Game content dynamic adjustment method and device, and electronic device

Abstract

The application provides a game content dynamic adjustment method and device and electronic equipment, which collects behavior data and world state data of a game player in multiple different scenes and multiple different dimensions, and then constructs a heterogeneous graph comprising multiple nodes and connected by directed edges. A player tendency vector is constructed based on the heterogeneous graph, and a world context vector is obtained based on the world state data. The triggered target is determined by combining the player tendency vector and the world context vector, and a candidate action combination corresponding to the triggered target is determined. The target action combination with the lowest cost is selected from the candidate action combination, and the game content is dynamically adjusted. In this scheme, multiple scene behaviors are collected to cover multiple dimensions of game interaction of the player, and the player's preference and the state of the whole server are accurately quantified through the player tendency vector and the world context vector. The game content is dynamically adjusted based on the action combination with the lowest cost, ensuring the real-time and collaboration of the adjustment.

Description

Technical Field

[0001] This invention relates to the field of game technology, and more specifically, to a method, apparatus, and electronic device for dynamically adjusting game content. Background Technology

[0002] As the gaming industry evolves towards open-world and highly free-form experiences, player demands have expanded beyond simple combat to include multi-dimensional and complex needs such as building and creating, social collaboration, quest exploration, and economic interaction. Players' demands for personalized and real-time adaptation of game content have significantly increased. Traditional games primarily rely on pre-built scripts or periodic version updates, failing to provide immediate and granular adjustments based on players' real-time behavior. This leads to some players losing engagement due to difficulties that are too high / too low, storylines that don't align with their interests, or rewards that are mismatched.

[0003] Existing technologies are limited to single-dimensional adjustments and have not formed a full-scenario collaborative mechanism, resulting in problems such as incomplete scenario coverage, insufficient quantification accuracy, lack of real-time performance, and insufficient collaboration. Summary of the Invention

[0004] The purpose of this invention is to provide a method, device, and electronic device for dynamically adjusting game content, so as to achieve precise quantification of player preferences and overall server status, and ensure the real-time and collaborative nature of the adjustment.

[0005] In a first aspect, the present invention provides a method for dynamically adjusting game content, the method comprising: Collect behavioral data and world state data of game players in multiple different scenarios and from different dimensions; A heterogeneous graph is constructed based on the behavioral data and world state data. The heterogeneous graph includes multiple nodes, which are connected by directed edges. A player preference vector is constructed based on the heterogeneous graph, and a world context vector is obtained based on the world state data. The triggered target is determined by combining the player preference vector and the world context vector, and the candidate action combination corresponding to the triggered target is determined. Select the target action combination with the lowest cost from the candidate action combinations; The game content is dynamically adjusted based on the target action combination.

[0006] In optional implementations, the scenarios include at least combat scenarios, life scenarios, social scenarios, economic scenarios, and mission scenarios; The nodes include at least player nodes, item nodes, NPC nodes, scene area nodes, quest nodes, currency nodes, and social group nodes; The directed edges between nodes have an edge type attribute and a weight attribute, and the weight attribute is calculated based on the interaction information between nodes represented by the edge type attribute.

[0007] In an optional implementation, the player preference vector includes multiple first-dimensional vectors, and the step of constructing the player preference vector based on the heterogeneous graph includes: For each first-dimensional vector in the player preference vector, determine the node and directed edge corresponding to the first-dimensional vector in the heterogeneous graph, and obtain the weight of the directed edge; By combining the weights of the directed edges with the behavioral data and world state data, the vector value of the first dimension vector is calculated.

[0008] In an optional implementation, the world context vector includes multiple second-dimensional vectors; The step of obtaining the world context vector based on the world state data includes: For each second dimension vector in the world context vector, determine the world state data contained in the world state data required to calculate the second dimension vector; Based on the determined required world state data, the vector value of the second dimension vector is calculated according to preset rules.

