KR102187916B1 - Method and apparatus for game event guide - Google Patents

Abstract

A method and apparatus for guiding a game event are disclosed. A game event guide method according to an embodiment includes game information about a game of interest, event information that can be performed on a game of interest, and a game operation index measured at a first point in time about the game of interest, in a pre-learned index prediction model. By inputting, predicting the game operation indicators of the target time interval related to the game of interest, and selecting the optimal events for the game of interest for each time point within the target time interval by considering the game operation indexes of the predicted target time interval With reference to an event effect table for storing detailed information of each event of the step of, and each game information on the sample games, each event information performed on the sample games, and each event information performed on the sample games, Based on the step of generating an event execution strategy for each of the selected optimal events, and the event execution strategy for each of the selected optimal events and the selected events, guide information on event performance for the game of interest is provided. And providing.

Description

METHOD AND APPARATUS FOR GAME EVENT GUIDE}

The following embodiments relate to a method and apparatus for a game event guide.

When a game is released, game-related events may be provided at an appropriate time to activate the game. For example, a game event may include a reward or promotion. The effect of providing such a game event may vary depending on the timing of the event and the type of event to be provided. Depending on the game genre, operating country, operating period, update status, etc., there may be an appropriate type of event or an appropriate event provision time. In connection with this, user analysis, game index analysis, and user churn prediction have been performed. On the other hand, as a solution to the problem of classifying input patterns into specific groups in recent years, research to apply an efficient pattern recognition method possessed by humans to real computers has been actively conducted. As one of these studies, there is a study on an artificial neural network that models the characteristics of human biological neurons by mathematical expressions. The artificial neural network can generate a mapping between the input pattern and the output pattern, and the characteristics of the artificial neural network can be utilized for providing events or guiding other game operations.

According to an embodiment, a game event guide method performed by a game operating device includes game information on a game of interest, event information that can be performed on the game of interest, and a game at a first point in time measured on the game of interest. Inputting an operation index into a pre-learned index prediction model, and predicting game operation indexes in a target time interval related to the game of interest; Selecting optimal events for the game of interest for each time point in the target time interval, taking into account the game operation indicators of the predicted target time interval; The selection is made by referring to an event effect table that stores detailed information on each game of the sample games, information on each event performed on the sample games, and detailed information on each event performed on the sample games. Generating an event execution strategy for each of the optimized events; And providing guide information on performing an event for the game of interest based on the selected optimal events and an event execution strategy for each of the selected events.

Selecting the optimal events may include selecting the optimal events based on each event information corresponding to the maximum value of the game operation indicators in the predicted target time interval.

The event effect table may further store game operation indicators measured for the sample games, and the step of generating the event execution strategy includes a maximum value of the game operation indicators measured for the sample games or the sample games It may include the step of generating the event execution strategy, based on the detailed event information corresponding to the maximum value of the increase amount of the game operation indicators measured for.

The game event guide method is performed after the step of predicting the game operation indexes, comparing the game operation indexes of the predicted target time interval with a first threshold, or the predicted target time interval. Comparing a change amount of game operation indicators with a second threshold; Determining a state of operation of the game of the game of interest based on the comparison result; And providing a risk warning regarding the game operation of the game of interest based on the determined state of the game operation.

The target time interval may include a second time point after the first time point, and a third time point after the second time point, and the index prediction model is the game information regarding the game of interest, and the game of interest. Based on the event information that can be performed, and the measured game operation index at the first time point, the game operation index at the second time point may be predicted, and the game information regarding the game of interest and the game of interest The game operation index at the third time point may be predicted based on the event information that can be performed on the related event and the predicted game operation index at the second time point.

The game operation indicator may include a game state indicator and a product life cycle (PLC) indicator. The index prediction model is based on the game information on the game of interest, the event information that can be performed on the game of interest, and the game state index measured on the game of interest at the first time point, the target time A game state indicator prediction model for predicting game state indicators of a section, and the game information on the game of interest, the event information that can be performed on the game of interest, and the first time point measured for the game of interest. Based on the PLC index, a PLC index prediction model for predicting PLC indexes of the target time interval may be included.

The game information may include genre information and operation information. The operation information may include operation country information, operation period information, and game update information.

The game event guide method includes: collecting game log information, game meta information, and event history, which is performed before the step of predicting the game operation indicators; Generating a descriptor table by processing the collected game log information, the game meta information, and the event history; And training the index prediction model by using the descriptor table.

According to an embodiment, a game operating device includes a processor; And a memory for storing instructions that can be read by the processor, and when the instructions are executed by the processor, the processor provides game information on a game of interest, event information that can be performed on the game of interest, and By inputting the game operation index of the first time point measured about the game into the pre-learned index prediction model, predicting the game operation index of the target time interval related to the game of interest, and the game operation of the predicted target time interval In consideration of the indicators, optimal events for the game of interest are selected for each time point within the target time interval, and game information about sample games, each event information performed on the sample games, and the sample An event execution strategy for each of the selected optimal events is generated by referring to an event effect table storing detailed event information of each event information performed on games, and the selected optimal events and Guide information on event execution for the game of interest is provided based on the event execution strategy for each of the selected events.

