Information processing method and device for turn game, electronic equipment and storage medium

Abstract

The invention provides an information processing method and device for turn games, electronic equipment and a storage medium, and relates to the technical field of games. Comprising the following steps: acquiring state data of each game unit under a target analysis granularity in a to-be-replayed game; according to the state data of each game unit under the target analysis granularity, determining a situation value of the target combat party under the target analysis granularity; and according to the situation value of the target combat party under the target analysis granularity, determining the situation difference of the target combat party or the situation difference of a non-target combat party. The concept of a situation difference is introduced into a replica of a turn-system game, situation values of two combat parties are analyzed and determined, the situation difference of the two combat parties is determined based on the situation values, and through the situation difference, players are helped to more visually analyze the influence of action strategies under analysis granularity on game winning and losing of the players. And more refined and more basis game replaying is realized, and players are helped to optimize subsequent game strategies.

Description

Technical Field

[0001] The present application relates to the field of game technology, and in particular to an information processing method, device, electronic device, and storage medium for a turn-based game. Background Art

[0002] In many turn-based combat games, players control their characters in turn-based combat scenes, gradually reducing the enemy's health through attacks and the use of skills. When all enemy characters' health is reduced to 0, the player wins; conversely, if all their own characters' health is reduced to 0, the player loses. Players must carefully plan their own character's actions and attack targets based on the enemy's characteristics, skills, and positions. Consequently, the final outcome of turn-based games often fluctuates due to randomness or split-second decision-making.

[0003] The game replay function can help players review game play, but existing replays for turn-based games only display information about changes in various status data of game units after they take action. The data display is relatively discrete and cannot effectively integrate the impact of various status data on the final outcome, resulting in a poor game replay experience for players. Summary of the Invention

[0004] The purpose of this application is to provide an information processing method, device, electronic device and storage medium for turn-based games, so as to enhance the player's game review experience and assist the player in performing more refined game review.

[0005] To achieve the above objectives, the technical solutions adopted in the embodiments of the present application are as follows:

[0006] In a first aspect, an embodiment of the present application provides an information processing method for a turn-based game, which is applied to a turn-based game, wherein the game includes a first combatant and a second combatant, each of which includes at least one game unit; the method includes:

[0007] Obtaining status data of each game unit at a target analysis granularity in the game to be reviewed; the target analysis granularity includes: a single round as the minimum analysis unit or a single game unit action as the minimum analysis unit;

[0008] Determine, based on the status data of each game unit at the target analysis granularity, a situation value of a target combatant at the target analysis granularity, the target combatant being the first combatant or the second combatant;

[0009] According to the situation value of the target combatant at the target analysis granularity, the situation difference of the target combatant or the situation difference of the non-target combatant is determined, wherein the situation difference of the target combatant is used to characterize the winning and losing trend of the target combatant at the target analysis granularity.

[0010] In a second aspect, an embodiment of the present application further provides an information processing device for a turn-based game, which is applied to a turn-based game, wherein the game includes a first combatant and a second combatant, each of the first combatant and the second combatant includes at least one game unit, and the device includes: an acquisition module and a determination module;

[0011] The acquisition module is used to obtain the status data of each game unit at the target analysis granularity in the game to be reviewed; the target analysis granularity includes: a single round as the minimum analysis unit or a single game unit action as the minimum analysis unit;

[0012] The determination module is configured to determine, based on the status data of each game unit at the target analysis granularity, a situation value of a target combatant at the target analysis granularity, wherein the target combatant is the first combatant or the second combatant;

[0013] The determination module is used to determine the situation difference of the target combatant or the situation difference of the non-target combatant based on the situation value of the target combatant at the target analysis granularity, wherein the situation difference of the target combatant is used to characterize the winning and losing trend of the target combatant at the target analysis granularity.

[0014] In a third aspect, an embodiment of the present application provides an electronic device comprising: a processor, a storage medium and a bus, wherein the storage medium stores machine-readable instructions executable by the processor. When the electronic device is running, the processor and the storage medium communicate through the bus, and the processor executes the machine-readable instructions to execute the information processing method of the turn-based game provided in the first aspect.

[0015] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the information processing method for a turn-based game as provided in the first aspect is executed.

[0016] The beneficial effects of this application are:

[0017] The present application provides an information processing method, device, electronic device, and storage medium for a turn-based game, including: obtaining status data of each game unit at a target analysis granularity in a game to be reviewed; determining the situation value of the target combatant at the target analysis granularity based on the status data of each game unit at the target analysis granularity; and determining the situation difference of the target combatant or the situation difference of the non-target combatant based on the situation value of the target combatant at the target analysis granularity. The method introduces the concept of situation difference into the review of a turn-based game, uses a set action node as the analysis granularity, obtains the status data of each game unit of both combatants at the analysis granularity, analyzes and determines the situation value from the perspective of the target combatant, and finally determines the situation difference from the perspective of the target combatant or the situation difference from the perspective of the non-target combatant based on the situation value from the perspective of the target combatant. The situation difference helps players more intuitively analyze the impact of their action strategies at the analysis granularity on their game wins and losses, thereby achieving more refined and more reliable game review and helping players optimize subsequent game strategies.

[0018] Among them, for low-frequency state data, the first prediction model is independently trained to process the low-frequency state data separately, and the processing result of the low-frequency state data, that is, the first prediction information, is used instead of the low-frequency state data to predict the final situation value, so that the influence of the low-frequency state data on the final prediction result of the model can be retained, and the direct use of the low-frequency state data can be avoided to cause the model to overfit or misjudgment, thereby improving the accuracy of the prediction results of the second prediction model.

[0019] By converting the situation difference into a situation diagram and visually displaying the game's win-loss trend through the situation diagram, players can intuitively view the win-loss trend at each node and intuitively identify some key nodes that have a key impact on the final outcome of the game. The game data of the key nodes can be reviewed in a targeted manner to assist in formulating more reasonable game strategies in subsequent games. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without creative work.

[0021] Figure 1 A schematic flow chart of an information processing method for a turn-based game provided in an embodiment of the present application;

[0022] Figure 2 A flowchart of another information processing method for a turn-based game provided in an embodiment of the present application;

[0023] Figure 3 A flowchart of another information processing method for a turn-based game provided in an embodiment of the present application;

[0024] Figure 4 A flowchart of another information processing method for a turn-based game provided in an embodiment of the present application;

[0025] Figure 5 A flowchart of a model training method provided in an embodiment of the present application;

[0026] Figure 6 A flowchart of another model training method provided in an embodiment of the present application;

[0027] Figure 7 A flowchart of another model training method provided in an embodiment of the present application;

[0028] Figure 8 A flowchart of another model training method provided in an embodiment of the present application;

[0029] Figure 9 A flowchart of another information processing method for a turn-based game provided in an embodiment of the present application;

[0030] Figure 10 A schematic diagram of a game situation review interface provided in an embodiment of the present application;

[0031] Figure 11 A schematic diagram of an information processing device for a turn-based game provided in an embodiment of the present application;

[0032] Figure 12 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0033] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. It should be understood that the drawings in the present application only serve the purpose of illustration and description and are not used to limit the scope of protection of the present application. In addition, it should be understood that the schematic drawings are not drawn to scale. The flowcharts used in this application illustrate the operations implemented according to some embodiments of the present application. It should be understood that the operations of the flowcharts can be implemented out of sequence, and steps without logical context can be reversed or implemented simultaneously. In addition, those skilled in the art, under the guidance of the contents of this application, can add one or more other operations to the flowchart, or remove one or more operations from the flowchart.

[0034] In addition, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. The components of the embodiments of the present application generally described and shown in the drawings here can be arranged and designed in various configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of the present application.

[0035] It should be noted that the term "comprising" will be used in the embodiments of the present application to indicate the existence of the features declared thereafter, but does not exclude the addition of other features.

[0036] First, a brief explanation of the game mechanics of turn-based games:

[0037] Turn-based games usually determine the winner through continuous battles over multiple rounds. The core mechanism of turn-based games usually revolves around the "take-turn actions" of game units, emphasizing strategic planning rather than operation speed. In battle, players and enemies can act in a fixed order (for example, the order of actions is determined by the "speed" attribute of the game units). When it is the player's turn to act, the player can issue instructions to the game units under control (such as attack, use skills, props, defense, etc.) to execute the current unit's actions in the current round; whether it is the player's own or the enemy, when a game unit acts, it may affect the game status data of some or all of the player's or the enemy's game units. For example: a player's game unit with healing attributes may increase the health of all the player's game units after taking action; after the player's game unit uses a group skill, it may damage the health of all the enemy's game units.

[0038] After all the game rounds are over, the winner is determined by counting the game status data of each game unit on both sides; or after a game round, if the health of all the game units of one side is 0 or the game units escape, the game can be ended directly and the other side wins.

