WO2018080272A1

WO2018080272A1 - Method, server, system and computer program for providing real-time personalised information to game user

Info

Publication number: WO2018080272A1
Application number: PCT/KR2017/012113
Authority: WO
Inventors: 방준혁; 박우석; 이경준; 석영민; 전선영
Original assignee: 넷마블 주식회사
Priority date: 2016-10-31
Filing date: 2017-10-31
Publication date: 2018-05-03

Abstract

The present invention relates to a method for providing real-time personalised information to a game user, and more specifically to a personalised information providing server, that is distinguished from a game server, and that provides personalised information to a game user's terminal having a game application installed therein. The method comprises the steps of: generating state information of a game user from game log information; generating a plurality of user codes by accumulating the game log information for a certain time period and batch processing same, and by combining the user codes, generating user pattern codes with respect to the game user; matching the game user and real-time campaign information customized with respect to the game user, by using the game user state information and user pattern code; and transmitting the matched customized real-time campaign information to a terminal of the game user. The present invention also relates to a server, a system and a computer program for realising the previously described method for providing personalised information.

Description

Methods, servers, systems and computer programs that provide game users with personalized information in real time

The present invention relates to a method, a server, a system, and a computer program for providing real-time personalized information to a game user. Specifically, the game pattern is utilized to analyze a user pattern and to optimize a user pattern in response to a game user's state. It relates to a configuration that can provide real-time information.

Many people play various games using mobile devices such as smartphones or tablet PCs.

As the mobile game market grows in size and competition among game companies is intensifying, measures are being taken to prevent game users from escaping and increase interest and loyalty to the game.

Game companies and publishing companies provide guidebooks or game guides on released games through online bulletin boards or game applications to encourage game users to understand and play the game easily.

In addition, in order to increase the game user's interest and increase the sustainability of the game, the game users are introduced to the gameplay tips periodically or non-periodically, and various promotions such as item specials and recommender points are earned.

However, such game-related information and promotions are provided to all game users in common, and thus are insufficient to satisfy game users having various propensities and game patterns.

As such, conventional game-related information and promotions commonly applied to game users become useless because many game users do not show interest except for some game users, and thus have a great effect on increasing the interest and survival rate of game users. There is a problem that can not be harvested.

In addition, unlike online games (PC games), mobile games do not have a procedure for entering personal information such as a social security number, so there is no game ID, and game users are identified by code generated when downloading and installing a game application. have.

As game users are identified only by codes of game applications, it is difficult to accurately identify game users in mobile games. In such a situation, it is more difficult to provide game-related information or promotions that meet the needs of mobile game users.

The present invention was devised to solve the above problems, and an object of the present invention is to increase the interest and sustainability of the game user by satisfying the game user having various tendencies and game patterns.

Another object of the present invention is to accurately identify game users who can play through various routes in a mobile game, and to provide specialized services (game guides, promotions, marketing) for a particular game user.

To this end, according to the personalization information providing server according to an embodiment of the present invention, a real-time event processing unit for generating a game user status information from the game log information, and accumulates and batches the game log information for a certain period of time; A user pattern analysis unit which generates a plurality of user codes and combines them to generate a user pattern code for a game user; and the game user and the game user using the state information of the game user and the user pattern code. A campaign providing unit for matching real-time customized campaign information, and a communication interface for receiving the game log information and transmitting the matched real-time customized campaign information to the terminal of the game user, and distinguished from a game server. It is done.

According to an embodiment, the communication interface may receive the game log information from at least one of a game user terminal having a game application installed therein or a game server synchronized with the game application.

According to an embodiment, the communication interface may communicate the game log information and the campaign information with an integrated service module included in a game application in the game user terminal.

According to an embodiment of the present disclosure, the personalization providing server further includes an LTV analyzer configured to calculate a life time value (LTV) of a game user using the game log information, wherein the LTV generates one of the plurality of user codes. It is characterized by being the basis. In addition, the LTV may be used to evaluate the adequacy of the campaign and determine whether the campaign is continued.

According to an embodiment, the LTV analyzer may calculate an LTV based on a survival probability and a purchase probability of a game user in a game application.

According to an embodiment, the LTV analyzer may calculate the LTV by further considering an average purchase amount, a survival probability attenuation rate, and a purchase probability attenuation rate of a unit user's unit period.

According to one embodiment, the response information of the game user to the campaign received from the game user terminal is characterized in that reflected in the survival probability or the purchase probability.

According to an embodiment of the present disclosure, the campaign provider determines a campaign constraint to determine whether to expose campaign information to the game user.

According to an embodiment of the present disclosure, the campaign constraint checking may include checking at least one of a campaign active period, repetition setting, frequency setting, limit setting, and real-time exposure setting.

According to an embodiment of the present disclosure, the campaign provider may determine whether to provide a campaign, whether to continue, change or modify the campaign by grasping the response information of the game user to the campaign received from the terminal of the game user in real time. .

According to an embodiment of the present disclosure, the personalization providing server further includes a natural inference unit that determines an actual game user, and the natural inference unit may play information on one or more games on a plurality of terminals or a plurality of games on one terminal. Infer whether the information is actually game play information by one game user, and the user pattern analyzer generates a user pattern code for the deduced one game user.

According to an embodiment, the natural inference unit extracts a plurality of kit values representing a game user from the game log information, and determines an actual game user based on the plurality of kit values.

According to an embodiment, the natural inference unit uses an install kit value generated when installing a game application as a plurality of kit values and a device kit value generated in a device on which the game application is installed. It features.

According to an embodiment, the natural inference unit may further use a channel interlocking kit value (Channel ID) or an advertising kit value (Advertising ID).

According to one embodiment, the game is characterized in that the game can be started without login.

According to one embodiment, the game is characterized in that the mobile game.

According to an embodiment, the personalization information providing server may be operated by a company serving a plurality of types of games, and the communication interface may receive game log information of a plurality of types of games.

According to an embodiment, the integrated service module may be embedded in a software development kit (SDK) of a game application provided to a game developer by a company serving a plurality of games.

The personalization information providing system according to an embodiment includes the above-described personalization information providing server and a user terminal which accesses a game server and transmits game log information in real time through an integrated service module included in a game application while playing a game. Characterized in that.