1 is a diagram illustrating a process of generating a descriptor table and an event effect table according to an embodiment.
2 is a diagram illustrating a learning process of an index prediction model according to an embodiment.
3 is a diagram illustrating a process of setting a global parameter according to an embodiment.
4 is a diagram illustrating an indicator prediction model according to an embodiment.
5 is a diagram illustrating an event simulation process according to an embodiment.
6 is a view showing an event simulation result according to an embodiment.
7 is a diagram illustrating a process of selecting an optimal event according to an embodiment.
8 is a diagram showing an event effect table according to an embodiment.
9 is a view showing a game operating device according to an embodiment.
10 is a view showing a game operation guide method according to an embodiment.

Specific structural or functional descriptions of the embodiments according to the concept of the present invention disclosed in this specification are exemplified only for the purpose of describing the embodiments according to the concept of the present invention, and embodiments according to the concept of the present invention They may be implemented in various forms and are not limited to the embodiments described herein.

Since the embodiments according to the concept of the present invention can apply various changes and have various forms, the embodiments will be illustrated in the drawings and described in detail herein. However, this is not intended to limit the embodiments according to the concept of the present invention to specific disclosed forms, and includes changes, equivalents, or substitutes included in the spirit and scope of the present invention.

Terms such as first or second may be used to describe various elements, but the elements should not be limited by the terms. The above terms are only for the purpose of distinguishing one component from other components, for example, without departing from the scope of rights according to the concept of the present invention, the first component may be named as the second component, Similarly, the second component may also be referred to as a first component.

When a component is referred to as being “connected” or “connected” to another component, it is understood that it may be directly connected or connected to the other component, but other components may exist in the middle. Should be. On the other hand, when a component is referred to as being “directly connected” or “directly connected” to another component, it should be understood that there is no other component in the middle. Expressions that describe the relationship between components, for example, “between” and “just between” or “directly adjacent to” should be interpreted as well.

The terms used in the present specification are only used to describe specific embodiments and are not intended to limit the present invention. Singular expressions include plural expressions unless the context clearly indicates otherwise. In the present specification, terms such as "comprise" or "have" are intended to designate that the specified features, numbers, steps, actions, components, parts, or combinations thereof exist, but one or more other features or numbers, It is to be understood that the presence or addition of steps, actions, components, parts, or combinations thereof, does not preclude the possibility of preliminary exclusion.

Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by a person of ordinary skill in the art to which the present invention belongs. Terms as defined in a commonly used dictionary should be interpreted as having a meaning consistent with the meaning in the context of the related technology, and should not be interpreted as an ideal or excessively formal meaning unless explicitly defined in this specification. Does not.

Hereinafter, embodiments will be described in detail with reference to the accompanying drawings. However, the scope of the patent application is not limited or limited by these embodiments. The same reference numerals in each drawing indicate the same members.

1 is a diagram illustrating a process of generating a descriptor table and an event effect table according to an embodiment. Referring to FIG. 1, a descriptor table 140 and an event effect table 150 are generated based on data collected from a game client 111 and data collected from a game database (DB) 112.

The game log is collected from the game client 111 and the game log table 121 may be generated. In addition, game meta information may be collected from the game DB 112 to generate a game meta information table 112, and event history may be collected from the game DB 112 to generate an event history table 123.

This data collection can be performed for a number of different games. For example, game log records related to a plurality of games may be collected to generate a game log table 121, and game meta information of a plurality of games may be collected to generate a game meta information table 122. , The event history table 123 may be generated by collecting the history of events executed on a plurality of games. Game meta information may also be referred to as game information.

Information in the game DB 112 or each table 121, 122, 123 may be input, managed, and updated by the game operator. For example, when a game operator is registered for each game, game meta information such as genre information or operation information about each game, event history such as event type or event period executed for each game, server of the game DB 112 You can enter the IP (internet protocol), the name of each table, the schema of each table, and the start date of each game.

The descriptor table 140 and the event effect table 150 may be generated through a certain process of processing the game log, game meta information, and event history. For example, the temporary aggregate table 131 may be generated through an aggregate query operation regarding game logs in the game log table 121. In addition, the temporary non-aggregation table 132 may be generated through a non-aggregate query operation on the game meta information in the game meta information table 122 and the event history in the event history table 123. Thereafter, the descriptor table 140 and the event effect table 150 may be generated based on a table join query operation on the temporary aggregate table 131 and the temporary non-aggregate table 132.

The descriptor table 140 may include a game state index descriptor, a product life cycle (PLC) index descriptor, a game information descriptor, and an event information descriptor. Each descriptor in the descriptor table 140 may be used for learning the index prediction model. Each descriptor in the descriptor table 140 is based on a collected value or a measured value, which may be distinguished from a predicted value.

The game state indicator descriptor refers to data representing the state of the game, which is continuously measured in relation to the game itself or users in the game. In order to cover a plurality of games, a game state indicator descriptor may be generated by constructing a vector describing the game state by using a field common to all games and operations related thereto. For example, a relationship between a common field constituting the game state indicator descriptor and an individual field of each game may be expressed as Equation 1 below.

Various aggregation operations can be used to construct a vector through data in a common field. The operation used here may be defined in advance. For example, the sum, average, and variance of each field may be used, and the aggregation operation may be performed based on all users including non-access users. The game state indicator descriptor can be expressed as Equation 2 below.