[0009] In an optional implementation, the step of determining the triggered target by combining the player preference vector and the world context vector includes: Obtain each target from the preset target library and the corresponding trigger conditions for each target; For each target, the player preference vector and world context vector are combined to determine whether the triggering condition of the target is triggered. If the triggering condition for the target is determined, then the target is identified as the triggered target.

[0010] In an optional implementation, the step of determining the candidate action combination corresponding to the triggered target includes: Obtain multiple actions from the preset action library; Find the action that drives the current state to the corresponding state of the triggered target from the multiple actions; The found actions are combined into candidate action combinations.

[0011] In an optional implementation, the step of forming a combination of candidate actions from the found actions includes: For the identified action, perform a judgment on the constraint conditions containing multiple constraint items; Actions that satisfy the constraints are selected to form a candidate action combination.

[0012] In an optional implementation, the step of selecting the lowest-cost target action combination from the candidate action combinations includes: For each candidate action combination, the corresponding execution cost, effect decay rate, and player acceptance are calculated based on the candidate action combination; The execution cost, effect decay rate, and player acceptance are summed up according to a set weight to obtain the comprehensive cost; The candidate action combination with the lowest overall cost is selected as the target action combination.

[0013] Secondly, the present invention provides a device for dynamically adjusting game content, the device comprising: The data acquisition module is used to collect behavioral data and world state data of game players in multiple different scenarios and from different dimensions. The construction module is used to construct a heterogeneous graph based on the behavioral data and world state data. The heterogeneous graph includes multiple nodes connected by directed edges. The generation module is used to construct a player preference vector based on the heterogeneous graph and obtain a world context vector based on the world state data. The determination module is used to combine the player preference vector and the world context vector to determine the triggered target and to determine the candidate action combination corresponding to the triggered target; The filtering module is used to select the target action combination with the lowest cost from the candidate action combinations; The adjustment module is used to dynamically adjust the game content based on the target action combination.

[0014] Thirdly, the present invention provides an electronic device, comprising: One or more processors; Storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors perform the method as described in any of the foregoing embodiments.

[0015] This invention provides a method, apparatus, and electronic device for dynamically adjusting game content. It collects behavioral data and world state data from players across multiple scenarios and dimensions to construct a heterogeneous graph. This graph includes multiple nodes connected by directed edges. A player preference vector is constructed based on the heterogeneous graph, and a world context vector is obtained based on the world state data. The triggered target is determined by combining the player preference vector and the world context vector, and candidate action combinations corresponding to the triggered target are identified. The lowest-cost target action combination is selected from the candidate action combinations, and the game content is dynamically adjusted based on this target action combination. This solution, through multi-scenario behavioral data collection, covers multi-dimensional player game interactions. By using player preference vectors and world context vectors, it achieves precise quantification of player preferences and overall server status. Dynamic adjustment of game content is performed based on the lowest-cost action combination, ensuring the real-time nature and synergy of the adjustment. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments of the present invention will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 A flowchart of a method for dynamically adjusting game content provided in an embodiment of the present invention; Figure 2 for Figure 1 A flowchart of the sub-steps included in S13; Figure 3 for Figure 1 A flowchart of the sub-steps included in S14; Figure 4 for Figure 1 A flowchart of the sub-steps included in S15; Figure 5 for Figure 1 Another flowchart of the sub-steps included in S15; Figure 6 for Figure 1 A flowchart of the sub-steps included in S16; Figure 7 This is a functional module block diagram of the game content dynamic adjustment device provided in an embodiment of the present invention; Figure 8 This is a structural block diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0018] The technical solutions of the present invention will now be described with reference to the accompanying drawings in the embodiments of the present invention.

[0019] Please see Figure 1 This is a flowchart illustrating the method for dynamically adjusting game content provided in an embodiment of the present invention. It should be understood that in other embodiments, the order of some steps in the method for dynamically adjusting game content in this embodiment can be interchanged according to actual needs, or some steps can be omitted or deleted. The detailed steps of the method for dynamically adjusting game content are described below.