[0039] The core appeal of turn-based games lies in long-term strategic play, with outcomes often fluctuating due to randomness or instantaneous player decisions. This is different from real-time games, where the trend of winning and losing often changes dynamically in real time as the battle progresses. In turn-based games, however, the trend of winning and losing does not change with each unit action, but only at the end of the round, with a phased transition due to the results of the actions. Therefore, the trend of winning and losing in turn-based games is usually discrete.

[0040] Turn-based players pay more attention to strategic tolerance. Discrete situation changes allow players to make action decisions more calmly and pay more attention to long-term strategic planning.

[0041] The game data playback function can help players review the game after the game ends by viewing historical game battle scenes and game battle data, thereby helping players to review battle details, analyze the pros and cons of decisions, and then optimize strategies to better achieve victory in future game battles.

[0042] The existing turn-based game replay function only supports the independent display of various game status data change information of each game unit on both sides of the battle, for example, statistical display of game unit health value change information, or display of game unit damage value change information, etc.

[0043] The above-mentioned playback data display method results in the fragmentation of various status data of game units and an imbalance in information density. Players find it difficult to connect different data during the review process, and are unable to intuitively analyze the impact of different action strategies on the game's outcome, resulting in poor game review results for players.

[0044] Based on this, this solution provides an information processing method for turn-based games. By obtaining various status data of each game unit under unit actions or a single round, and integrating various status data of each game unit, the situation value of the two combatants under unit actions or a single round is predicted, thereby determining the situation difference between the two combatants. The situation difference is used to characterize the game winning and losing trends of the two combatants, and by displaying the game winning and losing trends, players can intuitively analyze the impact of different unit action strategies on the game winning and losing trends, helping players optimize game strategies.

[0045] Figure 1 A flow chart of an information processing method for a turn-based game provided in an embodiment of the present application; the method can be applied to a turn-based game, wherein the game includes a first combatant and a second combatant, and the first combatant and the second combatant each include at least one game unit; a game unit can be understood as a character, for example: other player characters or summoned beasts of player characters, etc. Any object that can perform game actions can be called a game unit.

[0046] like Figure 1 As shown, the method may include:

[0047] S101. Obtain status data of each game unit at the target analysis granularity in the game to be reviewed.

[0048] The target analysis granularity includes: a single round as the minimum analysis unit or a single game unit action as the minimum analysis unit.

[0049] The game to be reviewed is a historical game game that has ended. In turn-based games, the target analysis granularity can be a single round or a single game unit action in a single round. The status data of the game unit at the target analysis granularity can be obtained by taking each game unit action in a round as a node, or by taking the end of a single round as a node.

[0050] It is worth noting that the game units here include all game units of both combatants in the game. For example, the first combatant includes five game units and the second combatant includes five game units. Therefore, when obtaining status data, the status data of these ten game units is obtained.

[0051] Taking the target analysis granularity of a single game unit action as an example, after any game unit of the first combatant side takes action, or after any game unit of the second combatant side takes action, it is necessary to obtain the various status data of a total of ten game units of both combatants. The status data here refers to the game status data, or game attribute data.

[0052] Then, when the game has a total of 5 rounds, each round includes ten game unit actions, and the status data of each game unit at the target analysis granularity obtained will include 50 data, each of which includes the status data of each game unit after one game unit action.

[0053] S102: Determine the situation value of the target combatant at the target analysis granularity based on the status data of each game unit at the target analysis granularity.

[0054] The target combatant is the first combatant or the second combatant.

[0055] In turn-based games, the outcome of a battle is closely related to the player's status data after the battle. Therefore, based on the status data of each game unit at the target analysis granularity, the situation value of the target combatant at the target analysis granularity can be analyzed and determined.

[0056] Here, we use either side as the target side. We analyze the impact of each unit's status data on the target side's victory or defeat to determine the target side's Situation Value. This Situation Value represents the target side's chance of winning, to a certain extent. The larger the Situation Value, the greater the probability of victory. Of course, the Situation Value here is relative. For example, if the sum of the Situation Values ​​of both sides is 100, and the Situation Value of the target side is 70, while the Situation Value of the other side is 30, the target side's Situation Value is considered to be higher.

[0057] S103: Determine the situation difference of the target combatant or the situation difference of the non-target combatant according to the situation value of the target combatant at the target analysis granularity.

[0058] Among them, the situation difference of the target combatant is used to characterize the winning and losing trend of the target combatant at the target analysis granularity.

[0059] In some embodiments, based on the situation value of the target combatant at the target analysis granularity, the situation value of the non-target combatant can be determined. When the target combatant is the first combatant, the non-target combatant is the second combatant.

[0060] Based on the situation values ​​of the target combatant and the non-target combatant, the situation difference between the two sides from the perspective of the target combatant can be determined, or the situation difference between the two sides from the perspective of the non-target combatant can be determined.

[0061] Optionally, the situation difference of the target combatant can represent the game win-loss trend of the target combatant at the target analysis granularity. When the situation difference is greater than 0, it can be considered that the target combatant has an advantage and has a certain winning trend. When the situation difference is less than 0, it is considered that the non-target combatant has an advantage and the target combatant has a losing trend.

[0062] In this way, players can clearly understand the impact of action decisions at each target analysis granularity on their game victory, thereby helping players to plan strategies in the future to better achieve game victory and enhance the gaming experience.

[0063] In summary, the information processing method for a turn-based game provided in this embodiment includes: obtaining status data of each game unit at a target analysis granularity in a game to be reviewed; determining the situation value of the target combatant at the target analysis granularity based on the status data of each game unit at the target analysis granularity; and determining the situation difference of the target combatant or the situation difference of the non-target combatant based on the situation value of the target combatant at the target analysis granularity. This method introduces the concept of situation difference into the review of a turn-based game, uses a set action node as the analysis granularity, obtains the status data of each game unit of both combatants at the analysis granularity, analyzes and determines the situation value from the perspective of the target combatant, and finally determines the situation difference from the perspective of the target combatant or the situation difference from the perspective of the non-target combatant based on the situation value from the perspective of the target combatant. Through the situation difference, the player is helped to more intuitively analyze the impact of the action strategy at the analysis granularity on the player's game victory or defeat, thereby achieving a more refined and more based-on game review and helping the player optimize subsequent game strategies.

[0064] Optionally, in step S101, status data of each game unit at the target analysis granularity in the game to be reviewed is obtained, including: if the analysis granularity is a single game unit action as the minimum analysis unit, then the status data of each game unit after each game unit action in each round of the game to be reviewed is obtained.

[0065] In some embodiments, when a single game unit action is used as the analysis granularity, it is necessary to obtain the status data of all game units on the field after each game unit action in each round of the game to be reviewed, that is, one game unit action is used as a node, and what needs to be obtained at this node is the status data of each game unit on the field (that is, both sides of the game battle).

[0066] For example, each side of a battle contains 5 game units, namely game unit 1-game unit 10. The game consists of 3 rounds. In the first round, each game unit on both sides takes an action. After game unit 1 takes an action, the status data of game units 1-game unit 10 can be obtained. After game unit 2 takes an action, the status data of game units 1-game unit 10 can be obtained. And so on. In each round, 10 data points can be obtained, each containing the status data of 10 game units.

[0067] If the analysis granularity is a single round as the minimum analysis unit, the status data of each game unit at the end of each round in the game to be reviewed is obtained.

[0068] In other embodiments, when a single round is used as the analysis granularity, it is necessary to obtain the status data of each game unit at the end of each round in the game to be reviewed. Similarly, the status data of all game units of both sides of the battle at the end of the round is obtained.

[0069] In this scenario, one round corresponds to obtaining one piece of data, which may include status data of game units 1 to 10.

[0070] Figure 2 A flowchart of another information processing method for a turn-based game provided in an embodiment of the present application; optionally, in step S102, determining the situation value of the target combatant at the target analysis granularity based on the status data of each game unit at the target analysis granularity includes:

[0071] S201. Determine first prediction information corresponding to each game unit at the target analysis granularity using a pre-trained first prediction model corresponding to the target analysis granularity based on low-frequency state data in the state data of each game unit at the target analysis granularity.

[0072] Among them, the first prediction model is trained based on the actual victory or defeat results of the target combatant, and the first prediction information is used to characterize the degree of positivity of the status data of each game unit at the target analysis granularity to the victory or defeat of the target combatant.

[0073] In some embodiments, the status data may include multiple types of data. Taking a turn-based game as an example, the status data may include but is not limited to: health, buff / debuff, seal, fall, summoned spirit status, remaining resources, etc.

[0074] Among them, some low-frequency state data will directly affect the final combined judgment. If directly discarded, it will lead to misjudgment of the result. Based on this, this embodiment can perform prediction processing on the low-frequency state data separately through the first prediction model corresponding to the pre-trained target analysis granularity.