Method for providing personalized information according to an embodiment is characterized in that it comprises each step of operating the above-described personalized information providing server. Specifically, the personalization information providing method according to an embodiment is a method in which a personalization information providing server which is distinguished from a game server provides personalization information to a game user terminal in which a game application is installed, and generates game user state information from game log information. And generating a plurality of user codes by accumulating and batching the game log information for a predetermined period of time, and combining them to generate a user pattern code for a game user, the state information of the game user and Matching the real-time customized campaign information for the game user and the game user using the user pattern code, and transmitting the matched real-time customized campaign information to the terminal of the game user.

According to an embodiment of the present disclosure, the method further includes receiving the game log information from at least one of the game user terminal in which a game application is installed, or a game server synchronized with the game application.

According to an embodiment of the present disclosure, the personalization information providing server may communicate the game log information and the campaign information with an integrated service module included in a game application in the game user terminal.

According to an embodiment, the method for providing personalized information further includes an LTV analysis step of calculating a life time value (LTV) of a game user using the game log information, wherein the LTV generates one of the plurality of user codes. It is characterized by being the basis.

According to an embodiment, the LTV may be calculated based on a survival probability and a purchase probability of a game user in a game application. The LTV may be calculated by further considering an average purchase amount, a survival probability decay rate, and a purchase probability decay rate of a unit period of a game user.

In one embodiment, the method for providing personalized information further includes receiving response information of a game user for the campaign from the game user terminal, and the response information may be reflected in the survival probability or the purchase probability. .

In one embodiment, the method for providing personalized information further includes an LTV analysis step of calculating a life time value (LTV) of a game user using the game log information, wherein the LTV evaluates adequacy of the campaign and the duration of the campaign. It can be used to determine whether or not.

In one embodiment, the method for providing personalized information further includes checking campaign constraints to determine whether to expose campaign information to the game user. The checking of the campaign constraints may include checking at least one of a campaign active period, repetition setting, frequency setting, restriction setting, and real-time exposure setting.

In one embodiment, the method for providing personalized information includes receiving response information of a game user for the campaign from the game user terminal, and whether the campaign is provided, continued, changed, or modified according to the received response information. Determining may be further included.

In one embodiment, the method of providing personalized information further includes a natural inference step of determining an actual game user, wherein the natural inference is information played by one or more games on a plurality of terminals or a plurality of games on one terminal. It is inferred whether one information is actually game play information by one game user, and the generating of the user pattern code may generate user pattern code for the inferred one game user.

In an exemplary embodiment, the natural inference step may include extracting a plurality of kit values representing a game user from the game log information, and determining an actual game user based on the plurality of kit values.

In one embodiment, the natural inference step uses an install kit value generated when installing a game application as a plurality of kit values and a device kit value generated in a device on which the game application is installed. It is done. The natural inference step may further use a channel interlocking kit value (Channel ID) or an advertising kit value (Advertising ID).

In one embodiment, the game of the personalization information providing method is characterized in that the game can be started without login. In one embodiment the game is in particular a mobile game.

In an embodiment, the personalization information providing server of the personalization information providing method may be operated by a company that services a plurality of types of games and may receive game log information of a plurality of types of games.

In one embodiment, the integrated service module of the personalization information providing method is embedded in a software development kit (SDK) of a game application provided to a game developer by a company serving a plurality of games.

A computer program stored in a computer readable recording medium according to one embodiment

When executed on a computer, program code for causing the computer to execute each step of the above-described method of providing personalization information.

According to a game application stored in a computer-readable recording medium according to an embodiment, the game log information including at least the game log information that is the basis for generating an LTV while connecting to the game server and running the game when running in the game user terminal Collect and transmit to the at least one of a personalization information providing server that is distinguished from the game server and the game server that is linked with the game application, and receives real-time customized campaign information according to a user pattern of a game user from the personalization information providing server. And a program module configured to display the real-time customized campaign information on the screen of the game user terminal and to transmit the game user's response information to the personalized information providing server to the displayed campaign information. Doing the above G module is characterized in that the built-in software development kit (SDK) for the company that provides services to a plurality of game for game development companies.

A system for providing real-time personalized information to a game user according to an embodiment of the present invention includes a user terminal for transmitting game log information in real time through an integrated service module included in a game application while playing a game by accessing a game server; It includes a personalization information providing server for receiving game log information from the integrated service module while the game is in progress, analyzing the user pattern, and providing real-time customized campaign information according to the user pattern in response to the state of the game user.

The personalization information providing server according to an exemplary embodiment of the present invention receives a game log information from a game application and generates a game user's state information, and processes a user pattern by batch processing the game log information of the game user. A user pattern analyzer for generating a code, and a campaign provider for matching real-time customized campaign information about the game user and the game user by using the state information of the game user and the user pattern code. It is characterized by. The method for providing personalized information in the personalized information providing server according to an embodiment of the present invention is a real-time processing method for providing personalized information, the method including receiving real-time game log information from a game application and indicating a game user from the game log information. Extracting a kit value, generating state information of a game user from the game log information, obtaining a natural person ID corresponding to the kit value, obtaining a user pattern code corresponding to the natural person ID; Querying campaign information based on the status information and the user pattern code.

The personalization information providing method in the personalization information providing server according to an embodiment of the present invention is a batch processing method for providing personalization information, the method comprising: collecting game log information of a predetermined period from a game application; Extracting a kit value from log information and generating a natural ID based on the extracted kit value; generating a user code from the collected game log information, and combining user codes to generate a user pattern code corresponding to the natural ID; It includes.

A user analysis system according to an embodiment of the present invention includes a database for matching, patterning and storing information played by a user from among play information played by a user in an existing game without a series of subscriptions; A database for collecting log information of a game in which a corresponding campaign (marketing / guide) is expressed in real time to update a game propensity or behavior pattern of a user; In the course of the game in which the campaign (marketing / guide) is expressed, a checkpoint is preset at a specific point, and it is checked whether the user satisfies a specific condition based on the user's play pattern and the game information, and checks the game. A control unit for transmitting the campaign (marketing / guide) information to the terminal of the user when the point is reached; And an operation unit that monitors the response rate and the performance of the user for the campaign (marketing / guide) in real time and supports the decision making of the campaign, and based on the points, the campaign, the user orientation / pattern, and the user response, AI can be learned and based on this, it can automatically support personalization service.

As described above, the personalized information providing method according to the present invention is to identify the game user's propensity and game behavior pattern, and based on the game log information coming in real-time customized guide or marketing information that can help game users in real time By improving the gameplay convenience, game sustainability (survival rate or survival rate) can be increased, and game sales can be increased.

