare log data of at least two groups to obtain group characteristic information for at least one category, which is a standard for determining the user's level of immersion in the first game. . Additionally, the game providing device 10 may compare log data of at least two groups to obtain group characteristic information for subcategories of a predetermined category.

Referring to the graph 900 of FIG. 9, the game providing device 10 provides log data of a “third group” with a high payment probability (e.g., a payment probability of 60% or more), and log data of a “third group” with a medium payment probability. Based on the log data of the “second group” (e.g., payment probability of 30% or more and less than 60%), the third group or characteristic information of the second group for the “economic activity” category may be obtained.

For example, the game providing device 10 may include an “item upgrade” category, an “item launch” category, an “item acquisition” category, an “item creation” category, and a “game money usage” category that are subcategories of the “economic activity” category. , For the “game money acquisition” category, characteristic information between the third group and the second group can be obtained and displayed. Referring to the graph 900 of FIG. 9, the game providing device 10 shows that users in the “third group” with a high payment probability perform activities related to “item creation” more than users in the “second group” with a medium payment probability. It can be judged that the tendency is high. Additionally, the game providing device 10 may determine that users in the “third group” with a high payment probability and users in the “second group” with a medium payment probability have similar activity tendencies related to “item acquisition.”

FIG. 10 is a flowchart illustrating a method of operating a game providing device that obtains information on factors that influenced the payment behavior of a paying user, according to an embodiment.

In step S1010 of FIG. 10, the game providing device 10 may obtain behavior information of paying users and behavior information of non-paying users from payment data of users of the first group.

In step S1020, the game providing device 10 performs an operation through machine learning to generate a first payment behavior prediction model considering a first parameter among a plurality of parameters, and makes a second payment without considering the first parameter. A behavior prediction model can be created. For example, the game providing device 10 may generate a payment behavior prediction model using a random forest algorithm or an artificial neural network.

In step S1030, the game providing device 10 acquires information on factors that influenced the payment behavior of the paying user based on the result of comparing the performance of the first payment behavior prediction model and the performance of the second payment behavior prediction model. You can.

For example, if the performance difference between the first payment behavior prediction model and the second payment behavior prediction model exceeds a preset range, the game providing device 10 records the user's behavior information related to the first parameter to determine the payment behavior of the user. It can be determined based on information on factors that influenced.

For another example, if the performance difference between the first payment behavior prediction model and the second payment behavior prediction model is within a preset range, the game providing device 10 may transmit the user behavior information related to the first parameter to the payment behavior of the user. It can be decided based on element information that has little influence.

FIG. 11 is a diagram illustrating a process of obtaining information on factors that influenced the payment behavior of a payment user in a game providing device, according to an embodiment.

The game providing device 10 compares the performance of a payment behavior prediction model generated considering a predetermined parameter with the performance of a payment behavior prediction model generated without considering the predetermined parameter, and based on the comparison result, determines the user related to the predetermined parameter. It is possible to determine whether the behavior information is element information that influenced the payment behavior of the payment user.

For example, the game providing device 10 may learn log data of multiple users based on machine learning. Based on the learning result, the game providing device 10 may generate a first payment behavior prediction model by considering the first parameter. Likewise, the game providing device 10 may generate a second payment behavior prediction model based on the learning result without considering the first parameter. The game providing device 10 may compare actual payment data of a plurality of users with predicted data calculated using a payment behavior prediction model.

As shown in table 1100 of FIG. 11, the predicted data calculated from the first payment behavior prediction model has an error rate of 2% from the actual payment data, and the predicted data calculated from the second payment behavior prediction model has an error rate of 2% compared to the actual payment data. If there is an error rate of 15% with the data, the game providing device 10 may determine user behavior information related to the first parameter as information on factors that influenced the payment behavior of the paying user.

For another example, the game providing device 10 may generate a third payment behavior prediction model in which the second parameter is considered and a fourth payment behavior prediction model in which the second parameter is not considered based on the learning result. The game providing device 10 may compare actual payment data of a plurality of users with predicted data calculated using a payment behavior prediction model.