[0075] The low-frequency state data of each game unit at the target analysis granularity can be used as input to a first prediction model corresponding to the target analysis granularity to obtain first prediction information corresponding to each game unit at the target analysis granularity.

[0076] The first prediction result can be used as a dimensionality reduction result of the low-frequency state data, indicating the degree of positivity of the state data of the game unit at this time on the final game outcome.

[0077] It's worth noting that the first prediction model is trained from the perspective of the target combatant, using the target combatant's actual win / loss results as training labels. During prediction, the model also predicts the impact of low-frequency state data on the target combatant's win / loss.

[0078] It should be noted that for each type of low-frequency state data, a corresponding first prediction model can be trained separately to perform separate predictions for each type of low-frequency state data.

[0079] S202. Based on the status data of each game unit at the target analysis granularity and the first prediction information corresponding to each game unit at the target analysis granularity, a pre-trained second prediction model corresponding to the target analysis granularity is used to determine the second prediction information of the target combatant at the target analysis granularity.

[0080] Among them, the second prediction model is trained based on the actual victory or defeat results of the target combat party, and the second prediction information includes the situation value.

[0081] Optionally, the first prediction information corresponding to each game unit at the target analysis granularity can be used as a new feature, and combined with the status data of each game unit at the target analysis granularity to obtain new status data of each game unit at the target analysis granularity, and then based on the new status data of each game unit at the target analysis granularity, the pre-trained second prediction model corresponding to the target analysis granularity is used to predict the second prediction information of the target combatant at the target analysis granularity.

[0082] The second prediction model can be considered a global model, integrating all state data for prediction. The first prediction model can be considered a local model, making predictions based on a single state data point. The second prediction model is similarly trained from the perspective of the target combatant, using the target combatant's actual win-loss results.

[0083] The second prediction information may include a situation value. This second prediction information is essentially a numerical value, namely a situation value, which can represent the target combatant's combat situation. Through data transformation, the situation value can be blurred into a situation expression such as "forward, even, or unfavorable." When the situation is favorable, the target combatant is considered to be in an advantageous position, and when the situation is unfavorable, the target combatant is considered to be at a disadvantage.

[0084] It is worth noting that, taking the perspective of the target combatant as an example, the situation refers to the state of the target combatant at the tactical or strategic level at a specific point in time (that is, in this scenario, after a game unit takes action or at the end of a round). This state can be quantified in some way as an estimate of the probability of winning or losing.

[0085] In other words, the situation is a comprehensive assessment of the pros and cons of the current battle situation, reflecting "whether the current state is conducive to winning."

[0086] The concept of situation can help players clearly grasp the situation trends of both sides of the battle at the target analysis granularity, so as to help players analyze the impact of combat strategies on the situation and conduct reasonable strategic review.

[0087] Generally speaking, because many buffs in the game rarely appear (sparse), yet some rare buffs have a significant impact on victory or defeat, ignoring these buffs directly will result in inaccurate predictions. Therefore, this solution independently constructs a first prediction model based on low-frequency state data to analyze the impact of these low-frequency buffs and outputs a single value (the first prediction information) representing the combined impact of these buffs on victory or defeat. This value serves as a new feature and is incorporated into the final second prediction model, thereby preserving the role of these rare buffs while avoiding the issues caused by sparsity.

[0088] Figure 3 A flowchart of another information processing method for a turn-based game provided in an embodiment of the present application; optionally, in step S201, based on low-frequency state data in the state data of each game unit at the target analysis granularity, a pre-trained first prediction model corresponding to the target analysis granularity is used to determine first prediction information corresponding to each game unit at the target analysis granularity, including:

[0089] S301. Extract low-frequency state data of each game unit from state data of each game unit at a target analysis granularity according to at least one predetermined low-frequency state.

[0090] Generally, low-frequency status data may refer to features that appear less frequently in a game. Although they appear less frequently, they may have a significant impact on the outcome of the game at certain key points.

[0091] The game features numerous buffs (boost / debuff statuses), such as "Speed ​​Up," "Poison," "Seal," and "Shield," with a wide variety of buffs available in various combinations. Many of these buffs appear infrequently, meaning they only appear in a very small number of matches, such as rare special skill effects. While these buffs rarely appear, they can have a significant impact on the outcome of a match. For example, a rare buff that grants immunity to a critical attack could potentially turn the tide of battle at a crucial moment.

[0092] Alternatively, the frequency of occurrence of each buff can be counted based on data within a certain period, and the data can be sorted from high to low according to the frequency. The buff data of the lower sorted types are determined to be in a low-frequency state. The low-frequency state determined is not limited to one, but may be multiple.

[0093] Based on the determined low-frequency state, the low-frequency state data of each game unit can be extracted from the state data of each game unit at the target analysis granularity.

[0094] S302: Input the low-frequency state data of each game unit into a first prediction model corresponding to the target analysis granularity, and use the first prediction model to perform prediction to determine first prediction information corresponding to each game unit at the target analysis granularity.

[0095] Therefore, the low-frequency state data of each game unit can be input into the first prediction model corresponding to the target analysis granularity to predict and obtain the first prediction information corresponding to each game unit at the target analysis granularity.

[0096] Each low-frequency state data can be predicted using an independent first prediction model, so each low-frequency state data of each game unit can be input into the corresponding first prediction model for prediction processing. The game unit will obtain corresponding first prediction information under each low-frequency state data.

[0097] Figure 4 A flowchart of another information processing method for a turn-based game provided in an embodiment of the present application; optionally, in step S202, based on the status data of each game unit at the target analysis granularity and the first prediction information corresponding to each game unit at the target analysis granularity, a pre-trained second prediction model corresponding to the target analysis granularity is used to determine the second prediction information of the target combatant at the target analysis granularity, including:

[0098] S401: Merge the state data of each game unit at the target analysis granularity with the first prediction information corresponding to each game unit at the target analysis granularity to obtain target state data of each game unit at the target analysis granularity.

[0099] Optionally, the first prediction information corresponding to each game unit at the target analysis granularity can be combined with other state data of each game unit at the target analysis granularity except low-frequency state data to obtain target state data of each game unit at the target analysis granularity.

[0100] By merging the first prediction information corresponding to each game unit at the target analysis granularity with the other status data in the status data of each game unit at the target analysis granularity except the low-frequency status data, it is possible to retain the influence of the low-frequency status data on the final judgment result of the model, and avoid directly using the low-frequency status data, which may cause the model to overfit or misjudge due to sparsity.

[0101] For example, let's take a cooking scenario that's easy to understand. Suppose you're making a dish with dozens of spices, some of which you rarely use (such as saffron), but they have a huge impact on the taste. If you ignore these spices directly, the dish won't taste good.

[0102] So, each time I only add these rare spices to see how much impact they have on the taste, and then give them a "taste score"; later when I cook, I use this "taste score" to replace the original spices. This way I retain the influence of the spices without having to deal with so many spices every time.

[0103] The "flavor score" here is equivalent to the first prediction information output by the first prediction model.

[0104] S402: Input the target state data of each game unit at the target analysis granularity into the second prediction model, and use the second prediction model to perform prediction to determine the second prediction information of the target combatant at the target analysis granularity.

[0105] The target state data of each game unit at the target analysis granularity can be used as input data and input into the second prediction model to predict and obtain the second prediction information of the target combatant at the target analysis granularity.

[0106] Optionally, in step S401, the status data of each game unit at the target analysis granularity and the first prediction information corresponding to each game unit at the target analysis granularity are feature-merged to obtain the target status data of each game unit at the target analysis granularity, including: merging the first prediction information corresponding to each game unit at the target analysis granularity with other status data except low-frequency status data in the status data of each game unit at the target analysis granularity to obtain the target status data of each game unit at the target analysis granularity.

[0107] The first prediction information corresponding to each game unit at the target analysis granularity may be used to replace the low-frequency state data of each game unit to update the state data of each game unit and obtain the target state data of each game unit at the target analysis granularity.

[0108] Assume that the state data of a game unit at the target analysis granularity includes: state data 1, state data 2, state data 3, and state data 4. State data 1 is low-frequency state data, and the first prediction information corresponding to state data 1 is data 1. Therefore, the first prediction information corresponding to state data 1 is merged with the state data of the game unit at the target analysis granularity, resulting in the target state data of the game unit at the target analysis granularity being: data 1, state data 2, state data 3, and state data 4.

[0109] Figure 5 A flow chart of a model training method provided in an embodiment of the present application; the first prediction model mentioned above is trained in the following manner:

[0110] S501: Collect sample low-frequency status data of game units in each historical game within a historical period.

[0111] According to the pre-determined parameters of the low-frequency state type, sample low-frequency state data of each game unit can be extracted from the game data of each historical game within the historical time period.