[0020] S11 collects behavioral data and world state data from game players in multiple different scenarios across various dimensions. S12, Construct a heterogeneous graph based on the behavioral data and world state data. The heterogeneous graph includes multiple nodes, which are connected by directed edges. S13, Construct a player preference vector based on the heterogeneous graph; S14, Obtain the world context vector based on the world state data; S15, combine the player preference vector and the world context vector to determine the triggered target, and determine the candidate action combination corresponding to the triggered target; S16, Select the target action combination with the lowest cost from the candidate action combinations; S17, dynamically adjust the game content based on the target action combination.

[0021] In this embodiment, the behavior collection system first collects player behavior data and world state data in different scenarios and from multiple dimensions to provide basic data support for subsequent analysis.

[0022] The scenarios include at least combat scenarios, daily life scenarios, social scenarios, economic scenarios, and mission scenarios. The core data collection dimensions for each scenario are shown in Table 1.

[0023] Table 1 Game Scenarios

[0024] Based on this, data processing is performed through a behavior analysis system, and a heterogeneous graph is constructed based on the collected behavior data and world state data. The heterogeneous graph includes multiple nodes, including at least player nodes, item nodes, NPC nodes, scene area nodes, quest nodes, currency nodes, and social group nodes.

[0025] Table 2 schematically illustrates the nodes in the heterogeneous graph (including node representation, node name, and core attributes). Table 2 Node Types

[0026] In this embodiment, nodes in the heterogeneous graph are connected by directed edges. The directed edges between nodes have edge type attributes and weight attributes. The weight attributes are calculated based on the interaction information between nodes represented by the edge type attributes.

[0027] Specifically, Table 3 lists, for example, the relevant information of directed edges between nodes, including edge type attributes (identifiers), interaction information descriptions, and weight attribute calculation logic.

[0028] Table 3 Edge Types

[0029] Taking the directed edge P→(damage)→N in Table 3 as an example, this directed edge points from node P to node N, representing the interaction information between the two nodes as the damage dealt by the player to the BOSS. The weight attribute of this directed edge can be calculated based on the interaction information according to the following formula: (Single-battle damage / BOSS total HP) × 0.8 + (Damage frequency / Server-wide average frequency) × 0.2; The coefficients in the above calculation logic, such as 0.8 and 0.2, can be set according to requirements and are not limited to this.

[0030] Building upon the above, a player preference vector is constructed based on the heterogeneous graph. This player preference vector comprises multiple first-dimensional vectors; for details, please refer to [link to relevant documentation]. Figure 2 This can be achieved in the following ways: S131, for each first dimension vector in the player preference vector, determine the node and directed edge corresponding to the first dimension vector in the heterogeneous graph, and obtain the weight of the directed edge; S132, combining the weights of the directed edges with the behavioral data and world state data, calculate the vector value of the first dimension vector.

[0031] In this embodiment, player preferences are precisely quantified from multiple dimensions. The first dimension vector in the player preference vector may include vP[frus], vP[combat], vP[explore], vP[build], vP[coope], vP[chat], vP[consume], vP[task], etc., as shown in Table 4.

[0032] Here, vP[frus] represents the frustration index, reflecting factors such as combat failure and mission difficulty. Its quantification value can be calculated based on the relevant nodes and the weights of their directed edges, combined with behavioral data and world state data. For example, it can be calculated as follows: 0.4×(N→kill→P edge weight) + 0.3×(BOSS remaining health percentage) + 0.2×(P→fail→T edge weight) + 0.1×(negative chat percentage); Furthermore, the calculation logic for the other first-dimensional vectors in the player preference vector can be found in Table 4.

[0033] Table 4 Player Preference Vector

[0034] In addition, in this embodiment, a world context vector can be obtained based on world state data to accurately quantify the state of the entire server / scene.