As shown in table 1100 of FIG. 11, the predicted data calculated from the third payment behavior prediction model has an error rate of -1% from the actual payment data, and the predicted data calculated from the fourth payment behavior prediction model has an error rate of -1% from the actual payment data. If there is an error rate of -2% with the payment data, the game providing device 10 determines that the difference between the performance of the third payment behavior prediction model and the fourth payment behavior prediction model is not large, and the user related to the second parameter Behavior information can be determined as element information that has little effect on the payment behavior of the payment user.

If the difference between the performance of the payment behavior prediction model generated considering the predetermined parameter and the performance of the payment behavior prediction model generated without considering the predetermined parameter exceeds a preset range, the game providing device 10 may control the user related to the predetermined parameter. Behavior information can be determined as information on factors that influenced the payment behavior of the payment user. Additionally, the game providing device 10 may assign weights to predetermined parameters based on the extent to which the performance difference between payment behavior prediction models exceeds a preset range. The game providing device 10 may learn a payment probability model that predicts the possibility of payment for the first game by considering the weight.

The error rate and importance for each parameter described in the table 1100 of FIG. 11 are examples, and other values may be described.

FIG. 12 is a flowchart illustrating a method of operating a game providing device that provides purchasing power information and payment conversion rate information of a plurality of users, according to an embodiment.

In step S1210 of FIG. 12 , the game providing device 10 may obtain purchasing power information indicating the expected payment amount of a plurality of users based on the payment data of the plurality of users for the first game for a predetermined period. In this case, the game providing device 10 may predict purchasing power information of a plurality of users using a linear regression model, a random forest algorithm, or an artificial neural network. For example, the purchasing power information of the first user may include information on the expected payment amount for this month determined based on payment data of the first user for a predetermined period (e.g., last month, 6 months, or 12 months, etc.). You can.

In step S1220, the game providing device 10 may classify a plurality of users into a plurality of groups based on purchasing power information.

In step S1230, the game providing device 10 may obtain payment conversion rate information indicating the actual payment amount compared to the expected payment amount for each of the plurality of groups.

In step S1240, the game providing device 10 may provide payment conversion rate information. For example, the game providing device 10 may receive an input for selecting a first section of the payment conversion rate from a first group among a plurality of groups. The game providing device 10 may display purchasing power information and actual payment amount information of users belonging to the first section.

FIG. 13 is a diagram illustrating a process of providing payment conversion rate information in a game providing device, according to an embodiment.

The game providing device 10 may obtain payment conversion rate information for a plurality of groups classified according to purchasing power information.

Referring to the graph 1310 of FIG. 13, the game providing device 10 may classify a plurality of users into six groups according to the purchasing power information of the plurality of users. For example, based on purchasing power information of a plurality of users, the game providing device 10 divides the plurality of users into a first group with an expected payment amount of less than 5,000 won, a second group with an expected payment amount of more than 5,000 won and less than 20,000 won, 3rd group with an expected payment amount of 20,000 to 50,000 won, 4th group with an expected payment amount of 50,000 to 100,000 won, 5th group with an expected payment amount of 100,000 to 200,000 won, and 2nd group with an expected payment amount of 200,000 won or more. It can be classified into 6 groups.

In this case, the game providing device 10 acquires purchasing power information indicating the expected payment amount of multiple users based on payment data of multiple users over a predetermined period (e.g., the past 6 months, 12 months, etc.). can do.

The payment data may be at least one of the total payment amount paid by a plurality of users over a predetermined period of time, the average amount of the payment amount over a certain period of time, and the time of payment.

For example, if the first user pays 30,000 won per month, the game providing device 10 may determine the purchasing power of the first user to be 30,000 won and classify the first user into the third group.

For another example, if the second user pays an average of 150,000 won every May 5th of each year, the game providing device 10 determines the second user's purchasing power in May to be 150,000 won and classifies it into the fifth group. You can.

As shown in the graph 1310 of FIG. 13, the game providing device 10 acquires payment conversion rate information indicating the actual payment amount compared to the expected payment amount for each plurality of groups based on the current point in time, and provides payment conversion rate information. It can be displayed.