[0112] For example, if the parameter of the determined low-frequency state type is "acceleration buff", the specific value of the acceleration buff can be extracted as the sample low-frequency state data.

[0113] In actual applications, there may be multiple types of low-frequency state parameters, so the extracted sample low-frequency state data of a game unit may also contain multiple.

[0114] It's worth noting that in this solution, when applying the aforementioned model, because it's used to review game situations during replay, the input for model application also consists of historical game data. Model training also uses historical game data, but the historical game data used for model training and model application can be from different historical periods. This means that the two historical game data sets should be as non-repeated as possible.

[0115] S502: Construct a first training data set based on sample low-frequency state data of game units in each historical game and actual game results of the target combatant in each historical game.

[0116] Since the target combatant’s perspective is the main perspective during use, the actual game results of the target combatant are also used as labels during model training.

[0117] For a particular type of sample low-frequency state data, a first training dataset for that sample low-frequency state data can be constructed using the sample low-frequency state data for each game unit in historical matches and the actual game results of the target combatant. That is, a separate first training dataset is constructed for each type of low-frequency state data to train an independent first prediction model. The actual game results of the target combatant refer to the target combatant's win or loss outcome.

[0118] S503: Use the first training data set to train and obtain a first prediction model.

[0119] Therefore, by using a first training data set corresponding to a sample low-frequency state data, a first prediction model corresponding to a sample low-frequency state data can be trained.

[0120] For each type of sample low-frequency state data, the same training method is used to train a first prediction model corresponding to each type of sample low-frequency state data. As mentioned above, the first prediction model is a model from the perspective of the target combatant. Thus, the first prediction information obtained is also prediction information from the perspective of the target combatant.

[0121] Figure 6 This is a flow chart of another model training method provided in an embodiment of the present application. If the target analysis granularity is to use a single game unit action as the minimum analysis unit, in step S501, sample low-frequency state data of the game unit in each historical game within the historical period is collected, including:

[0122] S601. Collect sample low-frequency status data of each game unit after each game unit takes action in each historical round of each historical game within a historical period.

[0123] When the target analysis granularity is different, the nodes for obtaining the sample low-frequency state data are also different.

[0124] When the target analysis granularity is a single game unit action, the data collected is a sample of low-frequency state data for each game unit after each game unit action in each historical round of each game. This is similar to the above method of obtaining state data for each game unit at different analysis granularities, except that only the sample low-frequency state data from multiple state data types is collected.

[0125] S602: Use the sample low-frequency state data of each game unit after each game unit takes action in each historical round of the historical game as the sample low-frequency state data of the game unit in the historical game.

[0126] In this way, the obtained sample low-frequency state data of each game unit after each game unit takes action in each historical round of the historical game is used as the sample low-frequency state data of the game unit in the historical game when training the first prediction model.

[0127] Figure 7 A flowchart of another model training method provided in an embodiment of the present application; if the target analysis granularity is a single round as the minimum analysis unit; in step S501, sample low-frequency state data of game units in each historical game within a historical period is collected, including:

[0128] S701: Collect sample low-frequency status data of each game unit at the end of each historical round in each historical game within a historical period.

[0129] Similarly, when the target analysis granularity is a single round, what is obtained is the sample low-frequency state data of each game unit at the end of each historical round in each historical game, that is, the acquisition node here is at the end of the round.

[0130] S702: Use the sample low-frequency state data of each game unit at the end of each historical round in the historical game as the sample low-frequency state data of the game unit in the historical game.

[0131] Similarly, the sample low-frequency state data of each game unit at the end of each historical round in the historical game is used as the sample low-frequency state data of the game unit in the historical game when the analysis granularity is a single round.

[0132] Optionally, in step S501, before collecting the sample low-frequency state data of the game unit in each historical game within the historical period, it also includes: determining at least one sample low-frequency state data from each sub-state data according to the frequency of occurrence of each sub-state data under the specified type of state data within the historical period.

[0133] Taking the status data in the scenarios listed in the above embodiments as an example, the specified type of status data here can be buff data. Since buff data includes many types, such as acceleration, seal, shield, etc., the acceleration, seal, and shield here can all be used as a sub-status data.

[0134] The sub-state data may be sorted according to the frequency of occurrence of each sub-state data within a preset period, such as the last month, and according to the sorting result, several sub-state data with less frequent occurrence may be determined as sample low-frequency state data.

[0135] Of course, the low-frequency state data determined in different game scenarios are different, but generally speaking, the occurrence frequencies of all state data can be sorted to determine the state data with less occurrence frequency as the sample low-frequency state data.

[0136] Figure 8 A flow chart of another model training method provided in an embodiment of the present application; the second prediction model is trained in the following manner:

[0137] S801. Collect sample status data of game units in each historical game within a historical period.

[0138] The sample state data here refers to all state data including sample low-frequency state data. Similarly, the sample state data of game units in each historical game during the historical period is obtained.

[0139] S802: Determine new sample state data of the game unit in each historical game based on the sample state data of the game unit in each historical game within the historical period and the sample low-frequency state data of the game unit in each historical game.

[0140] The sample state data of the game unit in each historical match can be combined with the sample low-frequency state data of the game unit in each historical match obtained above to obtain new sample state data of the game unit in each historical match.

[0141] The merging process here is the same as that in the model usage process. First, based on the sample low-frequency state data of the game unit, the first prediction model can be used to obtain the first prediction information corresponding to the game unit; then the first prediction information of the game unit is used to replace the sample low-frequency state data in the sample state data of the game unit to obtain the new sample state data of the game unit in each historical game.

[0142] S803: Construct a second training data set based on the new sample state data of the game units in each historical game and the actual game results of the target combatant in each historical game.

[0143] Therefore, the actual game results of the target combatants in each historical game are used as training labels, and the second training data set is composed of the new sample state data of the game units in each historical game and the actual game results of the target combatants in each historical game.

[0144] S804: Use the second training data set to train and obtain a second prediction model.

[0145] The second prediction model mentioned above can be trained using the second training data set.

[0146] Through the model training method of this method, a separate first prediction model can be trained for low-frequency state data to process the low-frequency state data separately; and the prediction results of the first prediction model can be used as a replacement for the original state data to train the second prediction model, thereby retaining the influence of the low-frequency state data on the model prediction results and avoiding the problem of inaccurate model prediction results caused by directly using low-frequency state data.

[0147] Figure 9A flowchart of another information processing method for a turn-based game provided in an embodiment of the present application; optionally, in step S103, determining the situation difference of the target combatant or the situation difference of the non-target combatant based on the situation value of the target combatant at the target analysis granularity includes:

[0148] S901. Determine the situation value of the non-target combatant at the target analysis granularity based on the situation value of the target combatant at the target analysis granularity.

[0149] Usually, when one side has an advantage, the other side is naturally at a disadvantage. That is, the situation values ​​of the two sides in the battle are relative, and the sum of the situation values ​​of the two sides in the battle is a fixed value. When the situation value of one side is large, the situation value of the other side is relatively small.

[0150] Therefore, the situation value of the non-target combatant at the target analysis granularity can be calculated based on the situation value of the target combatant at the target analysis granularity.

[0151] For example: the sum of the situation values ​​of the two combatants is 100. When the situation value of the target combatant at the target analysis granularity is 70, the situation value of the non-target combatant at the target analysis granularity is 30. At this time, it can be considered that the target combatant is in a "favorable" situation and the non-target combatant is in a "bad" situation. Similarly, when the situation values ​​of the target combatant and the non-target combatant are both 50, the target combatant and the non-target combatant are considered to be tied.

[0152] S902: Determine the situation difference of the target combatant or the situation difference of the non-target combatant at the target analysis granularity based on the situation value of the target combatant at the target analysis granularity and the situation value of the non-target combatant at the target analysis granularity.

[0153] Optionally, by subtracting the situation value of the target combatant at the target analysis granularity from the situation value of the non-target combatant at the target analysis granularity, the situation difference of the target combatant or the situation difference of the non-target combatant at the target analysis granularity can be obtained.

[0154] Among them, when the situation value of the target combatant at the target analysis granularity is subtracted from the situation value of the non-target combatant at the target analysis granularity, the result is the situation difference of the target combatant; conversely, when the situation value of the non-target combatant at the target analysis granularity is subtracted from the situation value of the target combatant at the target analysis granularity, the result is the situation difference of the non-target combatant.

[0155] Optionally, the method of the present application may further include: generating a game situation review screen of the game to be reviewed based on the situation difference of the target combatant or the situation difference of the non-target combatant at the target analysis granularity; the game situation review screen displays the win-loss trend from the perspective of the target combatant or the win-loss trend from the perspective of the non-target combatant at the target analysis granularity.