In addition, according to the present invention, since it is possible to consider the LTV information that is the basis for generating one of the user codes in the user pattern code, that is, the revenue information that is expected to be generated by a specific game user for a certain period of time, In the long term, we can expect the effect of providing campaign information that can generate more sales.

In addition, according to the present invention, when analyzing the game user's propensity / game behavior pattern or LTV (Life Time Value), the game user's response rate and campaign performance for the customized campaign (guide and marketing) can be utilized together with more accurate customized campaign. It has the effect of providing information.

In addition, according to the present invention, the integrated service module included in the game application is produced and distributed by a company or a publishing company that services a plurality of games, not a game company, and a personalization information providing server that communicates with the integrated service module is a game server. It is operated by a service provider or a publishing company that is divided and operated, so that a game company or a publishing company analyzes game users in multiple games in real time without having to change the game application code. As a result, the campaign can be integrated and managed.

Conventionally, in order to provide a campaign to a user, the publishing company organizes the requirements and sends them to the game developer, and after the game developer completes updating the game program for providing the campaign, the inspection work of the mobile game download platform such as Apple or Google is performed. As it went through, it took at least 3-4 weeks. According to the configuration of the present invention, since the publishing company can be sent directly to the game user after development, it is possible to quickly provide a new campaign. By analyzing the game and operating subjects in the same way, game developers do not need to directly modify the game code, so that game operation can be facilitated.

In addition, according to the present invention, it is possible to specify a game user playing through various routes in a mobile game that can be executed without login, so that the game related information optimized for the actual game user can be provided. In addition, this has the effect of further increasing sustainability and sales. In other words, by identifying game users that are viewed as a plurality of game users at a glance as one game user, the user's behavior pattern can be analyzed more objectively and comprehensively, and thus, the most optimized campaign for each user. It can be provided. For example, a personalized store may be provided that consists only of products with high user preferences and products that are more likely to be purchased.

In addition, according to an embodiment of the present invention, in providing customized campaign information for individual game users using a user pattern code including LTV information, by additionally using an actual game user identification (natural identification method) Can be specified more precisely, resulting in the effect of providing accurate customized campaign information to precisely specified individuals.

The user's disposition and game behavior pattern can be grasped through various game play data of the user, and the user's behavior can be predicted in a new game through the user's existing game play data.

In addition, the performance and user response rate of the campaign (marketing / guide) can be identified in real time, and the decision on the sustainability of the campaign can be supported in real time.

1 is a schematic configuration diagram of a system for providing real-time personalization information to a game user according to the present invention.

2 is an internal configuration diagram of a personalization information server according to the present invention.

3 is a flowchart illustrating a batch processing process in a personalization information server according to the present invention.

4 is a signal flow diagram illustrating a process of providing real-time personalization information in a personalization information server according to the present invention.

5 is a flow chart illustrating a process for processing campaign constraints in accordance with the present invention.

6 is a flowchart illustrating a natural inference process for identifying a real game user according to the present invention.

7 and 8 are views showing the customized campaign information on the screen of the user terminal according to the present invention.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. The construction of the present invention and the effects thereof will be clearly understood through the following detailed description.

Prior to the detailed description of the present invention, the same components are denoted by the same reference numerals as much as possible even if displayed on different drawings, and the known components will be omitted if it is determined that the gist of the present invention may obscure the gist of the present invention. do.

1 shows a schematic configuration of a system for providing real-time personalized information to a game user (hereinafter, referred to as a personalized information providing system) according to the present invention.

As shown in FIG. 1, the personalization information providing system according to the present invention includes a user terminal 10, a game server 20, a personalization information providing server 30, and the like.

The user terminal 10 refers to a terminal possessed by a game user. The user terminal 10 installs a game application and accesses the game server 20 to proceed with the game. The game application installed in the user terminal 10 has a built-in integrated service module 12 (see FIG. 2). The integrated service module 12 performs a function of campaign, authentication, payment, etc. integrally.

The user terminal 10 transmits the game log information to the personalized information providing server 30 in real time through the integrated service module 12 while the game server 20 is connected to the game server 20. Game log information refers to all raw data generated in the process of installing, running, and removing a game application in a terminal of a game user. In one example, the game user's response information about the campaign to be described later may be included in the game log information. In another embodiment (not shown), the game server 20 is synchronized with the game application installed in the user terminal 10 so that the game log information generated in the user terminal is automatically stored in the game server 20, personalized information The providing server 30 may receive game log information from the game server 20. In the above cases, preferably, the user's response information about the campaign to be described later in the game log information may be transmitted directly from the integrated service module 12 to the personalization information providing server 30, the game about the goods of the game user Important information that should not be forged / modulated, such as log information (or goods information generated from this game log information), can be transmitted from the game server 20 synchronized with the game application to the personalized information providing server 30. Furthermore, in another embodiment, a part of the game log information is transmitted directly from the user terminal 10 to the personalization information providing server 30, and the remaining game log information is synchronized with the game application of the user terminal 10. From 20 may be transmitted to the personalized information providing server 30.

The user terminal 10 may be any kind of device as long as the device can install and run a mobile device or a game application such as a smartphone or a tablet PC.

The game server 20 is connected to the user terminal 10 and provides graphics or data necessary for game play to the user terminal 10. While the game is in progress, the game application of the user terminal 10 may directly process graphics or data, or when there is a large amount of data to be processed, the game server 20 may receive data from the game application, process it, and return it.

The game server 20 may be built and operated and managed by a game developer, or a company or a publishing company serving a plurality of games may be transferred from a game developer and directly operated and managed.

The personalization information providing server 30 receives game log information from a game application while a game is in progress on the user terminal 10 to analyze game behavior patterns (ie, user patterns) of game users and identify game user states. It provides real-time customized campaign information according to the game user status and user pattern.

The campaign refers to an action in which the personalized information providing server 30 intervenes in the game user's game according to a scenario in which a game user is made what proposal and in what situation. The customized campaign information may include game-related guide information that can be used for game users as a personalized game-related information, or promotion or marketing information such as personalized shops or item solicitation.

The personalization information providing server 30 is distinguished from the game server 20. The game server 20 processes and returns data so that the game can be played while the game is in progress, while the personalization information providing server ( 30) collects and processes game log information representing a game play situation from a game application in real time.

If the personalization information providing server 30 is managed by a publishing company rather than a game company, the personalization information providing server 30 utilizes the results generated by the collection and processing of game log information from various games to meet the needs of the game user. By providing tailored campaigns, you can maximize game revenue and advertising effectiveness.