The game providing device 10 may receive an input for selecting a section in which the current payment conversion rate is greater than 0% and less than 10% among the third group. The game providing device 10 may extract users belonging to a range in which the current payment conversion rate is between 0% and less than 10% among the third group, and display purchasing power information and payment amount information of the extracted users.

Figure 14 is a block diagram showing the configuration of a game providing device, according to one embodiment.

The game providing device 10 shown in FIG. 14 may include a communication device 1410, a user interface device 1420, a storage 1430, a memory 1440, and a processor 1450. However, not all of the illustrated components are essential components. The game providing device 10 may be implemented with more components than the illustrated components, and the game providing device 10 may also be implemented with fewer components. The game providing device 10 shown in FIG. 14 may correspond to the server 4000 described in FIG. 1.

The communication device 1410 can communicate with an external device. Specifically, the communication device 1410 may be connected to a network either wired or wirelessly to communicate with an external device. Here, the external device may be a server, smartphone, tablet, PC, etc. The communication device 1410 may include a communication module that supports one of various wired and wireless communication methods. For example, the communication module may be in the form of a chipset, or may be a sticker/barcode (e.g. a sticker including an NFC tag) containing information necessary for communication. Additionally, the communication module may be a short-distance communication module or a wired communication module.

For example, the communication device 1410 may use wireless LAN, Wi-Fi (Wireless Fidelity), WFD (Wi-Fi Direct), Bluetooth, BLE (Bluetooth Low Energy), Wired Lan, and NFC ( It can support at least one of (Near Field Communication), Zigbee (IrDA, infrared Data Association), 3G, 4G, and 5G.

The user interface device 1420 may refer to a device that receives data from a user to control the game providing device 10. The processor 1450 may control the user interface device 1420 to generate and output a user interface screen for receiving predetermined commands or data from the user. The user interface device 1420 includes an input unit for receiving inputs that control the operation of the game providing device 10 and information such as the results of the operation of the game providing device 10 or the status of the game providing device 10. It may include an output unit for display. For example, the user interface device 1420 may include a manipulation panel that receives user input, a display panel that displays a screen, etc.

Specifically, the input unit may include devices that can receive various types of user input, such as a keyboard, physical button, touch screen, camera, or microphone. Additionally, the output unit may include, for example, a display panel or a speaker. However, the user interface device 1420 is not limited to this and may include a device that supports various inputs and outputs.

The user interface device 1420 may display prediction information about the possibility of payment of a plurality of users for the first game. Additionally, the user interface device 1420 may display the results of classifying a plurality of users into a plurality of groups according to prediction information. Additionally, the user interface device 1420 may display characteristic information of the first group. For example, the user interface device 1420 may display the result of comparing characteristic information of the first group with characteristic information of the second group.

The user interface device 1420 may display purchasing power information indicating the expected payment amount of a plurality of users for the first game. Additionally, the user interface device 1420 may display the results of classifying a plurality of users into a plurality of groups according to purchasing power information. Additionally, the user interface device 1420 may display payment conversion rate information indicating the actual payment amount compared to the expected payment amount for each of the plurality of groups, based on the current time.

The storage 1430 is related to programs, applications, etc. used by the game providing device 10 to predict payment possibilities of multiple users, classify multiple users into multiple groups, and provide characteristic information for the multiple groups. Information can be saved.

In addition, the storage 1430 allows the game providing device 10 to obtain purchasing power information indicating the expected payment amount from the payment data of a plurality of users, classify the plurality of users into a plurality of groups according to the purchasing power information, and classify the plurality of users into a plurality of groups. Information related to programs and applications used to provide payment conversion rate information can be stored.

The storage 1430 can store programs running on the game providing device 10, applications, and various data used on the platform.

The storage 1430 may store the results of learning each of the actions of a plurality of users based on the log data of the plurality of users in the game providing device 10, the results of re-learning the model using the learned results, etc. .

Memory 1440 may store software or programs. The memory 1440 is at least one device for executing a method of providing characteristic information of groups classified according to payment possibilities of a plurality of users in the game providing device 10, or a method of providing payment conversion rate information of a plurality of users. You can save the program. At least one program stored in the memory 1440 may be classified into a plurality of modules according to function.