[0156] In some embodiments, the difference in the situation between the two combatants can be graphically displayed to generate a game win-loss trend chart, which is then displayed in the game situation review interface for the game to be reviewed. The game win-loss trend chart can help players intuitively understand the win-loss trend of the two combatants at each analysis node, thereby facilitating the analysis of status data at key nodes that affect the final outcome, thereby helping to formulate reasonable game strategies in subsequent games.

[0157] Figure 10 A schematic diagram of a game situation review interface provided in an embodiment of the present application is shown as follows: Figure 10 As shown, it shows the situation review from the perspective of the target combatant (that is, our side), in which the winning and losing trend of the target combatant is displayed in the form of a bar chart. It can be seen that in the first round, the target combatant is in a "reverse" trend, which is obtained through the situation difference conversion of the target combatant.

[0158] When the target side's situation value is lower than that of the non-target side, and the difference is less than 0, the target side is in a "disadvantageous" situation, and its situation is displayed as a red "disadvantageous" situation in the bar chart. The length of the red bar chart can be determined by the specific difference in situation value. The larger the situation difference, the longer the red bar chart, indicating that the "disadvantageous" situation of the target side is more serious. When the target side is in a "disadvantageous" situation, the non-target side is in a "favorable" situation.

[0159] When the situation value of the target combatant is greater than that of the non-target combatant, the difference is greater than 0. At this time, the target combatant is in a "favorable" situation, which is displayed as a green favorable situation in the bar chart.

[0160] Therefore, through the calculation results of the situation difference, the situation difference can be converted into a win-loss trend (situation chart) for display, so that players can more intuitively understand the game situation trends of the two fighting parties, to help players analyze the final game outcome. Moreover, through the situation chart, some key nodes that play a key role in the game outcome can be intuitively determined, so that players can review the game data at the key nodes to assist in formulating more reasonable game strategies in the future.

[0161] Optionally, the status data of the game unit includes one or more of the following: life attributes, combat attributes, and resource attributes of the game unit.

[0162] Typically, in a turn-based game, the status data of a game unit may include the life attributes of the game unit, such as the game unit's health value, blood volume value, etc.; combat attributes, such as the game unit's attack value, defense value, spell value, blood recovery value, etc.; resource attributes, such as the number of summoned spirits and medicines of the game unit, etc.

[0163] It should be noted that the above embodiment is based on the application of the model to information prediction in the game review process. In actual applications, the above-trained model can also be applied to predictions in real-time games. For example, during the live broadcast of the game, the situation value of the current game can be predicted in real time based on the real-time collected game data and the above-mentioned method, and a review screen can be generated in real time.

[0164] In summary, the information processing method for a turn-based game provided in this embodiment includes: obtaining status data of each game unit at a target analysis granularity in a game to be reviewed; determining the situation value of the target combatant at the target analysis granularity based on the status data of each game unit at the target analysis granularity; and determining the situation difference of the target combatant or the situation difference of the non-target combatant based on the situation value of the target combatant at the target analysis granularity. This method introduces the concept of situation difference into the review of a turn-based game, uses a set action node as the analysis granularity, obtains the status data of each game unit of both combatants at the analysis granularity, analyzes and determines the situation value from the perspective of the target combatant, and finally determines the situation difference from the perspective of the target combatant or the situation difference from the perspective of the non-target combatant based on the situation value from the perspective of the target combatant. Through the situation difference, the player is helped to more intuitively analyze the impact of the action strategy at the analysis granularity on the player's game victory or defeat, thereby achieving a more refined and more based-on game review and helping the player optimize subsequent game strategies.

[0165] Among them, for low-frequency state data, the first prediction model is independently trained to process the low-frequency state data separately, and the processing result of the low-frequency state data, that is, the first prediction information, is used instead of the low-frequency state data to predict the final situation value, so that the influence of the low-frequency state data on the final prediction result of the model can be retained, and the direct use of the low-frequency state data can be avoided to cause the model to overfit or misjudgment, thereby improving the accuracy of the prediction results of the second prediction model.

[0166] By converting the situation difference into a situation diagram and visually displaying the game's win-loss trend through the situation diagram, players can intuitively view the win-loss trend at each node and intuitively identify some key nodes that have a key impact on the final outcome of the game. The game data of the key nodes can be reviewed in a targeted manner to assist in formulating more reasonable game strategies in subsequent games.

[0167] The following describes the apparatus, device, storage medium, etc. used to execute the information processing method for the turn-based game provided in this application. The specific implementation process and technical effects are described above and will not be repeated below.

[0168] Figure 11A schematic diagram of an information processing device for a turn-based game provided in an embodiment of the present application, which is applied to a turn-based game, wherein the game includes a first combatant and a second combatant, each of which includes at least one game unit; the functions implemented by the information processing device for the turn-based game correspond to the steps performed by the above method. The device can be understood as a server, or a processor of a server, or as a component independent of the above server or processor that implements the functions of the present application under the control of the server, such as Figure 11 As shown, the apparatus may include: an acquisition module 100 and a determination module 200;

[0169] The acquisition module 100 is used to obtain the status data of each game unit at the target analysis granularity in the game to be reviewed; the target analysis granularity includes: a single round as the minimum analysis unit or a single game unit action as the minimum analysis unit;

[0170] A determination module 200 is configured to determine the situation value of a target combatant at the target analysis granularity based on the status data of each game unit at the target analysis granularity, where the target combatant is the first combatant or the second combatant;

[0171] The determination module 200 is used to determine the situation difference of the target combatant or the situation difference of the non-target combatant based on the situation value of the target combatant at the target analysis granularity, wherein the situation difference of the target combatant is used to characterize the winning and losing trend of the target combatant at the target analysis granularity.

[0172] Optionally, the acquisition module 100 is specifically configured to acquire the status data of each game unit after each game unit action in each round of the game to be reviewed, if the analysis granularity is a single game unit action as the minimum analysis unit;

[0173] If the analysis granularity is a single round as the minimum analysis unit, the status data of each game unit at the end of each round in the game to be reviewed is obtained.

[0174] Optionally, the determination module 200 is specifically configured to determine first prediction information corresponding to each game unit at the target analysis granularity based on low-frequency state data in the state data of each game unit at the target analysis granularity using a pre-trained first prediction model corresponding to the target analysis granularity; the first prediction model is trained based on actual win / loss results of the target combatant, and the first prediction information is used to represent the degree of positivity of the state data of each game unit at the target analysis granularity in affecting the win / loss of the target combatant;

[0175] According to the status data of each game unit at the target analysis granularity and the first prediction information corresponding to each game unit at the target analysis granularity, a pre-trained second prediction model corresponding to the target analysis granularity is used to determine the second prediction information of the target combatant at the target analysis granularity. The second prediction model is trained based on the actual win or loss results of the target combatant, and the second prediction information includes the situation value.

[0176] Optionally, the determination module 200 is specifically configured to extract low-frequency state data of each game unit from the state data of each game unit at a target analysis granularity according to at least one predetermined low-frequency state;

[0177] The low-frequency state data of each game unit is input into a first prediction model corresponding to the target analysis granularity, and the first prediction model performs prediction to determine first prediction information corresponding to each game unit at the target analysis granularity.

[0178] Optionally, the determination module 200 is specifically configured to perform feature merging on the state data of each game unit at the target analysis granularity and the first prediction information corresponding to each game unit at the target analysis granularity to obtain target state data of each game unit at the target analysis granularity;

[0179] The target state data of each game unit at the target analysis granularity is input into the second prediction model, and the second prediction model performs prediction to determine the second prediction information of the target combatant at the target analysis granularity.

[0180] Optionally, the determination module 200 is specifically used to merge the first prediction information corresponding to each game unit at the target analysis granularity with other status data of each game unit at the target analysis granularity except low-frequency status data to obtain the target status data of each game unit at the target analysis granularity.

[0181] Optionally, it further includes: a training module;

[0182] The training module is used to collect low-frequency status data of game units in each historical game within a historical period;

[0183] Constructing a first training dataset based on sample low-frequency state data of game units in each historical game and actual game results of the target combatant in each historical game;

[0184] A first training data set is used to train a first prediction model.

[0185] If the target analysis granularity is to use a single game unit action as the minimum analysis unit; the training module is specifically used to collect sample low-frequency state data of each game unit after each game unit action in each historical round of each historical game within the historical period;

[0186] The sample low-frequency state data of each game unit after each game unit takes action in each historical round of the historical game is used as the sample low-frequency state data of the game unit in the historical game.

[0187] If the target analysis granularity is a single round as the minimum analysis unit; the training module is specifically used to collect sample low-frequency status data of each game unit at the end of each historical round in each historical game within the historical period;

[0188] The sample low-frequency state data of each game unit at the end of each historical round in the historical game is used as the sample low-frequency state data of the game unit in the historical game.