[0035] The world context vector includes multiple second-dimensional vectors; for details, please refer to [link to relevant documentation]. Figure 3 It can be obtained through the following methods: S141, for each second dimension vector in the world context vector, determine the world state data contained in the world state data required to calculate the second dimension vector; S142, based on the determined required world state data, calculate the vector value of the second dimension vector according to preset rules.

[0036] In this embodiment, the second-dimensional vectors in the world context vector include, for example, vW[bossDie], vW[dropItem], vW[mainTaskComplete], vW[branchTaskAccept], vW[groupActive], vW[currencyFlow], vW[itemSupply], and vW[sceneExplore]. The meaning and calculation logic of each second-dimensional vector are shown in Table 5.

[0037] Taking the second-dimensional vector vW[bossDie] as an example, vW[bossDie] represents the BOSS kill rate, reflecting the success rate of challenging the scene's BOSS. The world state data required to calculate vW[bossDie] includes data such as the number of scene BOSS kills and the total number of challenges. Based on this, according to the calculation logic corresponding to the preset rules described below, the vector value corresponding to vW[bossDie] is calculated based on the number of scene BOSS kills and the total number of challenges: (Number of times the scene boss is killed / Total number of challenges) × 1.0; In addition, the definition, calculation logic, etc. of the second-dimensional vectors of other dimensions can be found in Table 5.

[0038] Table 5 World Context Vectors

[0039] Based on the above, the triggered target is determined by combining the player preference vector and the world context vector. For details, please refer to [link to relevant documentation]. Figure 4 This step can be achieved in the following ways: S151, obtain each target in the preset target library and obtain the trigger conditions corresponding to each target; S152, For each target, determine whether to trigger the target's triggering condition by combining the player's tendency vector and the world context vector; S153, if it is determined that the triggering condition of the target is triggered, then the target is determined to be the triggered target.

[0040] In this embodiment, the preset target library includes multiple targets, and the preset target library supports real-time hot updates in the background. Table 6 shows, for example, the targets included in the preset target library, as well as the core description, basic weight, and triggering conditions of each target.

[0041] The triggering conditions for each objective are based on the player's tendency vector and the world context vector. Specifically, the triggering conditions are based on the first dimension vector of the player's tendency vector and the second dimension vector of the world context vector.

[0042] For example, the target KillBoss represents increasing the boss kill rate and reducing player frustration. Its trigger condition is: vW[bossDie] < 0.2 and vP[frus] > 0.6. Therefore, if the calculated vW[bossDie] and vP[frus] satisfy the trigger condition of this target, then this target is considered the triggered target.

[0043] When a target is triggered, its base weight value can be increased accordingly. For example, the base weight of the target KillBoss is increased from 0.15 to 0.35.

[0044] For descriptions of other targets in the preset target library and their triggering conditions, please refer to Table 6.

[0045] Table 6. Targets included in the preset target library

[0046] The aforementioned triggered target represents the desired outcome after adjustment. To achieve this target, relevant actions must be taken. Therefore, it is also necessary to determine the actions corresponding to the triggered target. Actions are often combinations of multiple individual actions. Therefore, in this embodiment, the candidate action combinations corresponding to the triggered target are first determined. Specifically, please refer to [link to relevant documentation]. Figure 5 This step is achieved in the following way: S154, obtain multiple actions from the preset action library; S155, find the action that drives the current state to the state corresponding to the triggered target from the plurality of actions; S156, combine the found actions into candidate action combinations.

[0047] In this embodiment, the preset action library is officially pre-set and can be updated with version updates. The preset action library includes multiple actions, such as generating temporary high-damage items, reducing BOSS attributes, and increasing the drop rate of building materials. Table 7 exemplarily illustrates several actions in the preset action library, along with their core effects and applicable objectives.

[0048] Table 7. Actions included in the preset action library

[0049] Taking the generation of temporary high-damage items during the action as an example, its core effect is to increase the target player's attack power by 50%-100%. It can only be used in the current BOSS battle, and its applicable target is the target KillBoss.