The processor 1450 controls the overall operation of the game providing device 10 and may include at least one processor such as a CPU. The processor 1450 may include at least one specialized processor corresponding to each function, or may be an integrated processor.

The processor 1450 may execute a program stored in the memory 1440, read data or files stored in the memory 1440, or store a new file in the memory 1440. The processor 1450 may execute instructions stored in the memory 1440.

The processor 1450 may apply log data of a plurality of users for the first game to a payment probability model to obtain prediction information about the payment probability of a plurality of users. Log data of multiple users may include behavior information of multiple users.

Here, the payment probability model may be a model that predicts the possibility that a user will perform a payment action in the first game. For example, the payment probability model is based on the similarity between the behavior information of multiple users and the behavior information of users who performed payment actions during a predetermined period when log data of multiple users is input. It may be a model that predicts.

The processor 1450 may classify a plurality of users into a plurality of groups according to prediction information about the possibility of payment of the plurality of users. For example, the possibility of payment can be expressed as a probability. The processor 1450 groups users with a probability of payment between 0 and less than 0.3 into a first group, users with a probability of payment between 0.3 and less than 0.6 into a second group, and users with a probability of payment between 0.6 and 1 or less into a third group. Can be classified into groups.

The processor 1450 may obtain characteristic information of the first group based on log data of a user belonging to the first group among the plurality of groups.

For example, the processor 1450 compares the log data of the first group and the log data of the second group to generate first group characteristic information including at least one characteristic that can distinguish the first group from the second group. can be obtained.

Specifically, the processor 1450 compares the log data of the first group and the log data of the second group to determine the user's level of immersion in the first game. You can obtain group characteristic information.

For another example, the processor 1450 may obtain information on factors that influenced the payment behavior of the paying user based on the behavior information of the paying user and the behavior information of the non-paying user obtained from the actual payment data of the first group of users. You can.

Specifically, the processor 1450 may obtain behavior information of paying users and behavior information of non-paying users from payment data of users of the first group.

The processor 1450 performs an operation through machine learning to generate a first payment behavior prediction model that considers the first parameter among a plurality of parameters, and creates a second payment behavior prediction model that does not consider the first parameter. You can.

The processor 1450 may obtain information on factors that influenced the payment behavior of the paying user based on the result of comparing the performance of the first payment behavior prediction model and the performance of the second payment behavior prediction model.

For example, if the performance difference between the first payment behavior prediction model and the second payment behavior prediction model exceeds a preset range, the processor 1450 uses the user's behavior information related to the first parameter to influence the payment behavior of the paying user. can be determined with crazy element information.

For another example, if the performance difference between the first payment behavior prediction model and the second payment behavior prediction model is within a preset range, the processor 1450 may apply the user behavior information related to the first parameter to the payment behavior of the paying user. It can be decided based on information on elements that have no influence.

The processor 1450 may provide characteristic information of the first group. For example, the processor 1450 may display the result of comparing the characteristic information of the first group with the characteristic information of the second group through the user interface device 1420. Additionally, the processor 1450 may transmit the result of comparing the characteristic information of the first group with the characteristic information of the second group to an external device through the communication device 1410.

Meanwhile, the processor 1450 may obtain log data of a paying user who performed a payment action in the first game during a predetermined period. Additionally, the processor 1450 may obtain log data of non-paying users who have not performed a payment action in the first game for a predetermined period of time. The processor 1450 may perform operations through machine learning to learn behavior information of paying users from log data of paying users and learn behavior information of non-paying users from log data of non-paying users.

The processor 1450 may learn factors that influenced the payment behavior of the paying user based on the behavior information of the paying user and the behavior information of the non-paying user. The processor 1450 may create a payment probability model using the results of learning factors that influenced the payment behavior of the payment user.

Additionally, the payment probability model may be a data recognition model used to determine the user's payment probability for the first game. The data recognition model may be a model obtained through deep learning built on an artificial neural network (ANN). For example, learning models such as Deep Neural Network (DNN), Recurrent Neural Network (RNN), and Bidirectional Recurrent Deep Neural Network (BRDNN) may be used as data recognition models, but are not limited thereto.