[0189] Optionally, the determination module 200 is further configured to determine at least one sample low-frequency state data from each sub-state data according to the occurrence frequency of each sub-state data under the specified type of state data in a historical period.

[0190] Optionally, the training module is further used to collect sample status data of game units in each historical game within a historical period;

[0191] Determine new sample state data for each game unit in each historical match based on the sample state data for the game unit in each historical match and the sample low-frequency state data for the game unit in each historical match;

[0192] Constructing a second training dataset based on new sample state data of game units in each historical game and actual game results of the target combatant in each historical game;

[0193] A second prediction model is obtained by training using the second training data set.

[0194] Optionally, the determination module 200 is specifically configured to determine the situation value of the non-target combatant at the target analysis granularity based on the situation value of the target combatant at the target analysis granularity;

[0195] According to the situation value of the target combatant at the target analysis granularity and the situation value of the non-target combatant at the target analysis granularity, the situation difference of the target combatant or the situation difference of the non-target combatant at the target analysis granularity is determined.

[0196] Optionally, it further includes: a generation module;

[0197] The generation module is used to generate a game situation review screen for the game to be reviewed based on the situation difference of the target combatant or the situation difference of the non-target combatant at the target analysis granularity; the game situation review screen displays the win-loss trend from the perspective of the target combatant or the win-loss trend from the perspective of the non-target combatant at the target analysis granularity.

[0198] Optionally, the status data of the game unit includes one or more of the following: life attributes, combat attributes, and resource attributes of the game unit.

[0199] The above modules can be one or more integrated circuits configured to implement the above methods, such as one or more application specific integrated circuits (ASICs), one or more digital single processors (DSPs), or one or more field programmable gate arrays (FPGAs). For another example, when a module is implemented by scheduling program code through a processing element, the processing element can be a general-purpose processor, such as a central processing unit (CPU) or other processor that can call program code. For another example, these modules can be integrated together and implemented in the form of a system-on-a-chip (SOC).

[0200] The above modules can be connected or communicate with each other via a wired connection or a wireless connection. The wired connection may include a metal cable, an optical cable, a hybrid cable, etc., or any combination thereof. The wireless connection may include a connection in the form of a LAN, a WAN, Bluetooth, ZigBee, or NFC, or any combination thereof. Two or more modules can be combined into a single module, and any module can be divided into two or more units. Those skilled in the art will clearly understand that for the convenience and brevity of description, the specific working process of the system and device described above can refer to the corresponding process in the method embodiment, and will not be repeated in this application.

[0201] Figure 12 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application, including: a processor 801, a storage medium 802, and a bus 803. The storage medium 802 stores machine-readable instructions executable by the processor 801. When the electronic device runs an information processing method for a turn-based game in the embodiment, the processor 801 communicates with the storage medium 802 via the bus 803. The processor 801 executes the machine-readable instructions to perform the following steps:

[0202] Obtain the status data of each game unit at the target analysis granularity in the game to be reviewed; the target analysis granularity includes: a single round as the minimum analysis unit or a single game unit action as the minimum analysis unit;

[0203] Determine the situation value of the target combatant at the target analysis granularity based on the status data of each game unit at the target analysis granularity, and the target combatant is the first combatant or the second combatant;

[0204] According to the situation value of the target combatant at the target analysis granularity, the situation difference of the target combatant or the situation difference of the non-target combatant is determined, wherein the situation difference of the target combatant is used to characterize the winning and losing trend of the target combatant at the target analysis granularity.

[0205] In one feasible embodiment, when executing the process of obtaining the status data of each game unit at a target analysis granularity in the game to be reviewed, the processor 801 is specifically configured to: if the analysis granularity is a single game unit action as the minimum analysis unit, obtain the status data of each game unit after each game unit action in each round of the game to be reviewed;

[0206] If the analysis granularity is a single round as the minimum analysis unit, the status data of each game unit at the end of each round in the game to be reviewed is obtained.

[0207] In one feasible embodiment, when determining the situation value of the target combatant at the target analysis granularity based on the status data of each game unit at the target analysis granularity, the processor 801 is specifically configured to: determine first prediction information corresponding to each game unit at the target analysis granularity based on low-frequency status data in the status data of each game unit at the target analysis granularity using a pre-trained first prediction model corresponding to the target analysis granularity; the first prediction model is trained based on the actual victory or defeat results of the target combatant, and the first prediction information is used to represent the degree of positivity of the status data of each game unit at the target analysis granularity for the victory or defeat of the target combatant;

[0208] According to the status data of each game unit at the target analysis granularity and the first prediction information corresponding to each game unit at the target analysis granularity, a pre-trained second prediction model corresponding to the target analysis granularity is used to determine the second prediction information of the target combatant at the target analysis granularity. The second prediction model is trained based on the actual win or loss results of the target combatant, and the second prediction information includes the situation value.

[0209] In one feasible embodiment, when the processor 801 determines the first prediction information corresponding to each game unit at the target analysis granularity based on the low-frequency state data in the state data of each game unit at the target analysis granularity using a pre-trained first prediction model corresponding to the target analysis granularity, the processor 801 is specifically configured to: extract each low-frequency state data of each game unit from the state data of each game unit at the target analysis granularity based on at least one pre-determined low-frequency state;

[0210] The low-frequency state data of each game unit is input into a first prediction model corresponding to the target analysis granularity, and the first prediction model performs prediction to determine first prediction information corresponding to each game unit at the target analysis granularity.

[0211] In one feasible embodiment, when the processor 801 determines the second prediction information of the target combatant at the target analysis granularity based on the state data of each game unit at the target analysis granularity and the first prediction information corresponding to each game unit at the target analysis granularity, using a pre-trained second prediction model corresponding to the target analysis granularity, the processor 801 is specifically configured to: perform feature merging on the state data of each game unit at the target analysis granularity and the first prediction information corresponding to each game unit at the target analysis granularity to obtain target state data of each game unit at the target analysis granularity;

[0212] The target state data of each game unit at the target analysis granularity is input into the second prediction model, and the second prediction model performs prediction to determine the second prediction information of the target combatant at the target analysis granularity.

[0213] In a feasible implementation scheme, when the processor 801 performs feature merging of the status data of each game unit at the target analysis granularity and the first prediction information corresponding to each game unit at the target analysis granularity to obtain the target status data of each game unit at the target analysis granularity, it is specifically used to: merge the first prediction information corresponding to each game unit at the target analysis granularity with other status data except low-frequency status data in the status data of each game unit at the target analysis granularity to obtain the target status data of each game unit at the target analysis granularity.

[0214] In a feasible embodiment, when executing the training of the first prediction model, the processor 801 is specifically configured to: collect sample low-frequency state data of game units in each historical game within a historical period;

[0215] Constructing a first training dataset based on sample low-frequency state data of game units in each historical game and actual game results of the target combatant in each historical game;

[0216] A first training data set is used to train a first prediction model.

[0217] In one feasible embodiment, if the target analysis granularity is to use a single game unit action as the minimum analysis unit, the processor 801, when executing the collection of sample low-frequency state data of the game units in each historical game within the historical period, is specifically configured to: collect sample low-frequency state data of each game unit after each game unit action in each historical round of each historical game within the historical period;

[0218] The sample low-frequency state data of each game unit after each game unit takes action in each historical round of the historical game is used as the sample low-frequency state data of the game unit in the historical game.

[0219] In one feasible embodiment, if the target analysis granularity is a single round as the minimum analysis unit, the processor 801, when executing the collection of sample low-frequency status data of game units in each historical game within the historical period, is specifically configured to: collect sample low-frequency status data of each game unit at the end of each historical round in each historical game within the historical period;

[0220] The sample low-frequency state data of each game unit at the end of each historical round in the historical game is used as the sample low-frequency state data of the game unit in the historical game.

[0221] In a feasible implementation scheme, before executing the collection of sample low-frequency status data of game units in each historical game within the historical period, the processor 801 is also used to: determine at least one sample low-frequency status data from each sub-status data based on the frequency of occurrence of each sub-status data under the specified type of status data within the historical period.

[0222] In a feasible embodiment, when executing the training of the second prediction model, the processor 801 is specifically configured to: collect sample state data of game units in each historical game within a historical period;

[0223] Determine new sample state data for each game unit in each historical match based on the sample state data for the game unit in each historical match and the sample low-frequency state data for the game unit in each historical match;

[0224] Constructing a second training dataset based on new sample state data of game units in each historical game and actual game results of the target combatant in each historical game;

[0225] A second prediction model is obtained by training using the second training data set.