In Equation 2, S denotes a game state indicator descriptor.

The PLC indicator descriptor can consist of vectors representing sales and daily active users (DAU). For example, the PLC indicator may include DAU, total play time, average play time, total revenue, average revenue per user (ARPU), average revenue per paid user (ARPPU), and the like. The PLC index descriptor can be expressed as Equation 3 below.

In Equation 3, P represents a PLC index descriptor.

The event information descriptor is information describing an event and may indicate the type of a game event. For example, game events may include rewards and promotions. In addition, compensation may include addition of fixed-term compensation for existing content, addition of new content to which fixed-term compensation is applied, and access compensation for new users or returning users, and promotions may include products (e.g., item , Character, etc.). The event information descriptor can be represented by E.

As will be described again below, according to an embodiment, detailed event information that is distinguished from the event information may be provided. Event detailed information may include specific event content compared to event information. For example, if the type of event is specified according to the event information, the event detailed information is the specific execution history of the event (e.g., the type of reward added to the existing content, the type of reward content newly added, the degree of reward, promotion The type of product to which this is applied, the degree of discount, etc.) may be included.

The game information descriptor generalizes the type (genre) and operation information of the game to respond to various games as a system, and embodiments can be applied to a new game in which all data does not exist through the generalized system and model. The game information descriptor may include a genre information descriptor and an operation information descriptor. The game information descriptor can be expressed as Equation 4 below.

In Equation 4, G denotes a game information descriptor, G _k denotes a genre information descriptor, and G _o denotes an operation information descriptor.

The genre information descriptor may be composed of vectors representing genre information, and each vector may be set to have a true value for a corresponding genre with each genre as one dimension. For example, the genres are RPG (role playing game), arcade, puzzle, casual, collectible, action, sports, shooting, massively multiplayer online game (MMO), multiplayer online battle arena (MOBA), and gambling (go-stop, poker). And the like. For example, a specific game may be expressed as a vector such as [1 0 0 0 0 1 0 0 1 0 0 0 ...]. Genre information can be expressed as Equation 5 below.

The operation information descriptor may include operation country information, operation period information, and game update information. The operation information descriptor can be expressed as Equation 6 below.

In Equation 6, G _c represents operating country information, G _d represents operating period information, and G _u represents game update information. The operating country information may be composed of a one-hot vector having a dimension as many as the number of serviceable countries. The operating country information can be expressed as Equation 7 below.

The operating period information may indicate a day elapsed from the start of the game service. The initiation of a game service can also include cases with large-scale updates. The operating period information can be expressed as Equation 8 below.

The game update information is update information for each game, and may indicate a total number of updates, an elapsed date from the latest update date, a change size according to the latest update, a change size according to the total update, and the like. For example, if the version number is changed from 1.0.2 to 2.0.0, the size of the recent update change can be expressed as [0 9 8], and if the version number is 2.0.0, the total update change size is [2 0] It can be expressed as 0]. The game update information can be expressed as Equation 9 below.

The descriptor table 140 may store each descriptor in the form of (S ^t+1 , P ^t+1 , S ^t , P ^t , E, G). Here, S denotes a game state index descriptor, P denotes a PLC index descriptor, E denotes an event information descriptor, and G denotes a game information descriptor. Game status indicators and PLC indicators can constitute game operation indicators. The game operation index may be time dependent information, and the event information descriptor and the game information descriptor may be time independent information. Accordingly, each game operation index according to time may be identified as a time step t, and the descriptor table 140 may load each game operation index according to time in the order of time elapsed, such as t and t+1. The descriptor table 140 may match and store each descriptor.

The event effect table 150 may store detailed information of events performed on each game in the past and effects accordingly. The event effect table 150 may be referred to to present an event execution strategy when recommending an event. For example, an optimal event may be selected based on a game operation index predicted using an index prediction model, and a specific event execution strategy may be determined by accessing the event effect table 150 with the optimal event. A specific event execution strategy may be determined based on detailed event information.

The event effect table 150 may collect, process, measure and store PLC index, change amount of PLC index, game information, and event information. For example, the event effect table 150 may store each information in a form such as (P ^t+1 , P ^t+1 -P ^t , E, G, event specification). Here, P represents the PLC index. P ^t+1 represents the PLC index measured based on the time point t+1, and P ^t+1 -P ^t represents the difference between the PLC index measured based on the time t+1 and the PLC index measured based on the time t. In other words, the latter can cope with the amount of change in the PLC index. E represents an event information descriptor, G represents a game information descriptor, and an event specification represents event detailed information. The event specification can have a text format. The event effect table 150 may match and store each piece of information. For example, the event effect table 150 may store each information in a tuple form.

Operations related to the game event guide according to the embodiments may be performed by a game operating device (not shown, described later), and the game DB 112 or each table of FIG. 1, information, etc. are stored in the game operation server. Can be.

2 is a diagram illustrating a learning process of an index prediction model according to an embodiment. Referring to FIG. 2, when the descriptor table 210 is generated, training data 220 may be extracted from the descriptor table 210, and the index prediction model 230 may be trained through the training data 220. . The index prediction model 230 may be an artificial neural network (ANN), for example, a recurrent neural network (RNN) that considers the passage of time.