2 shows an internal configuration of the personalization information providing server 30 according to the present invention.

As shown in FIG. 2, the personalization information providing server 30 communicates with the integrated service module 12 included in the game application of the user terminal 10 through the communication interface 120.

Game developers generally develop game applications using a software development kit (SDK) provided by a publishing company. In an embodiment of the present invention, an integrated service module is embedded in a software development kit of a publishing company. In addition, the game developer may load an API (Application Programming Interface) related to the integrated service module from the SDK and mount the campaign function according to the present invention in the game application as well as authentication and payment provided by the integrated service module.

As such, the integrated service module 12 mounted in the game application transmits game log information to the personalized information providing server 30 through a network while the game is in progress, and the personalized information providing server 30 communicates with the communication interface 120. Receive game log information through and deliver it to each component inside.

If the integrated service module 12 is provided for each game application, the personalization information providing server 30 collects game log information from various types of game applications and analyzes various games played by the game user, thereby not analyzing the entire game but the entire game. Game game patterns of game users can be identified.

The user pattern analyzer 102 generates a user pattern code by using game log information input through the communication interface 120.

The user pattern analyzer 102 generates a user code by batching game log information accumulated for a predetermined time period (for example, one day or three hours), and generates a user code by combining the generated user codes. The user pattern code is stored in the user pattern database 103 to update the user pattern code at regular intervals.

The user pattern analysis unit 102 generates a user code by referring to the user code database 109 according to a predetermined rule from the game log information, and generates a user pattern code for a natural ID using the user code and the natural ID. The natural ID is an ID given to an actual game user, which will be described later.

The user pattern code is information about the game propensity and behavior pattern of the game user. The user pattern code is given to the actual game user as complex information generated by the combination of the user codes.

Here, the user code refers to a definition of a situation or information occurring in game play of a game user. For example, a high billing code, a middle billing code, a low billing code, a non billing code, and the like may be defined as a user code related to a user's payment (billing).

As another example, the user code may be generated into morning, afternoon, and evening based on the user's game time zone, and the user code such as impulse purchase, safety seeking, and competition-oriented may be generated based on the user's game propensity. Can be.

As another example, the user code may be generated by overcoming certain obstacles during the game. For example, a user code may be generated, such as a type of overcoming through payment (billing) of the user, a type of overcoming with pure game play, a type of overcoming with a friend in the game, and the like.

As another example, a user code such as a high LTV, a medium LTV, a low LTV, or the like may be generated based on a life time value (LTV) to be described later.

As another example, a user code such as high skill, medium skill, low skill, etc. may be generated based on a play progress speed in a game.

The user pattern code is a combination of user codes and has a high billing, morning and impulsive tendency, overcomes payment through obstacles in the game, and has a low LTV user, or a low billing and evening type, and a safety pursuit type. In addition, there may be a combination of complex user codes such as a user who overcomes pure game play only when an obstacle occurs in the game, and a user having a high LTV. Depending on the type and number of user codes to be combined, various user pattern codes can be generated and the number of users belonging to each user pattern code will vary.

The natural person inference unit 104 processes the game log information collected for a predetermined time to generate natural person data, and stores and updates the generated natural person data in the natural person inquiry database 105. In addition, the natural person reasoning unit 104 determines a real game user in real time using a value for identifying a game user included in the game log information.

In a mobile game, it is difficult to identify a real game user because the game user does not receive a unique ID through a game service subscription and does not require the user to log in. In the embodiment of the present invention, the natural person means an actual game user, and the natural person inference unit 104 may infer a plurality of game users as one natural person using real-time game log information and natural person data.

When the natural inference unit 104 identifies the real game user in real time, the natural inference unit 104 obtains a natural ID corresponding to the real game user.

The LTV analysis unit 106 analyzes a game user's life time value (LTV) by using game log information of a predetermined period and updates the LTV database 107.

LTV is the revenue expected to be generated by a certain game user over a period of time. Here, the predetermined period may be a period in which the service of the game is maintained (Product Life Cycle: PLC), or may be a period until the next update of the game.

In one embodiment, LTV can be obtained using the survival probability and the purchase probability. Survival probability refers to the probability of continuing to use a game without leaving it. In one embodiment, survival probability may be calculated from, for example, game log information such as the number of days of attendance in the last week, the number of days since the last connection, the number of plays per connection in the last week, and whether or not one week has passed. The probability of purchase refers to the probability of purchasing a particular game in a certain period of time. In one embodiment, the probability of purchase may include, for example, the number of visits for 7 days after joining, the rate of change in the number of visits according to the number of visits, the amount of the in-game goods (hereinafter referred to as primary goods) that can be purchased with real money, and the ratio of the amount used; Simply repeat the steps you have previously played to obtain farming out of the total number of game plays and the percentage of the amount and amount of goods you have purchased (hereinafter referred to as secondary goods) from the primary good. Act), game log information such as the mobile device used transaction price, and the like. Using the above game log information as a variable, the probability of purchase can be determined through inference methods such as decision trees, logistic regression, random forest, and SVM. For example, if LTV is a monthly period, (average monthly purchase) * (survival probability) * (purchase probability) * (1 + α ₁ * β ₁ + α ₂ * β ₂ +… + α _n * β _n ) can be generalized. Here, α denotes a survival probability decay rate, β denotes a purchase probability decay rate, and n denotes a month. The decay rate of each month may be set to an appropriate constant according to an embodiment. If a game user purchases an average of 100,000 won in a game that has made a purchase, and a one month survival probability is 0.8 and a monthly purchase probability is 0.9 in a new game, the estimated sales for three months is 100,000 won * 0.8 * 0.9 + 10 10,000 won * 0.8 * α ₁ * 0.9 * β ₁ + 100,000 won * 0.8 * α ₁ ² * 0.9 * β ₁ ² . In this calculation, α ₂ is α ₁ ² , β ₂ is β ₁ ² .

The real time event processor 108 receives game log information through the communication interface 120 and processes the game log information in real time to generate state information of the game user. The real-time event processor 108 may determine the current state of the game user from the game log information.

The game user's status information refers to the game user's status, which is captured in real time from the game log information during game play, such as the game user has won 5 consecutive stages, the item has changed using the item, or the current stamina (energy) is insufficient. It's about the state.