[0226] In one feasible embodiment, when the processor 801 determines the situation difference of the target combatant or the situation difference of the non-target combatant based on the situation value of the target combatant at the target analysis granularity, it is specifically configured to: determine the situation value of the non-target combatant at the target analysis granularity based on the situation value of the target combatant at the target analysis granularity;

[0227] According to the situation value of the target combatant at the target analysis granularity and the situation value of the non-target combatant at the target analysis granularity, the situation difference of the target combatant or the situation difference of the non-target combatant at the target analysis granularity is determined.

[0228] In a feasible implementation scheme, the processor 801 is also used to generate a game situation review screen for the game to be reviewed based on the situation difference of the target combatant or the situation difference of the non-target combatant at the target analysis granularity; the game situation review screen displays the win-loss trend from the perspective of the target combatant or the win-loss trend from the perspective of the non-target combatant at the target analysis granularity.

[0229] In a feasible implementation manner, the status data of the game unit includes one or more of the following: life attributes, combat attributes, and resource attributes of the game unit.

[0230] Among them, the storage medium 802 stores program code, and when the program code is executed by the processor 801, the processor 801 executes the various steps of the information processing method of the turn-based game according to various exemplary embodiments of the present application described in the above "Exemplary Method" section of this specification.

[0231] The processor 801 can be a general-purpose processor, such as a central processing unit (CPU), a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, and can implement or execute the various methods, steps and logic block diagrams disclosed in the embodiments of the present application. The general-purpose processor can be a microprocessor or any conventional processor, etc. The steps of the method disclosed in conjunction with the embodiments of the present application can be directly embodied as being executed by a hardware processor, or can be executed by a combination of hardware and software modules in the processor.

[0232] The storage medium 802 is a non-volatile computer-readable storage medium that can be used to store non-volatile software programs, non-volatile computer executable programs and modules. The memory can include at least one type of storage medium, such as a flash memory, a hard disk, a multimedia card, a card-type memory, a random access memory (Random Access Memory, RAM), a static random access memory (Static Random Access Memory, SRAM), a programmable read-only memory (Programmable Read Only Memory, PROM), a read-only memory (Read Only Memory, ROM), an electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, EEPROM), a magnetic memory, a disk, an optical disc, etc. The memory is any other medium that can be used to carry or store a desired program code in the form of an instruction or data structure and can be accessed by a computer, but is not limited thereto. The storage medium 802 in the embodiment of the present application can also be a circuit or any other device that can realize a storage function, for storing program instructions and / or data.

[0233] Optionally, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the processor performs the following steps:

[0234] Obtain the status data of each game unit at the target analysis granularity in the game to be reviewed; the target analysis granularity includes: a single round as the minimum analysis unit or a single game unit action as the minimum analysis unit;

[0235] Determine the situation value of the target combatant at the target analysis granularity based on the status data of each game unit at the target analysis granularity, and the target combatant is the first combatant or the second combatant;

[0236] According to the situation value of the target combatant at the target analysis granularity, the situation difference of the target combatant or the situation difference of the non-target combatant is determined, wherein the situation difference of the target combatant is used to characterize the winning and losing trend of the target combatant at the target analysis granularity.

[0237] In one feasible embodiment, when executing the process of obtaining the status data of each game unit at a target analysis granularity in the game to be reviewed, the processor 801 is specifically configured to: if the analysis granularity is a single game unit action as the minimum analysis unit, obtain the status data of each game unit after each game unit action in each round of the game to be reviewed;

[0238] If the analysis granularity is a single round as the minimum analysis unit, the status data of each game unit at the end of each round in the game to be reviewed is obtained.

[0239] In one feasible embodiment, when determining the situation value of the target combatant at the target analysis granularity based on the status data of each game unit at the target analysis granularity, the processor 801 is specifically configured to: determine first prediction information corresponding to each game unit at the target analysis granularity based on low-frequency status data in the status data of each game unit at the target analysis granularity using a pre-trained first prediction model corresponding to the target analysis granularity; the first prediction model is trained based on the actual victory or defeat results of the target combatant, and the first prediction information is used to represent the degree of positivity of the status data of each game unit at the target analysis granularity for the victory or defeat of the target combatant;

[0240] According to the status data of each game unit at the target analysis granularity and the first prediction information corresponding to each game unit at the target analysis granularity, a pre-trained second prediction model corresponding to the target analysis granularity is used to determine the second prediction information of the target combatant at the target analysis granularity. The second prediction model is trained based on the actual win or loss results of the target combatant, and the second prediction information includes the situation value.

[0241] In one feasible embodiment, when the processor 801 determines the first prediction information corresponding to each game unit at the target analysis granularity based on the low-frequency state data in the state data of each game unit at the target analysis granularity using a pre-trained first prediction model corresponding to the target analysis granularity, the processor 801 is specifically configured to: extract each low-frequency state data of each game unit from the state data of each game unit at the target analysis granularity based on at least one pre-determined low-frequency state;

[0242] The low-frequency state data of each game unit is input into a first prediction model corresponding to the target analysis granularity, and the first prediction model performs prediction to determine first prediction information corresponding to each game unit at the target analysis granularity.

[0243] In one feasible embodiment, when the processor 801 determines the second prediction information of the target combatant at the target analysis granularity based on the state data of each game unit at the target analysis granularity and the first prediction information corresponding to each game unit at the target analysis granularity, using a pre-trained second prediction model corresponding to the target analysis granularity, the processor 801 is specifically configured to: perform feature merging on the state data of each game unit at the target analysis granularity and the first prediction information corresponding to each game unit at the target analysis granularity to obtain target state data of each game unit at the target analysis granularity;

[0244] The target state data of each game unit at the target analysis granularity is input into the second prediction model, and the second prediction model performs prediction to determine the second prediction information of the target combatant at the target analysis granularity.

[0245] In a feasible implementation scheme, when the processor 801 performs feature merging of the status data of each game unit at the target analysis granularity and the first prediction information corresponding to each game unit at the target analysis granularity to obtain the target status data of each game unit at the target analysis granularity, it is specifically used to: merge the first prediction information corresponding to each game unit at the target analysis granularity with other status data except low-frequency status data in the status data of each game unit at the target analysis granularity to obtain the target status data of each game unit at the target analysis granularity.

[0246] In a feasible embodiment, when executing the training of the first prediction model, the processor 801 is specifically configured to: collect sample low-frequency state data of game units in each historical game within a historical period;

[0247] Constructing a first training dataset based on sample low-frequency state data of game units in each historical game and actual game results of the target combatant in each historical game;

[0248] A first training data set is used to train a first prediction model.

[0249] In one feasible embodiment, if the target analysis granularity is to use a single game unit action as the minimum analysis unit, the processor 801, when executing the collection of sample low-frequency state data of the game units in each historical game within the historical period, is specifically configured to: collect sample low-frequency state data of each game unit after each game unit action in each historical round of each historical game within the historical period;

[0250] The sample low-frequency state data of each game unit after each game unit takes action in each historical round of the historical game is used as the sample low-frequency state data of the game unit in the historical game.

[0251] In one feasible embodiment, if the target analysis granularity is a single round as the minimum analysis unit, the processor 801, when executing the collection of sample low-frequency status data of game units in each historical game within the historical period, is specifically configured to: collect sample low-frequency status data of each game unit at the end of each historical round in each historical game within the historical period;

[0252] The sample low-frequency state data of each game unit at the end of each historical round in the historical game is used as the sample low-frequency state data of the game unit in the historical game.

[0253] In a feasible implementation scheme, before executing the collection of sample low-frequency status data of game units in each historical game within the historical period, the processor 801 is also used to: determine at least one sample low-frequency status data from each sub-status data based on the frequency of occurrence of each sub-status data under the specified type of status data within the historical period.

[0254] In a feasible embodiment, when executing the training of the second prediction model, the processor 801 is specifically configured to: collect sample state data of game units in each historical game within a historical period;

[0255] Determine new sample state data for each game unit in each historical match based on the sample state data for the game unit in each historical match and the sample low-frequency state data for the game unit in each historical match;

[0256] Constructing a second training dataset based on new sample state data of game units in each historical game and actual game results of the target combatant in each historical game;

[0257] A second prediction model is obtained by training using the second training data set.

[0258] In one feasible embodiment, when the processor 801 determines the situation difference of the target combatant or the situation difference of the non-target combatant based on the situation value of the target combatant at the target analysis granularity, it is specifically configured to: determine the situation value of the non-target combatant at the target analysis granularity based on the situation value of the target combatant at the target analysis granularity;

[0259] According to the situation value of the target combatant at the target analysis granularity and the situation value of the non-target combatant at the target analysis granularity, the situation difference of the target combatant or the situation difference of the non-target combatant at the target analysis granularity is determined.

[0260] In a feasible implementation scheme, the processor 801 is also used to generate a game situation review screen for the game to be reviewed based on the situation difference of the target combatant or the situation difference of the non-target combatant at the target analysis granularity; the game situation review screen displays the win-loss trend from the perspective of the target combatant or the win-loss trend from the perspective of the non-target combatant at the target analysis granularity.