The index prediction model 230 may be trained to predict game operation indexes of the target section based on the game operation index, event information, and game information at the first time point. As described above, since the game operation index may be time-dependent information, it may be classified according to the time point, and may include a game state index and a PLC index.

When the game operation indicator includes a game state indicator and a PLC indicator, the indicator prediction model 230 may include a game state indicator prediction model for predicting the game state indicator and a PLC indicator prediction model for predicting the PLC indicator. . When the learning is completed, the game state indicator prediction model can predict the game state indicators of the target time interval based on the game information, event information, and the game state indicator at the first time point, and the PLC indicator prediction model is game information, event information. , And PLC indicators of the target time interval may be predicted based on the PLC indicators of the first time point.

Training data for learning the game state index prediction model extracted from the descriptor table 210 may be expressed as Equation 10 below, and the game state index prediction model may be modeled according to Equation 11 below.

The index prediction model 230 may be trained to minimize an error between a predicted value and an actual value using tuple data according to Equation 10. According to Equation 11, a game state index at time t+1 is predicted based on a game state index at time t, a PLC index at time t, event information, and game information. The game state indicator can express indices for representing the game state as a multi-dimension, and i corresponds to an index for classifying each dimension. The operation is repeated for each dimension related to the game state, so that prediction about the entire dimension related to the game state can be performed.

As the prediction is repeated, a certain target time interval or an index related to a certain target time point may be predicted. For example, the game state indicator at the time t+1 may be predicted based on the game state indicator at the time t and the PLC indicator, and the game state indicator at the predicted time t+1, and the PLC indicator t+ The game state indicator at point 2 can be predicted.

Meanwhile, training data for learning the PLC index prediction model extracted from the descriptor table 210 may be expressed as Equation 12 below, and the PLC index prediction model may be modeled according to Equation 13 below.

The PLC index can express the indexes for indicating the PLC as multi-dimensions, and i corresponds to an index for classifying each dimension. In addition, the above-described game state index prediction model may be applied correspondingly to the PLC index prediction model.

3 is a diagram illustrating a process of setting a global parameter according to an embodiment. In the index prediction model, event information and game information may correspond to global conditions that are not affected by each time step. As illustrated in FIG. 3, by modeling the relationship between events, and modeling the relationship between the operating countries, the relationship between the operating period, and the game update information, finally the relationship between the expressions can be modeled. The representation can be modeled through a fully connected layer, a convolution layer, a gate layer, and the like. The global condition (GL) for event information and game information may be expressed as Equation 14 below.

The expressed global condition GL may be used in the form of vector concatenation at all time steps as shown in Equation 15 below.

4 is a diagram illustrating an indicator prediction model according to an embodiment. According to an embodiment, the index prediction model may be implemented as an autoregressive model as illustrated in FIG. 4. A node receiving input data [GL, P ^t-2 , S ^t-2 ] can output output data y _t-1 , and input data corresponding to output data y _t-1 [GL, P ^t-1] , S ^t-1 ] can be input to the next consecutive node. If the game state index prediction models and PLC index prediction models that are implemented separately, the output data may be a s _t corresponding to the game status indicator, or may be a s _t corresponding to the PLC indicators. For the next time step, input data and output data may be determined through self-regressive calculations.

), 관심 게임에 관해 수행될 수 있는 이벤트 정보(

), 및 관심 게임에 관해 측정된 t 시점의 게임 운영 지표(

,

)가 입력됨에 응답하여, 관심 게임에 관한 타겟 시간 구간의 게임 운영 지표들(

,

)을 출력한다. 관심 게임은 현재 서비스 중인 다수의 게임들 중에 이벤트 수행에 관한 가이드 정보를 제공하고자 하는 대상을 나타낸다. t 시점에 관한 지표의 입력에 따라 t+1 시점의 지표가 예측될 수 있고, 예측된 t+1 시점의 지표에 따라 t+2 시점의 지표가 예측될 수 있다. 단위는 하루에 해당할 수 있다. 이와 같은 과정의 반복을 통해 타겟 시간 구간(t+1 내지 k)에 해당하는 지표들이 예측될 수 있다.5 is a diagram illustrating an event simulation process according to an embodiment. 5, the index prediction model 510 has completed learning, and game information about the game of interest (

), event information that can be performed on the game of interest (

), and the game operation index at point t measured for the game of interest (

,

In response to the input of ), the game operation indicators (

,

) Is displayed. The game of interest represents a target for providing guide information on event performance among a number of games currently in service. The index of time t+1 may be predicted according to the input of the index of time t, and the index of time t+2 may be predicted according to the index of time t+1. Units can correspond to one day. Through repetition of this process, indices corresponding to the target time interval (t+1 to k) may be predicted.