The campaign provider 110 provides personalized campaign information to the game user based on the game user's state information and the user pattern code. In one embodiment, the LTV information may be included in the user pattern code. In this embodiment, when providing campaign information, it is possible to consider LTV, i.e., sales that are "expected" to be generated by a particular game user over a period of time, so that campaign information is based only on the average amount of ARPU spent during a particular period of time. Compared to the case of providing, it is possible to expect the effect of providing game users with campaign information that can generate more sales in the long term. In the short term, campaigns with higher ARPU may have lower LTV values, and campaigns with lower ARPU may have higher LTV values.

The campaign provider 110 receives the game user's status information from the real-time event processor 108, receives the natural ID from the natural inference unit 104, and checks the user pattern code corresponding to the natural ID. Campaign information according to the user pattern code is retrieved from the campaign database 113.

The campaign database 113 stores campaign information corresponding to status information and user pattern codes. Campaign information is generated by a campaign management tool (not shown) that is present inside or separately from the campaign provider 110. The campaign management tool generates campaign information optimized for various predefined status information and user pattern codes, and edits or changes the campaign information by reflecting state information, user pattern codes, and campaign response information, which are added later. 111) can be updated.

The campaign provider 110 transmits the customized campaign information retrieved from the campaign database 111 to the integrated service module 12 of the game application through the communication interface 120.

In addition, the campaign provider 110 may determine whether to provide or continue the campaign by checking the campaign response information of the game user included in the game log information, and may change or modify the campaign information.

The campaign response information is information about how the game user reacted to the campaign (game guide, personalized store, marketing, promotion) provided to the game user. For example, the game user does not check the campaign popup window. A game user's behavior in response to a campaign, such as closing it immediately, buying an item with guidance, or using a skill.

3 shows a batch process performed in the personalization information providing server according to the present invention.

Referring to FIG. 3, the personalization information providing server 30 collects game log information while storing game log information received from a game application in an internal buffer memory (not shown) (S10).

After checking whether it is time to batch process the game log information (S20), if the batch processing time has not yet reached, game log information is continuously collected. That is, the personalization information providing server 30 performs an ETL (Extract, Transform, Load; Extract, Transform, Load) process for every game log information generated in the game at predetermined intervals.

When the batch processing time arrives, game log information stored in the internal buffer memory is transferred to each component to perform batch processing.

The batch processing is a batch processing of data accumulated for a predetermined time. In the embodiment of the present invention, batch processing is performed for user pattern analysis (S30), natural inference (S40), and LTV analysis (S50).

Batch processing time and period can be set in the system. For example, user pattern analysis, natural inference, and LTV analysis can be set to 1 hour period, 3 hours period, 24 hours period, etc. ) Or batch processing may be performed based on 02:00 am. Therefore, if the batch is processed every 3 hours based on 02:00 am, batch processing may proceed to 02:00, 05:00, 08:00 ... 23:00 8 times a day.

The user pattern analysis S30 is performed by the user pattern analysis unit 102 of the personalization providing server 30. When the batch processing time arrives, the personalization providing server 30 analyzes the user pattern using the game log information stored in the internal buffer memory.

The personalization providing server 30 generates a user code from game log information and combines the user codes to generate a user pattern code corresponding to a natural ID. The user pattern code generated through the user pattern analysis is stored in the user pattern database 103 and the user pattern database 103 is updated (S32).

In general, user pattern codes have a four-step life cycle.

In the initial stage, the user pattern code is registered, and in the ready stage, the calculation is completed so that the user pattern code can be used for the campaign service. The running stage includes game user and campaign information. The user pattern code is exposed to be matched, and the stopped step is a state in which the user pattern code is inactivated and is not utilized for the campaign service.

Natural inference S40 is performed by the natural inference unit 104 of the personalization providing server 30. When the batch processing time arrives, the personalization providing server 30 performs natural inference using game log information stored in the internal buffer memory.

The personalization providing server 30 extracts kit values that may be the basis of game user identification from the game log information, and generates natural data using the kit values. Natural data generated through natural inference is stored in the natural query database 105 so that the natural query database 105 is updated (S42).

The natural inference process will be described with reference to FIG. 6.

In an embodiment of the present invention, a kit value for identifying a game user is utilized for natural inference. These kit values are included in the game log information, and the following kit values are used in the embodiment of the present invention.

An install kit value is a kit value generated when a game is executed after installing a game application. The install kit value is generated by the integrated service module 12.

The install kit value is maintained as long as the game user does not delete the game application. However, when the game application is deleted and a new game is installed, a new install kit value is issued.

When you play a game, you provide items or skills to the game user through a random draw product at a particular stage. Many people tend to repeat a particular stage to get the product they want. At this time, since the game application must be repeatedly deleted and reinstalled, there may be many imaginary install kit values in one game user.

The device kit value is a shared kit value generated for each mobile device. The device kit value is generated at the mobile device.

Although the device kit value may be known to be played by one game user even if several games are played on one device, the kit value changes when the game user changes the device.

Channel ID is a kit value generated when playing a game through an SNS account.

Even if one game user plays multiple games on multiple devices, the kit value is the same if they are authenticated through the same SNS account, but there is a problem that only a game with a channel interworking function is possible.

The advertising kit value (Advertising ID) is a kit value provided for receiving an advertisement in a mobile operating system (OS).

The advertising kit value has a disadvantage in that the value of identifying the terminal user to compensate or benefit when the advertisement is clicked or the advertising kit value is not known when the game user blocks the advertisement receiving function.

In this way, the personalization information providing server 30 extracts a plurality of kit values from the game log information collected for a predetermined time (S150). In an embodiment of the present invention, batch processing may be performed in various cycles. If 24 hours is a cycle, the game log information of the previous day may be used for batch processing. If used, game log information corresponding to the cycle time can be used for batch processing. In the specification of the present invention, the game log information of the previous day's game value or the game log information of the batch cycle time is expressed as D-1 game log information. That is, the game log information most recently ETL (extracted, converted, loaded) is defined as D-1 game log information.

Next, the same kit value is grouped from the plurality of kit values extracted from the D-1 game log information (S152).

The kit values extracted from the game log information are extracted as pairs of mutually correlated kit values, and are grouped into one group by binding a pair having at least one identical kit value from the pair of kit values.

The natural ID is generated and assigned to each grouped group as described above (S154). If a natural ID is given to a grouped kit value pair, D-1 natural data is generated. D-1 natural data refers to natural data generated from D-1 game log information.

The D-1 natural data is mapped with all natural data (D-2 natural data) before the current batch period (S156), and the new D-2 natural data is updated (S158). Here, the D-2 natural person data is the natural person data generated based on the most recent ETL game log information, that is, all previous game log information except the D-1 game log information.