The processor 1450 may update the payment probability model for the first game based on actual payment data of a plurality of users and information on factors that affect the payment behavior of the paying user.

Meanwhile, the processor 1450 may obtain purchasing power information indicating the expected payment amount of the plurality of users based on payment data of the plurality of users for the first game for a predetermined period of time. Here, the payment data may be at least one of the total payment amount paid by a plurality of users over a certain period of time, the average amount of payment over a certain period of time, and the time of payment.

The processor 1450 may obtain payment conversion rate information indicating the actual payment amount compared to the expected payment amount for each of the plurality of groups as of the current time. The processor 1450 may display payment conversion rate information through the user interface device 1420.

For example, the user interface device 1420 may receive an input for selecting a first section of the payment conversion rate from a first group among a plurality of groups. The user interface device 1420 may display purchasing power information and actual payment amount information of a user belonging to the first section.

Additionally, the processor 1450 may transmit payment conversion rate information to an external device through the communication device 1410.

The game providing device 10 described above may be implemented with hardware components, software components, and/or a combination of hardware components and software components. For example, devices and components described in embodiments may include, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable gate array (FPGA), etc. , may be implemented using one or more general-purpose or special-purpose computers, such as a programmable logic unit (PLU), a microprocessor, or any other device capable of executing and responding to instructions. A processing device may execute an operating system (OS) and one or more software applications that run on the operating system. Additionally, a processing device may access, store, manipulate, process, and generate data in response to the execution of software. For ease of understanding, a single processing device may be described as being used; however, those skilled in the art will understand that a processing device includes multiple processing elements and/or multiple types of processing elements. It can be seen that it may include. For example, a processing device may include a plurality of processors or one processor and one controller. Additionally, other processing configurations, such as parallel processors, are possible.

Software may include a computer program, code, instructions, or a combination of one or more of these, which may configure a processing unit to operate as desired, or may be processed independently or collectively. You can command the device. Software and/or data may be used on any type of machine, component, physical device, virtual equipment, computer storage medium or device to be interpreted by or to provide instructions or data to a processing device. , or may be permanently or temporarily embodied in a transmitted signal wave. Software may be distributed over networked computer systems and thus stored or executed in a distributed manner. Software and data may be stored on one or more computer-readable recording media.

The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded on a computer-readable recording medium. A computer-readable recording medium may include program instructions, data files, data structures, etc., singly or in combination. Program instructions recorded on the medium may be specially designed and configured for the embodiment or may be known and available to those skilled in the art of computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic media such as floptical disks. -Includes optical media (magneto-optical media) and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, flash memory, etc. Examples of program instructions include machine language code, such as that produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter, etc. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

As described above, although the embodiments have been described with limited examples and drawings, various modifications and variations can be made by those skilled in the art from the above description. For example, the described techniques are performed in a different order than the described method, and/or components of the described system, structure, device, circuit, etc. are combined or combined in a different form than the described method, or other components are used. Alternatively, appropriate results may be achieved even if substituted or substituted by an equivalent.

Therefore, other implementations, other embodiments and equivalents of the claims also fall within the scope of the following claims.

Claims (27)