) 및 이벤트 정보(

)가 확정된 상태에서, 특정 시점에 관한 게임 운영 지표가 주어지면, 지표 예측 모델(510)에 의해 주어진 게임 정보(

)에 대응하는 게임(예:관심 게임)에 관해 주어진 이벤트 정보(

)에 따른 이벤트가 수행되었을 때 타겟 시간 구간의 게임 운영 지표가 예측될 수 있다. 예를 들어, 사용자에 의해 관심 게임, 수행할 이벤트, 예측하려는 시간 구간(타겟 시간 구간)이 입력되면, 게임 운영 서버는 특정 시점 t(예: 현재 시점)을 기준으로 입력된 정보를 바탕으로 입력 데이터(

,

,

,

)를 구성하여 지표 예측 모델(510)에 입력할 수 있고, 지표 예측 모델(510)의 출력을 통해 타겟 시간 구간의 게임 운영 지표들(

,

)을 예측할 수 있다.In other words, game information (

) And event information (

) Is determined, when a game operation index for a specific time point is given, the game information given by the index prediction model 510 (

Event information (for example, the game of interest) corresponding to) (

When the event according to) is performed, the game operation index of the target time interval can be predicted. For example, when a game of interest, an event to be performed, and a time interval to be predicted (target time interval) are input by the user, the game operation server inputs the information based on the input information based on a specific time point t (eg, the current time point) data(

,

,

,

) Can be configured and input to the indicator prediction model 510, and game operation indicators in the target time interval through the output of the indicator prediction model 510 (

,

) Can be predicted.

,

)은 주어진 이벤트를 수행함에 따른 이벤트 효과를 나타낼 수 있다. 따라서, 이러한 이벤트 시뮬레이션을 통해 주어진 이벤트를 수행함에 따른 미래의 이벤트 효과가 예측될 수 있다. 예를 들어, 도 6은 일 실시예에 따른 이벤트 시뮬레이션 결과를 나타낸 도면이며, 도 6의 그래프(600)와 같이 타임 스텝(시점) 별로 게임 운영 지표들(S, P)이 예측될 수 있고, 예측된 게임 운영 지표들(S, P)을 통해 이벤트 효과가 분석될 수 있다.Predicted game operation indicators (

,

) May indicate an event effect according to the execution of a given event. Therefore, the future event effect can be predicted by performing a given event through such event simulation. For example, FIG. 6 is a diagram showing an event simulation result according to an embodiment, and game operation indicators S and P may be predicted for each time step (time point) as in the graph 600 of FIG. 6, The effect of the event may be analyzed through the predicted game operation indicators (S, P).

7 is a diagram illustrating a process of selecting an optimal event according to an embodiment. The event simulation as illustrated in FIG. 5 may be performed for various games and various events, and accordingly, an appropriate event may be recommended. 7 is for explaining such a series of processes.

), 각 게임에 관해 수행될 수 있는 이벤트 정보(

), 및 각 게임에 관해 측정된 t 시점의 게임 운영 지표(

,

)가 입력됨에 응답하여, 관심 게임에 관한 타겟 시간 구간의 게임 운영 지표들(

,

)을 출력한다. 지표 예측 모델(710)의 출력은 이벤트 추천을 위해 테이블과 같은 형식으로 저장될 수 있고, 해당 테이블을 참조하여 최적의 이벤트(720)가 선정될 수 있다.Referring to FIG. 7, the index prediction model 710 provides game information for each game (

), event information that can be performed for each game (

), and the game operation index at point t measured for each game (

,

In response to the input of ), the game operation indicators (

,

) Is displayed. The output of the index prediction model 710 may be stored in a table-like format for event recommendation, and an optimal event 720 may be selected by referring to the corresponding table.

For example, the table may include game operation indicators predicted during the target section with respect to the case where each event is performed for each game, and each time step (time point) of each game in consideration of the predicted game operation indicators. An optimal event 720 for can be selected. For example, an event corresponding to each event information corresponding to the maximum value of the game operation indicators in the target time interval may be selected as the optimal event 720. Since the PLC indicator can reflect the effect of the event well, the value of the PLC indicator (eg, the maximum value) can be used when selecting the optimal event 720.

In this case, an optimal event 720 may be selected for each time step of each game. For example, the first event is selected as the optimal event 720 for the first time step of the first game, and the second event is selected as the optimal event 720 for the second time step of the first game. Can be.

This series of processes for event recommendation can be represented as shown in Table 1 below.

,

)가 예측된다. 게임 운영 지표(

,

)의 각 디멘전에 관한 연산을 통해 최종적인 게임 운영 지표(

,

)가 예측될 수 있다.According to Table 1, the variable Games represents each game (game information), and the variable game represents any one of them (game information). In FIG. 7,'x' is used as an identifier for classifying games. (1, k) represents the target time interval. The variable Events represents each event (event information), and the variable event represents any one of these events (event information). In Fig. 7,'y' is used as an identifier for classifying events. In a state in which a game is specified as a variable game and a time step is specified as t, a game operation index (

,

) Is predicted. Game operation indicator (

,

) Through the calculation of each dimension of the final game operation index (

,

) Can be predicted.