As described above, a process of generating D-1 natural data and updating D-2 natural data will be described.

First, suppose that N0001: (P1-D1, P2-D1, P3-D1) is included in D-2 natural data. Here, N0001 represents a natural ID, P1, P2, and P3 represent an installation kit value (Install key), and D1 represents a device kit value (Device key).

P1-D1 is a kit value pair generated when a game user installs a game application on the D1 device and proceeds with the game. P2-D1 is a game user deletes the P1 game from the D1 device and reinstalls the game application. P3-D1 is a kit value pair generated when the user progresses, and P3-D1 is a kit value pair generated when the user deletes the P2 game from the D1 device and reinstalls the game application to proceed with the P3 game. In particular, it is assumed that after the user installs the P3 game on the D1 device, the user performs channel input through the SNS account to input additional account information.

In the presence of such D-2 natural data, if a game user plays a P3 game on a D1 device and then installs a game application through an SNS account on another device, that is, the D2 device, the D-1 game log information is displayed. P3-D1 and P3-D2 kit values can be extracted.

That is, P3-D1 contained in D-2 natural data has channel linking kit value, which is additional linkage information through SNS account. When installing game application in D2 device, new kit value P4 is generated, but the user has channel linking information. If the existing information is loaded by using, P3, which is an installation kit value in the D1 device, is loaded into D2 which is the current device, and a P3-D2 kit value pair is generated. Specifically, when the game application is installed on the D2 device, once the P4-D2 is generated, and the user works with the P3-D1 via the SNS account, the personalization information providing server 30 replaces the kit value with the P4 to the P3 D2 device. In communication with the P3-D2 kit value is generated in the D2 device.

Therefore, if P3-D1 and P3-D2 are extracted from the D-1 game log information, the installation kit value P3 is the same, and thus the same value (pair value) is given to P3-D1 and P3-D2. P3-D1 and P3-D2 given the same value in D-1 game log information are grouped into one group.

Natural IDs are generated in the grouped P3-D1 and P3-D2 to generate the natural data of D-1 including N0002: (P3-D1, P3-D2). D-2 natural data contains N0001: (P1-D1, P2-D1, P3-D1), and D-1 natural data contains N0002: (P3-D1, P3-D2), so P3 and D1 are common to be.

When D-1 natural data is mapped to D-2 natural data, P3 and D1 are common so that N0001: (P1-D1, P2-D1, P3-D1, P3-D2) is present in the updated D-2 natural data. do. Herein, the natural ID may maintain the N001 value of the D-2 natural data before updating.

The natural data generated during the natural inference process through the batch processing is stored in the natural query database 105, and the natural ID can be obtained by querying the natural query database 105 when identifying the actual game user through real-time processing. have.

In the above example, if the natural ID is queried in real time, P3-D3 kit value is first extracted from the game log information, and the natural inquiry database 105 searches the D-2 natural data to confirm that P3 is common. The game user who plays the game with D3 can be determined by the natural ID N001.

D-2 It is assumed that natural data includes N0001: (P1-D1, P2-D1, P3-C1-D1). Here, N0001 represents a natural ID, P1, P2, and P3 represent an installation kit value (Install key), C1 represents a channel interlocking kit value (Channel ID), and D1 represents a device kit value (Device key).

P1-D1 is a kit value set generated when a game user installs a game application (A) on a D1 device and proceeds to play a game. P2-D1 is a game user deletes a P1 game from a D1 device and the same game application (A). Is a set of kit values generated when P2 game is re-installed and P3-C1-D1 is a kit value generated when a user plays a P3 game by installing a new game application (B) through the SNS account on the D1 device. Three.

In the presence of such D-2 natural data, a game user plays a P3 game on a D1 device and then installs another new game application (C) on another device, that is, on the D2 device using the same SNS account. P3-C1-D1 and P4-C1-D2 kit values can be extracted from -1 game log information.

That is, since the other game application (C) is installed on the D2 device but installed through the same SNS account, the install kit values are different from P3 and P4, but the channel interlocking kit values are the same as C1.

Therefore, if P3-C1-D1 and P4-C1-D2 are extracted from the D-1 game log information, the same value (set value) is given to P3-C1-D1 and P4-C1-D2 because the channel interlocking kit value C1 is the same. . P3-C1-D1 and P4-C1-D2 given the same value in D-1 game log information are grouped into one group.

Natural IDs are generated in the grouped P3-C1-D1 and P4-C1-D2 to generate D-1 natural data including N0002: (P3-C1-D1, P4-C1-D2). D-2 natural data contains N0001: (P1-D1, P2-D1, P3-C1-D1), and D-1 natural data contains N0002: (P3-C1-D1, P4-C1-D2). Therefore, P3, C1, and D1 are common.

When D-1 natural data is mapped to D-2 natural data, P3, C1, and D1 are common, so the updated D-2 natural data is N0001: (P1-D1, P2-D1, P3-C1-D1, P4-C1 -D2) will be present. Herein, the N001 value of the D-2 natural data before the update may be maintained.

If the natural ID is queried in real time with the above example, when the P4-C1-D3 kit value is extracted from the game log information, the natural inquiry database 105 inquires the D-2 natural data and confirms that C1 and P4 are common. Thus, the game user who plays the game with P4-C1-D3 can be determined as the natural person ID N001.

On the other hand, in a kit value set extracted from game log information, a large number of kit value sets may be included in one natural ID.

For example, when a person who acts for game settlement performs a purchase settlement for a game item or the like, a large number of install kit values are generated in one device, and a natural ID is given to a large number of kit value sets in common with device kit values. . Since such kit value sets are included in natural data, reliability of natural inference may be deteriorated. Therefore, abnormal kit value sets need to be excluded from mapping of natural data.

In addition, when the game is tested for QA (Quality Assurance) before the game is released, the same installation kit value may be issued in another game. These test install kit values do not reflect general game users and should be excluded from the mapping of natural data. The test install kit value contains a specific game code value and can be distinguished from the normal install kit value.

There is also a device that continuously issues new install kit values for one game. Such a device is called a factory device, and the factory device sells an installation kit value obtained by acquiring good characters and items. The installation kit value created in the factory device does not reflect the game user who played the game normally and should be excluded from the mapping of natural data. The device kit value of the factory device can be distinguished from the device kit value of the general game user.