[0261] In a feasible implementation manner, the status data of the game unit includes one or more of the following: life attributes, combat attributes, and resource attributes of the game unit.

[0262] In the embodiment of the present application, the computer program can also execute other machine-readable instructions when run by the processor to execute other methods described in the embodiment. For the specific execution method steps and principles, please refer to the description of the embodiment and will not be repeated here.

[0263] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0264] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0265] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0266] The above-mentioned integrated unit implemented in the form of a software functional unit can be stored in a computer-readable storage medium. The above-mentioned software functional unit is stored in a storage medium and includes a number of instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) or a processor (English: processor) to perform some steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: a USB flash drive, a mobile hard disk, a read-only memory (English: Read-Only Memory, abbreviated: ROM), a random access memory (English: Random Access Memory, abbreviated: RAM), a disk or an optical disk, and other media that can store program code.

Claims

1. An information processing method for a turn-based game, characterized in that: Applicable to a turn-based game, the game includes a first combatant and a second combatant, and the first combatant and the second combatant each include at least one game unit; The method comprises: Obtaining status data of each game unit at a target analysis granularity in the game to be reviewed; the target analysis granularity includes: a single round as the minimum analysis unit or a single game unit action as the minimum analysis unit; Determine, based on the status data of each game unit at the target analysis granularity, a situation value of a target combatant at the target analysis granularity, the target combatant being the first combatant or the second combatant; According to the situation value of the target combatant at the target analysis granularity, the situation difference of the target combatant or the situation difference of the non-target combatant is determined, wherein the situation difference of the target combatant is used to characterize the winning and losing trend of the target combatant at the target analysis granularity.

2. The method according to claim 1, characterized in that The step of obtaining the status data of each game unit at the target analysis granularity in the game to be reviewed includes: If the analysis granularity is a single game unit action as the minimum analysis unit, then obtaining the status data of each game unit after each game unit action in each round of the game to be reviewed; If the analysis granularity is a single round as the minimum analysis unit, the status data of each game unit at the end of each round in the game to be reviewed is obtained.

3. The method according to claim 1, characterized in that Determining the situation value of the target combatant at the target analysis granularity based on the status data of each game unit at the target analysis granularity includes: Determining first prediction information corresponding to each game unit at the target analysis granularity based on low-frequency state data in the state data of each game unit at the target analysis granularity using a pre-trained first prediction model corresponding to the target analysis granularity; the first prediction model is trained based on the actual victory or defeat results of the target combatant, and the first prediction information is used to represent the degree of positivity of the state data of each game unit at the target analysis granularity in affecting the victory or defeat of the target combatant; According to the status data of each game unit at the target analysis granularity and the first prediction information corresponding to each game unit at the target analysis granularity, a pre-trained second prediction model corresponding to the target analysis granularity is used to determine the second prediction information of the target combatant at the target analysis granularity. The second prediction model is trained based on the actual win or loss results of the target combatant, and the second prediction information includes the situation value.

4. The method according to claim 3, characterized in that The method of determining first prediction information corresponding to each game unit at the target analysis granularity by using a pre-trained first prediction model corresponding to the target analysis granularity based on low-frequency state data in the state data of each game unit at the target analysis granularity includes: Extracting low-frequency state data of each game unit from the state data of each game unit at the target analysis granularity according to at least one predetermined low-frequency state; The low-frequency state data of each game unit is input into a first prediction model corresponding to the target analysis granularity, and the first prediction model is used to perform prediction to determine first prediction information corresponding to each game unit at the target analysis granularity.

5. The method according to claim 3, characterized in that The method further comprises: determining, based on the status data of each game unit at the target analysis granularity and the first prediction information corresponding to each game unit at the target analysis granularity, a second prediction model corresponding to the target analysis granularity that is pre-trained, and second prediction information of the target combatant at the target analysis granularity, comprising: Merging the state data of each game unit at the target analysis granularity with the first prediction information corresponding to each game unit at the target analysis granularity to obtain target state data of each game unit at the target analysis granularity; The target state data of each game unit at the target analysis granularity is input into the second prediction model, and the second prediction model performs prediction to determine the second prediction information of the target combatant at the target analysis granularity.

6. The method according to claim 5, characterized in that The step of merging the state data of each game unit at the target analysis granularity with the first prediction information corresponding to each game unit at the target analysis granularity to obtain target state data of each game unit at the target analysis granularity includes: The first prediction information corresponding to each game unit at the target analysis granularity is combined with other status data of each game unit at the target analysis granularity except low-frequency status data to obtain target status data of each game unit at the target analysis granularity.

7. The method according to any one of claims 3 to 6, characterized in that: The first prediction model is trained in the following way: Collect low-frequency status data of game units in each historical game within the historical period; Constructing a first training data set based on sample low-frequency state data of game units in each historical game and actual game results of the target combatant in each historical game; The first prediction model is trained using the first training data set.

8. The method according to claim 7, characterized in that If the target analysis granularity is to use a single game unit action as the minimum analysis unit; the collected low-frequency state data of the game units in each historical game within the historical period includes: Collect sample low-frequency status data of each game unit after each game unit takes action in each historical round of each historical game within the historical period; The sample low-frequency state data of each game unit after each game unit takes action in each historical round of the historical game is used as the sample low-frequency state data of the game unit in the historical game.

9. The method according to claim 7, characterized in that If the target analysis granularity is a single round as the minimum analysis unit; the low-frequency state data of the game units in each historical game in the historical period is collected, including: Collect sample low-frequency status data of each game unit at the end of each historical round in each historical game within the historical period; The sample low-frequency state data of each game unit at the end of each historical round in the historical game is used as the sample low-frequency state data of the game unit in the historical game.

10. The method according to claim 7, characterized in that Before collecting the sample low-frequency status data of the game units in each historical game within the historical period, the following steps are also included: According to the occurrence frequency of each sub-state data under the specified type of state data in the historical period, at least one sample low-frequency state data is determined from the sub-state data.

11. The method according to claim 7, characterized in that The second prediction model is trained in the following way: Collect sample status data of game units in each historical game within the historical period; Determine new sample state data for each game unit in each historical match based on the sample state data for the game unit in each historical match and the sample low-frequency state data for the game unit in each historical match; Constructing a second training data set based on new sample state data of game units in each historical game and actual game results of the target combatant in each historical game; The second prediction model is trained using the second training data set.

12. The method according to claim 1, characterized in that Determining the situation difference of the target combatant or the situation difference of the non-target combatant based on the situation value of the target combatant at the target analysis granularity includes: Determining the situation value of the non-target combatant at the target analysis granularity according to the situation value of the target combatant at the target analysis granularity; According to the situation value of the target combatant at the target analysis granularity and the situation value of the non-target combatant at the target analysis granularity, the situation difference of the target combatant or the situation difference of the non-target combatant at the target analysis granularity is determined.

13. The method according to claim 12, characterized in that Also includes: A game situation review screen of the game to be reviewed is generated based on the situation difference of the target combatant or the situation difference of the non-target combatant at the target analysis granularity; the game situation review screen displays the win-loss trend from the perspective of the target combatant or the win-loss trend from the perspective of the non-target combatant at the target analysis granularity.

14. The method according to any one of claims 1 to 6, characterized in that The state data of the game unit includes one or more of the following: life attributes, combat attributes, and resource attributes of the game unit.

15. An information processing device for a turn-based game, characterized in that: Applied to a turn-based game, the game includes a first combatant and a second combatant, each of the first combatant and the second combatant includes at least one game unit, the apparatus includes: an acquisition module and a determination module; The acquisition module is used to obtain the status data of each game unit at the target analysis granularity in the game to be reviewed; the target analysis granularity includes: a single round as the minimum analysis unit or a single game unit action as the minimum analysis unit; The determination module is configured to determine, based on the status data of each game unit at the target analysis granularity, a situation value of a target combatant at the target analysis granularity, wherein the target combatant is the first combatant or the second combatant; The determination module is used to determine the situation difference of the target combatant or the situation difference of the non-target combatant based on the situation value of the target combatant at the target analysis granularity, wherein the situation difference of the target combatant is used to characterize the winning and losing trend of the target combatant at the target analysis granularity.

16. An electronic device, characterized in that: include: A processor, a storage medium, and a bus, wherein the storage medium stores program instructions executable by the processor. When the electronic device is running, the processor and the storage medium communicate via the bus, and the processor executes the program instructions to perform the information processing method for a turn-based game as described in any one of claims 1 to 14.

17. A computer-readable storage medium, characterized in that The storage medium stores a computer program, and when the computer program is executed by a processor, the information processing method for a turn-based game according to any one of claims 1 to 14 is executed.

Images