,

)가 예측되면, 관한 게임 운영 지표를 고려하여 각 이벤트 중 최적의 이벤트(720)에 해당하는 것이 변수 Bestevent에 추가될 수 있다. PLC 지표가 이벤트 효과를 적절히 반영할 수 있으므로,

를 고려하여 최적의 이벤트(720)가 결정될 수 있다. 예를 들어, 각 이벤트에 관해

의 각 디멘전의 합이나 평균을 통해

의 대표 값이 결정될 수 있고, 각 이벤트에 관해 결정된 대표 값들 중 값이 최대인 것에 대응하는 이벤트가 최적의 이벤트(720)로 결정될 수 있다. 이러한 대표 값은 아래에서 간단히 최대 값으로 표현될 수도 있다. 표 1의 마지막 두 열을 통해 계산된

,

는 다음 스텝의 예측 연산에 사용될 수 있다.Game operation indicator (

,

) Is predicted, one corresponding to the optimal event 720 among each event may be added to the variable Bestevent in consideration of the related game operation index. Since PLC indicators can properly reflect event effects,

In consideration of the optimal event 720 may be determined. For example, for each event

Through the sum or average of each dimension of

A representative value of may be determined, and an event corresponding to the largest value among representative values determined for each event may be determined as the optimal event 720. These representative values may be simply expressed as a maximum value below. Calculated over the last two columns in Table 1

,

Can be used in the next step's prediction operation.

Through the above-described process, an optimal event 720 for each time system may be selected for each game, and the selected optimal event 720 may be included in the guide information and provided to the user. For example, a plurality of optimal events selected for each time system may be provided in the form of a list as guide information, and in this case, effects (eg, predicted PLC indicator values) are provided in a form arranged in the highest order. It is also possible.

On the other hand, if the game operation index is predicted, the future game operation state may be predicted through the predicted game operation index. For example, if a game operation state may reach a risk level at a specific point in the future, a risk warning regarding game operation may be provided in advance as guide information. According to an embodiment, when game operation indicators are predicted for a target time interval of a game of interest, the predicted game operation indicators and a first threshold are compared, or a change amount of the predicted game operation indicators and a second threshold are compared. I can. Also, based on the comparison result, a game operation state of the game of interest may be determined, and a risk warning regarding the game operation of the game of interest may be provided based on the determined game operation state.

8 is a diagram illustrating an event effect table according to an embodiment. When an optimal event is selected through the process of FIG. 7, an event execution strategy for each optimal event selected with reference to the event effect table 800 of FIG. 8 may be generated.

), PLC 지표의 변화량(

), 게임 정보(

), 이벤트 정보(

), 및 이벤트 상세 정보(

)를 저장한다. 여기서, 'x'는 게임(게임 정보)을 구분하고, 'y'는 이벤트(이벤트 정보)를 구분한다. 이벤트 효과 테이블(800)은 기 수행된 이벤트 결과를 저장하며, 도 1의 이벤트 효과 테이블(150)에 대응할 수 있다. 이벤트 효과 테이블(800)의 작성을 위해 참조된 게임은 샘플 게임으로 지칭될 수 있다.8, the event effect table 800 is a PLC indicator (

), PLC index change amount (

), game information (

), event information (

), and event details (

). Here,'x' identifies a game (game information), and'y' identifies an event (event information). The event effect table 800 stores a result of a previously performed event and may correspond to the event effect table 150 of FIG. 1. The game referenced for the creation of the event effect table 800 may be referred to as a sample game.

에 대응하고, 최적의 이벤트가

에 대응하는 것을 가정한다. 이 경우, 게임 정보(

) 및 이벤트 정보(

)에 대응하는 PLC 지표(

), PLC 지표의 변화량(

), 및 이벤트 상세 정보(

)가 검색될 수 있다.When the optimal event is selected, the event effect table 800 is accessed with the specific game (hereinafter, the game of interest) and each optimal event selected for the game of interest, and corresponding to the game of interest and each selected optimal event. Information can be searched. For example, the game of interest

And the optimal event

Assume that it corresponds to. In this case, game information (

) And event information (

PLC index corresponding to) (

), PLC index change amount (

), and event details (

) Can be searched.

)는 관심 게임에 대응하는 게임 정보를 갖는 참조 게임들 및 참조 게임들에 관해 수행된 최적의 이벤트에 대응하는 이벤트들에 관한 이벤트 수행 이력을 포함할 수 있다. 따라서, 검색된 이벤트 상세 정보(

)에 포함된 과거 이벤트 수행 이력을 통해 이벤트 수행 전략이 생성 및 제공될 수 있다.Event details (

) May include reference games having game information corresponding to the game of interest, and event execution history regarding events corresponding to optimal events performed on the reference games. Therefore, the searched event details (

An event execution strategy may be created and provided through the past event execution history included in ).

) 및 PLC 지표의 변화량(

)은 이벤트 정보(

)에 따른 최적의 이벤트의 효과를 나타낼 수 있고, 그 효과가 최대가 되었을 때 수행되었던 구체적인 방안들이 이벤트 상세 정보로서 제공될 수 있다. 예를 들어, PLC 지표(

) 중 최대 값에 대응하는 이벤트 상세 정보(

) 또는 PLC 지표의 변화량(

) 중 최대 값에 대응하는 이벤트 상세 정보(

)에 기초하여 이벤트 수행 전략이 생성될 수 있다. 이러한 이벤트 수행 전략은 이벤트 수행에 관한 가이드 정보로서 사용자에게 제공될 수 있다. 실시예에 따라, 가이드 정보는 최적의 이벤트에 관한 정보, 이벤트 수행 전략에 관한 정보, 및 위험 경고 중 적어도 하나를 포함할 수 있다.Also, the searched PLC indicator (

) And the amount of change in PLC indicators (

) Is the event information (

), the effect of the optimal event may be indicated, and specific measures performed when the effect was maximized may be provided as event detail information. For example, PLC indicators (

) Of the event details corresponding to the maximum value (

) Or the amount of change in PLC indicators (

) Of the event details corresponding to the maximum value (

) Based on the event execution strategy may be generated. This event execution strategy may be provided to a user as guide information on event execution. Depending on the embodiment, the guide information may include at least one of information on an optimal event, information on an event execution strategy, and a risk warning.