In addition, when playing a game using an emulator, all game users using the emulator have the same device kit value. Games played in the emulator are not normal play and should be excluded from the mapping of natural data. The device kit value of the device for the emulator can be distinguished from the device kit value of a general game user.

The personalization information providing server uses the natural data generated through the natural inference process to confirm that the information played by the game on the plurality of terminals and / or the information played by the plurality of games is actually game play information by one game user. Can be identified.

One feature of the present invention is to provide customized campaign information for individual game users using a user pattern code including LTV information, and by using the above-described natural identification method, the individual can be more accurately identified. As a result, it is possible to provide accurate customized campaign information to precisely specified individuals.

The LTV analysis S50 is performed by the LTV analysis unit 106 of the personalization providing server 30. When the batch processing time arrives, the personalization providing server 30 performs LTV analysis using game log information and user pattern information stored in an internal buffer memory. The LTV analysis result generated through the LTV analysis is stored in the user pattern database 103 and the LTV database 107 is updated (S52).

The personalization information providing server 30 generates the LTV corresponding to the natural person ID using the game log information. The personalized information providing server 30 may use the campaign response information among the game log information for the LTV analysis. By using campaign response information in LTV analysis, you can numerically check the performance of your campaign and use it in your marketing strategy.

The personalization information providing server 30 may obtain the LTV reflecting the campaign response result by further considering the campaign response information when obtaining the parameter of the LTV function for calculating the LTV. For example, an event that closes within 1 second of exposure of a campaign pop-up window, an event that closes 1 second after exposure of a campaign pop-up window, an event that explicitly closes the campaign pop-up window, or the user moves away from the campaign pop-up. Or events that expire after the campaign expires, events that enter the first phase of the campaign, events that enter the second phase of the campaign, events that purchase items proposed by the campaign, or events that utilize skills. By analyzing a variety of predefined events, these event analysis results can be reflected in the probability of survival or purchase.

If a certain campaign is provided to a user and the LTV rises, the campaign can be recognized as an effective campaign, and if the same campaign is provided to other users of the same user pattern code as the above user, the campaign effect will be improved.

4 shows a process of providing real-time personalized information in the personalized information providing server according to the present invention.

Referring to FIG. 4, first, game log information received from the integrated service module 12 of the game application is input to the real-time event processor 108 and the natural inference unit 104 (S60).

The natural person inference unit 104 extracts a plurality of kit values from the input game log information and inquires the natural person inquiry database 105 (S62) to obtain a natural person ID based on the kit value (S63).

The natural person inference unit 104 transmits the natural person ID obtained from the natural person inquiry database 105 to the campaign provider 110 (S64).

On the other hand, the real-time event processing unit 108 generates the state information of the game user from the received game log information and delivers it to the campaign provider 110 (S66).

The campaign provider 110 checks the user pattern database 103 based on the natural ID received from the natural inference unit 104 (S68), and checks the user pattern code corresponding to the natural ID (S69).

Next, the campaign provider 110 searches the campaign constraints database 113 to determine whether the campaign is exposed (S70), and obtains the constraints (S71).

The campaign provider 110 processes the constraint (S72) to determine the exposure or non-exposure of the campaign. Constraints processing of the campaign provider 110 will be described with reference to FIG. 5.

As shown in FIG. 5, first, the campaign provider 104 checks whether the campaign is active now (S100), and determines that the campaign is not exposed (S120).

If the campaign is active, it is checked whether there is a repetition setting (S102), and if there is a repetition setting, it is checked whether it is a corresponding date on the repetition setting criteria (S104). If there is no repetition setting, the flow proceeds directly to the frequency setting checking step (S106).

Here, the repetition setting refers to repetition of the period during which the campaign exposure is allowed, such as the campaign exposure is allowed from 1st to 5th of each month. There may be a plurality of repetition setting conditions, and if any one condition is satisfied, it is passed.

If it is not the corresponding day on the repetition setting criteria, it is determined as the campaign non-exposure (S120).

If there is a frequency setting, it is checked whether the current exposure number is the maximum exposure number or more (S108). If there is no frequency setting, the control proceeds directly to the constraint setting check step (S112).

If the maximum number of exposures or more is determined as the campaign non-exposure (S120), if not more than the maximum number of exposures to determine whether the exposure interval remains (S110).

In this case, the exposure interval refers to a time interval after which the campaign information is exposed to wait for the next campaign information to be exposed.

Therefore, if the exposure interval remains, it is determined as the campaign non-exposure (S120). If the exposure interval does not remain, it is checked whether the restriction condition is set (S112).

If the restriction condition is set, it is checked whether the current product purchase number is the product purchase limit number or more (S114). If the restriction condition is not set, the process immediately enters the real time exposure condition check step (S116).

If the product purchase limit is more than the number of times to determine the campaign non-exposure (S120), if not more than the product purchase limit is determined whether the real time exposure conditions are set (S116).

The real-time exposure condition refers to determining the campaign exposure based on the current state of the game user. For example, you can expose a campaign when one or more combinations of states are identified, such as a game user who lost three consecutive games, a stamina (less than 2 energy), or a gift in the mailbox.

If the real-time exposure conditions are not set, the campaign is exposed immediately (S122). If the real-time exposure conditions are set, it is checked whether all the real-time exposure conditions are satisfied (S118).

If any of the real-time exposure conditions are not satisfied, it is determined that the campaign is not exposed (S120), and if all the real-time exposure conditions are satisfied, the campaign is exposed (S122).

As described above, if the campaign constraints are determined to be the campaign exposure, the campaign provider 110 queries the campaign database 111 (S74).

The campaign provider 110 matches the game user identified by the natural ID to the campaign information based on the game user's state information and the user pattern code to obtain customized campaign information optimized for the game user (S75). The information is transmitted to the integrated service module 12 (S76).

The user terminal 10 receives the customized campaign information through the integrated service module 12 of the game application and displays it on a specific portion of the game screen in progress.

7 and 8 show that the customized campaign information according to the present invention is displayed on the game screen of the user terminal 10.

In FIG. 7, (a) shows campaign information 1 for guiding a growth package, and (b) shows campaign information 2 for recommending item purchases to a store.

The growth package campaign information 1 is displayed in an opaque state covering a part of the game on the game screen, but the shop guide campaign information 2 is displayed in a translucent or transparent state that does not cover the game screen.

In FIG. 8, (a) shows campaign information 3 for recommending a product, and (b) shows recommended product information displayed when entering according to the campaign information 3.