[0050] For players, if the current state needs to be driven to the state corresponding to the triggered target, then relevant actions must be activated. For example, if the triggered target includes KillBoss, the actions to be activated would include generating temporary high-damage items and reducing BOSS attributes. If the triggered target includes PromoteBuild, the actions to be activated would include increasing the drop rate of building materials and generating building speed-up items.

[0051] The triggered target may include one or more, and the corresponding required action may include one or more, often multiple actions. Furthermore, a single triggered target can be achieved through multiple combinations of actions. Therefore, the found actions are combined into candidate action combinations, and each candidate action combination includes one or more actions.

[0052] In this embodiment, when deriving the required action from the triggered target, the GOAP reverse action derivation method can be used. During the derivation process, in order to meet conditions such as actual resource availability, timeliness, and no action conflicts, the candidate action combination needs to be filtered. Specifically, for the found actions, a constraint condition containing multiple constraint items is judged, and actions that meet the constraint conditions are filtered to form a candidate action combination.

[0053] The constraints may include feasibility constraints, timeliness constraints, and superimposed optimal constraints. Feasibility constraints mainly ensure that resources are available, timeliness constraints mainly ensure that the action takes effect within a specified time, and superimposed optimal constraints mainly ensure that there are no conflicts between multiple actions.

[0054] Based on the selection of candidate action combinations, this embodiment selects the target action combination with the lowest cost from the candidate action combinations. Specifically, please refer to... Figure 6 This step is achieved in the following way: S161, For each candidate action combination, calculate the corresponding execution cost, effect decay rate and player acceptance based on the candidate action combination; S162, the execution cost, effect decay rate and player acceptance are summed according to the set weights to obtain the comprehensive cost; S163, select the candidate action combination with the lowest overall cost as the target action combination.

[0055] In this embodiment, multiple indicators are used to evaluate each candidate action combination, and the candidate action combination with the lowest overall cost is selected as the target action combination. Optionally, the A* algorithm can be used to perform optimal selection. Each candidate action combination is evaluated from three dimensions: execution cost, effect decay rate, and player acceptance. Then, the evaluation results of the three dimensions are accumulated according to a set weight, for example, the weights can be 0.6, 0.3, and 0.1 respectively. The overall cost of each candidate action combination is calculated, and the candidate action combination with the lowest overall cost is selected.

[0056] In addition, in this embodiment, actions that disrupt the game's ecosystem (such as infinitely generating rare items or excessively reducing the difficulty of bosses) can be excluded by preset rules, and the optimal target action combination can be transformed into a JSON-formatted structured instruction.

[0057] Finally, the content generation system (content execution layer) parses the structured instructions output by the decision-making system and dynamically adjusts the game content through procedural generation technology, ensuring that the adjustments take effect in real time without waiting for version updates.

[0058] Table 8 provides examples of dynamic adjustment actions for different combinations of actions, along with examples of how these actions take effect.

[0059] Table 8 Dynamic adjustment actions and examples of their effects

[0060] The following examples illustrate the implementation process and effects of the solution provided by this invention in specific scenarios.

[0061] Taking the scenario of "challenging a dragon-shaped BOSS + the side quest of 'finding the lost scroll'" as an example, the implementation process and effects of this invention are fully demonstrated: Step 1: Behavior Data Collection (Behavior data collection for player UID:123 is as follows): 1. Battle scenario: 20 BOSS challenges, 10 deaths, 3500 damage per battle (BOSS's initial HP is 17500, remaining at 80%), lightsaber usage time is 20 minutes (total battle time is 20 minutes), skills are released 20 times, and hit 18 times.

[0062] 2. Task Scenario: The side quest was accepted 5 times, completed 1 time, failed 2 times, and the "Find the Lost Scroll" side quest was not accepted. The response time for accepting the quest is 5 minutes (the quest is open for 30 minutes).

[0063] 3. Life scenario: Collect 20 pieces of wood per hour (average of 15 pieces across the server), build 2 houses (average of 1 house across the server), explore 35% of the area, and discover 1 hidden spot (out of a total of 5 hidden spots in the area).