For example, assuming that the game of interest is a CCG (collectible card game) genre in service with an operating period of 1 year, the operating period is among the past event performance information of other sample games of the CCG game genre with an operating period of 3 years. Events that were effective when they were about a year old and specific event execution strategies at that time can be recommended. In addition, in the case of a new puzzle game, game indicators can be predicted using information collected through the closed beta service, and event simulations can be performed using this information to receive recommendations for events related to the influx of promising new users. Even in situations where there is not much information as it is a new game, it is possible to simulate an event through the game information of the new game (eg, genre information, operation information) and search the event history for other similar games. It is also possible to recommend events according to the embodiment.

9 is a diagram showing a game operating device according to an embodiment. Referring to FIG. 9, the game operating device 900 includes a processor 910 and a memory 920. The memory 920 may store instructions that can be read by the processor 910, and when the instructions stored in the memory 920 are executed by the processor 910, the processor 910 operates on a game event guide according to the embodiment. Can perform. For example, the processor 910 inputs game information about a game of interest, event information that can be performed on a game of interest, and a game operation index measured at a first point in time about the game of interest into a pre-learned index prediction model. Thus, it is possible to predict the game operation indicators of the target time interval related to the game of interest, and to select the optimal events for the game of interest for each time point within the target time interval by considering the game operation indexes of the predicted target time interval. . In addition, the processor 910 may generate an event effect table for storing detailed information on each event of each game information on the sample games, information on each event performed on the sample games, and information on each event performed on the sample games. With reference, an event execution strategy for each of the optimal events is generated, and guide information on event execution for the game of interest is provided based on the selected optimal events and the event execution strategy for each of the selected events. can do. In addition, with respect to the game operating device 900, the contents described with reference to FIGS. 1 to 8 and 10 may be applied.

10 is a diagram illustrating a game operation guide method according to an embodiment. Referring to FIG. 10, the game management guide method includes predicting game operation indicators in a target time interval using an indicator prediction model (1010), and selecting an optimal event in consideration of the predicted game operation indicators (1020). , Generating (1030) an event execution strategy with reference to the event effect table, and providing (1040) guide information on event execution. In addition, the content described through FIGS. 1 to 9 may be applied to the game operation guide method.

The apparatus described above may be implemented as a hardware component, a software component, and/or a combination of a hardware component and a software component. For example, the devices and components described in the embodiments are, for example, a processor, a controller, an Arithmetic Logic Unit (ALU), a digital signal processor, a microcomputer, a Field Programmable Gate Array (FPGA). , PLU (Programmable Logic Unit), a microprocessor, or any other device capable of executing and responding to instructions, it may be implemented using one or more general purpose computers or special purpose computers. The processing device may execute an operating system (OS) and one or more software applications executed on the operating system. In addition, the processing device may access, store, manipulate, process, and generate data in response to the execution of software. For the convenience of understanding, although it is sometimes described that one processing device is used, one of ordinary skill in the art, the processing device is a plurality of processing elements and/or a plurality of types of processing elements. It can be seen that it may include. For example, the processing device may include a plurality of processors or one processor and one controller. In addition, other processing configurations are possible, such as a parallel processor.

The software may include a computer program, code, instructions, or a combination of one or more of these, configuring the processing unit to behave as desired or processed independently or collectively. You can command the device. Software and/or data may be interpreted by a processing device or to provide instructions or data to a processing device, of any type of machine, component, physical device, virtual equipment, computer storage medium or device. , Or may be permanently or temporarily embodyed in a transmitted signal wave. The software may be distributed over networked computer systems and 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 in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, and the like alone or in combination. The program instructions recorded on the medium may be specially designed and configured for the embodiment, or may be known and usable to those skilled in 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. -A hardware device specially configured to store and execute program instructions such as magneto-optical media, and ROM, RAM, flash memory, and the like. Examples of program instructions include not only machine language codes such as those produced by a compiler but also high-level language codes that can be executed by a computer using an interpreter or the like. The hardware device described above may be configured to operate as one or more software modules to perform the operation of the embodiment, and vice versa.

As described above, although the embodiments have been described by the limited embodiments and drawings, various modifications and variations are possible from the above description by those of ordinary skill in the art. For example, the described techniques are performed in a different order from the described method, and/or components such as a system, structure, device, circuit, etc. described are combined or combined in a form different from the described method, or other components Alternatively, even if substituted or substituted by an equivalent, an appropriate result can be achieved.

Therefore, other implementations, other embodiments, and claims and equivalents fall within the scope of the claims to be described later.