The product recommendation campaign information 3 is displayed in a minimized form on a specific portion in a direction that does not disturb the game of the game user.

The recommended product information according to the product recommendation campaign information 3 is not information common to all users but personalized information personalized according to the current state of the game user and the game behavior pattern.

For example, if there is a game user who shows a game behavior pattern that gives up the game and purchases no items at least 4 consecutive games, and the game user confirms the 3 consecutive losing status, the item is purchased at 50% discount of the current price. Inviting campaign information may be exposed to the game user in real time.

The exemplary methods according to the present invention are computer program products recorded on a recording medium readable by a computer (including an apparatus having an information processing function), program instructions executed by a processor, a software module, microcode, a computer, Applications, logic circuits, application specific semiconductors, or firmware can be implemented in a variety of ways. Examples of the computer-readable recording medium include, but are not limited to, a ROM, a RAM, a CD, a DVD, a magnetic tape, a hard disk, a floppy disk, a hard disk, an optical data storage device, and the like. The computer readable recording medium can also be distributed over network coupled computer systems so that the computer readable code is stored and executed in a distributed fashion.

The above description is merely illustrative of the present invention, and various modifications may be made by those skilled in the art without departing from the technical spirit of the present invention.

Therefore, the embodiments disclosed in the specification of the present invention are not intended to limit the present invention. The scope of the present invention should be construed by the claims below, and all techniques within the scope equivalent thereto will be construed as being included in the scope of the present invention.

Claims

A method of providing personalized information to a game user terminal provided with a game application by a server for providing personalized information distinguished from a game server,

Generating state information of the game user from the game log information;

Accumulating and batching the game log information for a predetermined period to generate a plurality of user codes, and combining them to generate user pattern codes for game users;

Matching real-time customized campaign information for the game user and the game user using the state information of the game user and the user pattern code;

And transmitting the matched real-time customized campaign information to the terminal of the game user.
The method of claim 1,

And receiving the game log information from at least one of the game user terminal in which a game application is installed, or a game server synchronized with the game application.
The method of claim 1,

And the personalization information providing server communicates the game log information and the campaign information with an integrated service module included in a game application in the game user terminal.
The method of claim 1,

The method further includes an LTV analysis step of calculating a life time value (LTV) of a game user using the game log information, wherein the LTV is based on generating one of the plurality of user codes. Way.
The method of claim 4, wherein

And the LTV is calculated based on a survival probability and a purchase probability of a game user in a game application.
The method of claim 5, wherein the LTV is calculated by further considering an average purchase amount, a survival probability decay rate, and a purchase probability decay rate of a unit period of a game user.
The method of claim 5,

And receiving response information of a game user for the campaign from the game user terminal, wherein the response information is reflected in the survival probability or the purchase probability.
The method of claim 1,

LTV analysis step of calculating a game user's LTV (Life Time Value) using the game log information, the LTV is characterized in that it is used to evaluate the appropriateness of the campaign and to determine whether or not to continue the campaign. How to provide personalized information.
The method of claim 1,

Identifying campaign constraints to determine whether to expose campaign information to the game user.
The method of claim 9,

The checking of the campaign constraints may include identifying at least one of a campaign active period, repetition setting, frequency setting, restriction setting, and real-time exposure setting.
The method of claim 1,

Receiving response information of a game user for the campaign from the game user terminal;

And determining whether to provide, continue, change, or modify the campaign according to the received reaction information.
The method of claim 1,

It further includes natural reasoning steps to determine the actual game user,

The natural inference is inferring whether information of playing more than one game on a plurality of terminals or playing information of a plurality of games on a single terminal is actually game play information by one game user, and generating the user pattern code. Personalization information providing method for generating a user pattern code for the inferred one game user.
The method of claim 12,

The natural inference step may include extracting a plurality of kit values representing a game user from the game log information, and determining an actual game user based on the plurality of kit values.
The method of claim 13,

The natural inference step may include an installation kit value generated when installing a game application as a plurality of kit values and a device kit value generated by a device on which the game application is installed. How to Provide.
The method of claim 14,

The natural inference step of providing personalized information, characterized in that additionally using a channel linkage kit value (Channel ID) or advertising kit value (Advertising ID).
The method of claim 12,

The game is a personal information providing method characterized in that the game can be started without login.
The method of claim 16,

The game is a personalized information providing method, characterized in that the mobile game.
The method of claim 1,

The personalization information providing server is operated by a company that services a plurality of types of games, personal information providing method characterized in that it can receive the game log information of the plurality of types of games.
The method of claim 3,

The integrated service module is a method of providing personalized information, characterized in that embedded in a software development kit (SDK) of a game application provided to a game developer by a company that services a plurality of games.
Personalized information providing server that is distinct from the game server,

A real-time event processor for generating game user status information from game log information;

A user pattern analyzer for accumulating and batching the game log information for a predetermined period to generate a plurality of user codes, and combining them to generate a user pattern code for a game user;

A campaign provider for matching real-time customized campaign information about the game user and the game user by using the state information of the game user and the user pattern code;

A communication interface for receiving the game log information and transmitting the matched real-time customized campaign information to the terminal of the game user,

Each step of the method for providing personalized information according to any one of claims 1 to 19 can be executed.

Personalization information server.
The method of claim 20,

Personalization information providing server further comprises an LTV analysis unit for calculating a game user's LTV (Life Time Value) using the game log information.
The method of claim 20,

Personalization information providing server further comprises a natural reasoning unit for determining the actual game user.
A user terminal which accesses a game server and transmits game log information in real time through an integrated service module included in a game application while playing a game;

A system for providing real-time personalized information to a game user, comprising a server for providing personalized information according to claim 20.
A computer program stored in a computer readable recording medium,

20. A computer program comprising program codes which, when executed on a computer, cause the computer to execute each step of the method for providing personalization information according to any one of claims 1 to 19.
A game application stored on a computer readable recording medium.

When executed in a game user terminal, the game log information including at least the game log information that is the basis for generating the LTV while connecting to the game server to proceed with the game to collect, the personalization information providing server that is distinct from the game server, and Transmit to at least one of the game servers linked with a game application,

Receive real-time customized campaign information according to the user pattern of the game user from the personalization information providing server,

Displaying the real-time customized campaign information on the screen of the game user terminal,

And a program module composed of computer program code for implementing transmission of the reaction information of the game user to the displayed campaign information to the personalized information providing server.

Wherein the program module is embedded in a software development kit (SDK) provided to a game developer by a company serving a plurality of games.