[0064] 4. Social Scenarios: Join a guild, be active in the group for 60 minutes (total player activity time is 120 minutes), help other players 3 times, and send 10 messages in the group chat (70% of which are positive emotions).

[0065] 5. Economic scenario: Consume 500 gold coins, deposit 800 gold coins (total holding 2000 gold coins), trade items 2 times, and browse the shop for 10 minutes.

[0066] Step Two: Behavior Analysis (Heterogeneous Graph Modeling + Two-Vector Generation): Heterogeneous graph modeling: Generate a heterogeneous graph containing seven types of nodes: P, I, N, A, T, C, and G. The core edge weight calculation results are: N→kill→P=0.67, P→build→A=1.0, P→accept→T=0.28, P→consume→C=0.5, etc.

[0067] Vector generation: Player preference vector: vP[frus=0.72, combat=0.35, explore=0.43, build=0.79, coop=0.75, chat=0.80, consume=0.38, task=0.70].

[0068] World context vector: vW[bossDie=0.1, dropItem=0.4, mainTaskComplete=0.35, branchTaskAccept=0.4, groupActive=0.84, currencyFlow=0.2, itemSupply=0.8, sceneExplore=0.36].

[0069] Step 3: Content Decision (GOAP + A* Algorithm Filtering): Target Triggering: Based on the vP and vW vectors, two main targets are triggered: KillBoss Objective: vW [bossDie] = 0.1 < 0.2 and vP [frus] = 0.72 > 0.6, weight increased to 0.35.

[0070] BoostTaskAccept objective: vW[branchTaskAccept]=0.4<0.5 and vP[task]=0.70>0.5, weight increased to 0.2.

[0071] 1. Action Combination Derivation: The candidate action combination "Generate temporary high-damage items + reduce BOSS attributes + double side quest rewards" is derived using the GOAP algorithm.

[0072] 2.A * Algorithm Screening: Calculate the overall cost of the combination to be 0.32 (lowest) and there is no balance risk, and output a structured decision instruction.

[0073] Step 4: Content Generation 1. Battle scenario: The server pushes the command to generate temporary high-damage weapons (attack power 800) in real time, and at the same time updates the BOSS attributes (attack damage 700, range 7 meters).

[0074] 2. Quest Scenario: The server has updated the configuration of the "Find the Lost Scroll" side quest (1000 experience points, reward: high-level potion) and pushed a "limited-time reward doubled" pop-up to players.

[0075] 3. Effect Feedback: Players who defeat the dragon-shaped BOSS within 5 minutes (significantly reduced from the previous average of 30 minutes) and accept and complete the "Find the Lost Scroll" side quest will send "Finally cleared! The reward is amazing" in the chat (positive emotion percentage increased to 90%).

[0076] The game content dynamic adjustment method provided in this embodiment has at least the following beneficial effects compared with the prior art: Full-scene coverage: For the first time, five major scenes of "combat, life, social, economy and mission" have been coordinated and adjusted, significantly improving the activity of building and social players, and significantly reducing the complaint rate in non-combat scenes.

[0077] High personalization accuracy and segmented preference adaptation: vP vectors can accurately segment player groups, completely solving the "one-size-fits-all" experience problem.

[0078] Real-time response and instant resolution of experience issues: From data collection to content implementation, the entire process can be completed without waiting for version updates, and short-term player experience issues (such as consecutive deaths when challenging a BOSS or getting stuck on a mission) can be addressed instantly.

[0079] Strong balance and controllability, improved ecological stability: vW vector quantification of the entire server status, combined with action cost calculation and balance rule filtering, significantly improves the stability of the game economy and the participation rate of the task system.

[0080] Significant commercial value and optimized core metrics: Simulation tests show that after applying this solution, the weekly player retention rate increases by 25%-32%, the conversion rate of rare items to paid purchases increases by 15%-22%, and the game lifecycle is extended by 15%-20%.