Claims (17)

In the game event guide method performed by the game operating device,
Game information on the game of interest, event information that can be performed on the game of interest, and the game operation index of the first time point measured on the game of interest are input into a pre-learned index prediction model, Predicting game operation indicators in the target time interval;
Selecting optimal events for the game of interest for each time point in the target time interval, taking into account the game operation indicators of the predicted target time interval;
The selection is made by referring to an event effect table that stores detailed information on each game of the sample games, information on each event performed on the sample games, and detailed information on each event performed on the sample games. Generating an event execution strategy for each of the optimized events; And
Providing guide information on event execution for the game of interest based on the selected optimal events and event execution strategies for each of the selected events
Including,
The event effect table further stores game operation indicators measured for the sample games,
Generating the event execution strategy
The step of generating the event execution strategy based on event detail information corresponding to a maximum value of the game management indicators measured for the sample games or the maximum value of the increase amount of the game management indicators measured for the sample games. Included,
How to guide game events.
The method of claim 1,
The step of selecting the optimal events
Including the step of selecting the optimal events based on each event information corresponding to the maximum value of the game operation index of the predicted target time interval,
How to guide game events.
delete The method of claim 1,
Performed after the step of predicting the game operation indicators,
Comparing the predicted game operation indicators of the target time interval with a first threshold value, or comparing the predicted change amount of the game operation indicators of the target time interval with a second threshold value;
Determining a state of operation of the game of the game of interest based on the comparison result; And
Providing a risk warning about the game operation of the game of interest based on the determined state of the game operation
Including further, the game event guide method.
The method of claim 1,
The target time interval is
A second time point after the first time point, and a third time point after the second time point,
The indicator prediction model is
Predicting the game operation index at the second viewpoint based on the game information regarding the game of interest, the event information that can be performed on the game of interest, and the measured game operation index at the first viewpoint,
Predicting a game operation index at the third time point based on the game information regarding the game of interest, the event information that may be performed on the game of interest, and the predicted game operation index at the second time point,
How to guide game events.
The method of claim 1,
The above game operation indicator is
Including a game state indicator and a PLC (product life cycle) indicator,
How to guide game events.
The method of claim 6,
The indicator prediction model is
Based on the game information about the game of interest, the event information that can be performed on the game of interest, and the game state index of the first time point measured with respect to the game of interest, a game state index of the target time interval A game state indicator prediction model that predicts them, and
Based on the game information related to the game of interest, the event information that can be performed on the game of interest, and the PLC index of the first time point measured for the game of interest, the PLC indexes of the target time interval are predicted PLC indicator prediction model
Including, the game event guide method.
The method of claim 1,
The above game information
Including genre information and operational information,
How to guide game events.
The method of claim 8,
The above operation information is
Including operating country information, operating period information, and game update information,
How to guide game events.
The method of claim 1,
Performed before the step of predicting the game operation indicators,
Collecting game log information, game meta information, and event history;
Generating a descriptor table by processing the collected game log information, the game meta information, and the event history; And
Training the index prediction model using the descriptor table
Including further, the game event guide method.
A computer-readable storage medium storing one or more programs including instructions for performing the method of any one of claims 1, 2, and 4 to 10.
In the game operating device,
Processor; And
Memory for storing instructions readable by the processor
Including, and when the instructions are executed in the processor, the processor
Game information on the game of interest, event information that can be performed on the game of interest, and the game operation index of the first time point measured on the game of interest are input into a pre-learned index prediction model, Predicts game operation indicators in the target time interval,
In consideration of the game operation indicators of the predicted target time period, optimal events for the game of interest are selected for each time point within the target time period,
The selection is made by referring to an event effect table that stores detailed information on each game of the sample games, information on each event performed on the sample games, and detailed information on each event performed on the sample games. Generating an event execution strategy for each of the optimal events,
Based on the selected optimal events and an event execution strategy for each of the selected events, guide information on event execution for the game of interest is provided,
The event effect table further stores game operation indicators measured for the sample games,
The processor generates the event execution strategy based on event detail information corresponding to a maximum value of game operation indicators measured for the sample games or a maximum value of an increase amount of game operation indicators measured for the sample games. doing,
Game operating device.
The method of claim 12,
The processor is
Selecting the optimal events based on each event information corresponding to the maximum value of the game operation indicators in the predicted target time interval,
Game operating device.
delete The method of claim 12,
The processor is
Comparing the predicted game operation indices of the target time interval with a first threshold, or comparing a change amount of the game operation indices of the predicted target time interval with a second threshold,
Determining a state of the game operation of the game of interest based on the comparison result,
Providing a risk warning related to the game operation of the game of interest based on the determined state of the game operation,
Game operating device.
The method of claim 12,
The above game operation indicator is
Including game state indicators and PLC (product life cycle) indicators,
The indicator prediction model is
Based on the game information about the game of interest, the event information that can be performed on the game of interest, and the game state index of the first time point measured with respect to the game of interest, a game state index of the target time interval A game state indicator prediction model that predicts them, and
Based on the game information related to the game of interest, the event information that can be performed on the game of interest, and the PLC index of the first time point measured for the game of interest, the PLC indexes of the target time interval are predicted PLC indicator prediction model
Containing, game operating device.
The method of claim 12,
The processor is
Collect game log information, game meta information, and event history,
The collected game log information, the game meta information, and the event history are processed to generate a descriptor table,
Training the index prediction model using the descriptor table,
Game operating device.

Images