By applying log data containing behavior information of a plurality of users for the first game to a payment probability model that predicts the possibility of a user performing a payment action in the first game, prediction of the possibility of payment of the plurality of users is made. Obtaining information;
Classifying the plurality of users into a plurality of groups according to prediction information about the possibility of payment of the plurality of users;
Obtaining characteristic information of the first group based on log data of a user belonging to the first group among the plurality of groups; and
A method of operating a game providing device, including providing characteristic information of the first group. According to paragraph 1,
The payment probability model is,
When log data of the plurality of users is input, a model that predicts the possibility of payment of the plurality of users based on the similarity between the behavior information of the plurality of users and the behavior information of users who performed payment actions during a predetermined period, How the game delivery device operates. According to paragraph 1,
Generating the payment probability model based on log data of a paying user who performed a payment action in the first game during a predetermined period and log data of a non-paying user who did not perform a payment action in the first game during the predetermined period. A method of operating a game providing device, further comprising steps. According to paragraph 3,
The step of generating the payment probability model is,
Performing an operation through machine learning to learn behavior information of the paying user from log data of the paying user and learning behavior information of the non-paying user from log data of the non-paying user; and
A method of operating a game providing device, comprising the step of learning factors that influenced the payment behavior of the paying user based on the behavior information of the paying user and the behavior information of the non-paying user. According to paragraph 1,
The step of obtaining characteristic information of the first group includes:
Characteristic information of the first group including at least one characteristic that can distinguish the first group from the second group by comparing the log data of the first group with the log data of the second group among the plurality of groups A method of operating a game providing device, including the step of acquiring. According to clause 5,
The step of obtaining characteristic information of the first group includes:
Characteristic information of the first group for at least one category that is a standard for determining the user's level of immersion in the first game by comparing the log data of the first group and the log data of the second group A method of operating a game providing device, including the step of acquiring. According to paragraph 1,
The step of obtaining characteristic information of the first group includes:
Providing a game comprising the step of obtaining information on factors that influenced the payment behavior of the paying user based on the behavior information of the paying user and the behavior information of the non-paying user obtained from the actual payment data of the users of the first group. How the device works. In clause 7,
The step of obtaining information on factors that influenced the payment behavior of the payment user is,
By performing calculations through machine learning, the performance of the first payment behavior prediction model generated considering the first parameter among the plurality of parameters and the second payment behavior generated without considering the first parameter among the plurality of parameters Comparing the performance of behavior prediction models; and
According to the comparison result, if the performance difference between the first payment behavior prediction model and the second payment behavior prediction model exceeds a preset range, the user's behavior information related to the first parameter is used to influence the payment behavior of the payment user. A method of operating a game providing device, comprising determining with crazy factor information. In clause 7,
A method of operating a game providing device, further comprising updating the payment probability model for the first game based on the actual payment data of the plurality of users and information on factors that influenced the payment behavior of the paying users. . According to paragraph 1,
The step of providing characteristic information of the first group includes:
A method of operating a game providing device, comprising displaying a result of comparing characteristic information of the first group with characteristic information of a second group among the plurality of groups. Obtaining purchasing power information indicating an expected payment amount of the plurality of users based on payment data of the plurality of users for a first game for a predetermined period of time;
classifying the plurality of users into a plurality of groups according to the purchasing power information;
Obtaining payment conversion rate information representing the actual payment amount compared to the expected payment amount for each of the plurality of groups; and
A method of operating a game providing device, including providing the payment conversion rate information. According to clause 11,
The step of providing the payment conversion rate information is,
Receiving an input for selecting a first section of the payment conversion rate from a first group among the plurality of groups; and
A method of operating a game providing device, comprising displaying purchasing power information and actual payment amount information of a user belonging to the first section. communication device;
user interface device;
storage;
processor; and
Includes a memory that stores instructions executable by the processor,
The processor, by executing the instructions,
By applying log data containing behavior information of a plurality of users for the first game to a payment probability model that predicts the possibility of a user performing a payment action in the first game, prediction of the possibility of payment of the plurality of users is made. obtain information,
Classifying the plurality of users into a plurality of groups according to prediction information about the possibility of payment of the plurality of users,
Based on log data of a user belonging to a first group among the plurality of groups, obtain characteristic information of the first group,
A game providing device that provides characteristic information of the first group through the user interface device. According to clause 13,
The payment probability model is,
When log data of the plurality of users is input, a model that predicts the possibility of payment of the plurality of users based on the similarity between the behavior information of the plurality of users and the behavior information of users who performed payment actions during a predetermined period, Game delivery device. According to clause 13,