[0081] Based on the same inventive concept, please refer to Figure 7 This invention also provides a functional module diagram of a game content dynamic adjustment device. This embodiment can divide the game content dynamic adjustment device into functional modules based on the above method embodiments. For example, each function can be divided into its own functional modules, or two or more functions can be integrated into one processing module. The integrated modules can be implemented in hardware or as software functional modules. It should be noted that the module division in this invention embodiment is illustrative and only represents one logical functional division; other division methods may be used in actual implementation.

[0082] The game content dynamic adjustment device may include a collection module, a construction module, a generation module, a determination module, a filtering module, and an adjustment module. The functions of each module of the game content dynamic adjustment device will be described in detail below.

[0083] The data acquisition module is used to collect behavioral data and world state data of game players in multiple different scenarios and from different dimensions. The construction module is used to construct a heterogeneous graph based on the behavioral data and world state data. The heterogeneous graph includes multiple nodes connected by directed edges. The generation module is used to construct a player preference vector based on the heterogeneous graph and obtain a world context vector based on the world state data. The determination module is used to combine the player preference vector and the world context vector to determine the triggered target and to determine the candidate action combination corresponding to the triggered target; The filtering module is used to select the target action combination with the lowest cost from the candidate action combinations; The adjustment module is used to dynamically adjust the game content based on the target action combination.

[0084] The game content dynamic adjustment device provided in this embodiment can be used to execute the game content dynamic adjustment method under any of the above embodiments. For details not covered in this embodiment, please refer to the corresponding descriptions in the above embodiments. This embodiment will not be elaborated here.

[0085] Please see Figure 8 This is a structural block diagram of an electronic device provided in an embodiment of the present invention. The electronic device can be, for example, a computer device or a server. The electronic device includes a memory, a processor, and a communication module. The memory, processor, and communication module are electrically connected directly or indirectly to each other to achieve data transmission or interaction. For example, these components can be electrically connected to each other through one or more communication buses or signal lines.

[0086] The memory is used to store computer programs or data. Memory can be, but is not limited to, Random Access Memory (RAM), Read Only Memory (ROM), Programmable Read-Only Memory (PROM), Erasable Programmable Read-Only Memory (EPROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), etc.

[0087] The processor is used to read / write data or programs stored in the memory and execute the game content dynamic adjustment method provided in any embodiment of the present invention.

[0088] The communication module is used to establish communication connections between electronic devices and other communication terminals via a network, and to send and receive data via the network.

[0089] It should be understood that, Figure 8 The structure shown is only a schematic diagram of an electronic device; the electronic device may also include components that are larger than those shown. Figure 8 The more or fewer components shown, or having the same Figure 8 The different configurations shown.

[0090] Furthermore, embodiments of the present invention also provide a computer-readable storage medium storing machine-executable instructions, which, when executed, implement the game content dynamic adjustment method provided in the above embodiments.

[0091] Specifically, the computer-readable storage medium can be a general-purpose storage medium, such as a removable disk or hard disk. When the computer program on the computer-readable storage medium is run, it can execute the aforementioned method for dynamically adjusting game content. The processes involved in the execution of the executable instructions on the computer-readable storage medium can be referred to the relevant descriptions in the above method embodiments, and will not be detailed here.

[0092] In the embodiments provided by this invention, it should be understood that the disclosed apparatus and method can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0093] Furthermore, the units described as separate components may or may not be physically separate. 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 the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0094] Furthermore, the functional modules in the various embodiments of the present invention can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0095] It should be noted that if the functionality is implemented as a software module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0096] In this document, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, without necessarily requiring or implying any such actual relationship or order between these entities or operations.