The processor, by executing the instructions,
Generating the payment probability model based on log data of a paying user who performed a payment action in the first game during a predetermined period and log data of a non-paying user who did not perform a payment action in the first game during the predetermined period. , game delivery device. According to clause 15,
The processor, by executing the instructions,
By performing calculations through machine learning, behavior information of the paying user is learned from the log data of the paying user, and behavior information of the non-paying user is learned from the log data of the non-paying user,
A game providing device that learns factors that influenced the payment behavior of the paying user based on the behavior information of the paying user and the behavior information of the non-paying user. According to clause 13,
The processor, by executing the instructions,
Characteristic information of the first group including at least one characteristic that can distinguish the first group from the second group by comparing the log data of the first group with the log data of the second group among the plurality of groups A game providing device that acquires a. According to clause 17,
The processor, by executing the instructions,
Characteristic information of the first group for at least one category that is a standard for determining the user's level of immersion in the first game by comparing the log data of the first group and the log data of the second group A game providing device that acquires a. According to clause 13,
The processor, by executing the instructions,
A game providing device that acquires information on factors that influenced the payment behavior of the paying user based on the behavior information of the paying user and the behavior information of the non-paying user obtained from the actual payment data of the users of the first group. According to clause 19,
The processor, by executing the instructions,
By performing calculations through machine learning, the performance of the first payment behavior prediction model generated by including the first parameter among the plurality of parameters and the second payment behavior generated by excluding the first parameter among the plurality of parameters Compare the performance of prediction models,
According to the comparison result, if the performance difference between the first payment behavior prediction model and the second payment behavior prediction model exceeds a preset range, the user's behavior information related to the first parameter is used to influence the payment behavior of the payment user. A game providing device that decides with crazy factor information. According to clause 19,
The processor, by executing the instructions,
A game providing device that updates the payment probability model for the first game based on the actual payment data of the plurality of users and information on factors that influenced the payment behavior of the paying users. According to clause 13,
The processor, by executing the instructions,
A game providing device that displays a result of comparing characteristic information of the first group with characteristic information of a second group among the plurality of groups through the user interface device. communication device;
user interface device;
storage;
processor; and
Includes a memory that stores instructions executable by the processor,
The processor, by executing the instructions,
Based on payment data of a plurality of users for a first game for a predetermined period, purchasing power information indicating the expected payment amount of the plurality of users is acquired,
Classifying the plurality of users into a plurality of groups according to the purchasing power information,
For each of the plurality of groups, obtain payment conversion rate information indicating the actual payment amount compared to the expected payment amount,
A game providing device that provides information on the payment conversion rate through the user interface device. By applying log data containing behavior information of a plurality of users for the first game to a payment probability model that predicts the possibility of a user performing a payment action in the first game, prediction of the possibility of payment of the plurality of users is made. Obtaining information;
Classifying the plurality of users into a plurality of groups according to prediction information about the possibility of payment of the plurality of users;
Obtaining characteristic information of the first group based on log data of a user belonging to the first group among the plurality of groups; and
A computer program stored on a medium for executing a method in conjunction with hardware, comprising providing the first group of characteristic information. Obtaining purchasing power information indicating an expected payment amount of the plurality of users based on payment data of the plurality of users for a first game for a predetermined period of time;
classifying the plurality of users into a plurality of groups according to the purchasing power information;
Obtaining payment conversion rate information representing the actual payment amount compared to the expected payment amount for each of the plurality of groups; and
A computer program stored on a medium for executing a method in conjunction with hardware, comprising providing the payment conversion rate information. By applying log data containing behavior information of a plurality of users for the first game to a payment probability model that predicts the possibility of a user performing a payment action in the first game, prediction of the possibility of payment of the plurality of users is made. Obtaining information;
Classifying the plurality of users into a plurality of groups according to prediction information about the possibility of payment of the plurality of users;
Obtaining characteristic information of the first group based on log data of a user belonging to the first group among the plurality of groups; and
A computer-readable recording medium recording a program for executing on a computer a method including providing characteristic information of the first group. Obtaining purchasing power information indicating an expected payment amount of the plurality of users based on payment data of the plurality of users for a first game for a predetermined period of time;
classifying the plurality of users into a plurality of groups according to the purchasing power information;
Obtaining payment conversion rate information representing the actual payment amount compared to the expected payment amount for each of the plurality of groups; and
A computer-readable recording medium recording a program for executing a method on a computer, including providing the payment conversion rate information.

Images