[0097] The above description is merely an embodiment of the present invention and is not intended to limit the scope of protection of the present invention. For those skilled in the art, the present invention can have various modifications and variations. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for dynamically adjusting game content, characterized in that, The method includes: Collect behavioral data and world state data of game players in multiple different scenarios and from different dimensions; A heterogeneous graph is constructed based on the behavioral data and world state data. The heterogeneous graph includes multiple nodes, which are connected by directed edges. A player preference vector is constructed based on the heterogeneous graph, and a world context vector is obtained based on the world state data. The triggered target is determined by combining the player preference vector and the world context vector, and the candidate action combination corresponding to the triggered target is determined. Select the target action combination with the lowest cost from the candidate action combinations; The game content is dynamically adjusted based on the target action combination.

2. The method for dynamically adjusting game content according to claim 1, characterized in that, The scenarios include at least combat scenarios, life scenarios, social scenarios, economic scenarios, and mission scenarios; The nodes include at least player nodes, item nodes, NPC nodes, scene area nodes, quest nodes, currency nodes, and social group nodes; The directed edges between nodes have an edge type attribute and a weight attribute, and the weight attribute is calculated based on the interaction information between nodes represented by the edge type attribute.

3. The method for dynamically adjusting game content according to claim 1, characterized in that, The player preference vector includes multiple first-dimensional vectors, and the step of constructing the player preference vector based on the heterogeneous graph includes: For each first-dimensional vector in the player preference vector, determine the node and directed edge corresponding to the first-dimensional vector in the heterogeneous graph, and obtain the weight of the directed edge; By combining the weights of the directed edges with the behavioral data and world state data, the vector value of the first dimension vector is calculated.

4. The method for dynamically adjusting game content according to claim 1, characterized in that, The world context vector includes multiple second-dimensional vectors; The step of obtaining the world context vector based on the world state data includes: For each second dimension vector in the world context vector, determine the world state data contained in the world state data required to calculate the second dimension vector; Based on the determined required world state data, the vector value of the second dimension vector is calculated according to preset rules.

5. The method for dynamically adjusting game content according to claim 1, characterized in that, The step of determining the triggered target by combining the player preference vector and the world context vector includes: Obtain each target from the preset target library and the corresponding trigger conditions for each target; For each target, the player preference vector and world context vector are combined to determine whether the triggering condition of the target is triggered. If the triggering condition for the target is determined, then the target is identified as the triggered target.

6. The method for dynamically adjusting game content according to claim 1, characterized in that, The step of determining the candidate action combination corresponding to the triggered target includes: Obtain multiple actions from the preset action library; Find the action that drives the current state to the corresponding state of the triggered target from the multiple actions; The found actions are combined into candidate action combinations.

7. The method for dynamically adjusting game content according to claim 6, characterized in that, The step of forming a candidate action combination from the found actions includes: For the identified action, perform a judgment on the constraint conditions containing multiple constraint items; Actions that satisfy the constraints are selected to form a candidate action combination.

8. The method for dynamically adjusting game content according to claim 1, characterized in that, The step of selecting the lowest-cost target action combination from the candidate action combinations includes: For each candidate action combination, the corresponding execution cost, effect decay rate, and player acceptance are calculated based on the candidate action combination; The execution cost, effect decay rate, and player acceptance are summed up according to a set weight to obtain the comprehensive cost; The candidate action combination with the lowest overall cost is selected as the target action combination.

9. A device for dynamically adjusting game content, characterized in that, The device includes: The data acquisition module is used to collect behavioral data and world state data of game players in multiple different scenarios and from different dimensions. The construction module is used to construct a heterogeneous graph based on the behavioral data and world state data. The heterogeneous graph includes multiple nodes connected by directed edges. The generation module is used to construct a player preference vector based on the heterogeneous graph and obtain a world context vector based on the world state data. The determination module is used to combine the player preference vector and the world context vector to determine the triggered target and to determine the candidate action combination corresponding to the triggered target; The filtering module is used to select the target action combination with the lowest cost from the candidate action combinations; The adjustment module is used to dynamically adjust the game content based on the target action combination.

10. An electronic device, characterized in that, include: One or more processors; Storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any one of claims 1 to 8.

Images