Methods and devices for generating game content

Abstract

This application provides a method and apparatus for generating game content. The method includes: constructing a reference difficulty sequence based on an initial content unit sequence of the game, where each content unit in the initial content unit sequence is the smallest content carrier with an independent logical structure in the game, and the reference difficulty sequence includes reference difficulty values ​​for each content unit that change with key nodes in the game; evaluating the logical difficulty of each content unit in the initial content unit sequence to obtain an evaluated difficulty value for each content unit; determining the difference between the reference difficulty value and the evaluated difficulty value for each content unit, and adjusting each content unit in the initial content unit sequence based on the difference to obtain a target content unit sequence; and generating game content based on the target content unit sequence. This application improves the efficiency of game content generation, thereby increasing system utilization.

Description

Technical Field

[0001] This application relates to the field of game technology, and in particular to a method and apparatus for generating game content. Background Technology

[0002] In the game development process, the design and generation of game content are crucial to the game experience, with the control of difficulty and pacing being a core design element. In related technologies, game developers often rely on their subjective experience or player data to deduce game content. These methods struggle to control the overall game pacing and are heavily influenced by player states. Summary of the Invention

[0003] This application provides a method, apparatus, electronic device, computer-readable storage medium, and computer program product for generating game content, which can control the overall rhythm of the game, improve the efficiency of game content generation, and thus improve system utilization.

[0004] The technical solution of this application embodiment is implemented as follows: This application provides a method for generating game content, the method comprising: A reference difficulty sequence is constructed based on the initial content unit sequence of the game. The content units in the initial content unit sequence are the smallest content carriers with independent logical structures in the game. The reference difficulty sequence includes reference difficulty values ​​for each content unit that change with key nodes in the game. The logical difficulty of each content unit in the initial content unit sequence is evaluated to obtain an evaluated difficulty value for each content unit. The difference between the reference difficulty value and the evaluated difficulty value of each content unit is determined, and each content unit in the initial content unit sequence is adjusted based on the difference to obtain a target content unit sequence. The game content of the game is generated based on the target content unit sequence.

[0005] This application provides a device for generating game content, comprising: A construction module is used to construct a reference difficulty sequence based on the initial content unit sequence of the game. The content unit in the initial content unit sequence is the smallest content carrier with an independent logical structure in the game. The reference difficulty sequence includes the reference difficulty value of each content unit as the game's key nodes change. An evaluation module is used to evaluate the logical difficulty of each content unit in the initial content unit sequence to obtain an evaluation difficulty value for each content unit. An adjustment module is used to determine the difference between the reference difficulty value and the evaluation difficulty value of each content unit, and to adjust each content unit in the initial content unit sequence based on the difference to obtain a target content unit sequence. The generation module is used to generate game content for the game based on the target content unit sequence.

[0006] In the above scheme, the construction module is further configured to construct a difficulty trend map representing the difficulty change trend of the game based on the initial content unit sequence of the game; for each key node in the game, discretize sampling is performed on the difficulty trend map to obtain the difficulty height corresponding to each key node; a reference difficulty value for each key node is determined based on the difficulty height corresponding to each key node, and the reference difficulty values ​​of each key node are combined to obtain the reference difficulty sequence.

[0007] In the above scheme, the construction module is further configured to determine the difficulty gradient corresponding to the difficulty height of each key node in the difficulty trend graph, and to determine the first key node in the game whose difficulty gradient exceeds a preset gradient; adjust the difficulty height corresponding to the first key node in the difficulty trend graph to obtain an adjusted difficulty trend graph; the difficulty gradient corresponding to the adjusted first key node is lower than the preset gradient; and determine a reference difficulty value for each key node based on the difficulty height corresponding to each key node in the adjusted difficulty trend graph.

[0008] In the above scheme, the evaluation module is further configured to extract evaluation elements that can characterize the logical difficulty of each content unit in the initial content unit sequence, wherein the evaluation elements include at least one of the following: mechanism complexity, task chain depth, and fault tolerance space; to score the difficulty of each evaluation element of the content unit to obtain the element score corresponding to each evaluation element; and to fuse the element scores of each evaluation element to obtain the evaluation difficulty value of the content unit.

[0009] In the above scheme, the evaluation module is further used to determine the operation preferences of different player groups based on the operation data of the player groups, and to determine the weight of each evaluation element based on the operation preferences; based on the weight of each evaluation element, the element scores of each evaluation element are weighted and summed to obtain the evaluation difficulty value of the content unit.

[0010] In the above scheme, the evaluation module is further configured to determine the first difference between the reference difficulty value and the evaluation difficulty value of each content unit in the initial content unit sequence as the difference of the content unit; the adjustment module is further configured to adjust the logical difficulty of the content unit in response to the first difference corresponding to the content unit exceeding a first preset difference; or, insert a new content unit before or after the content unit.

[0011] In the above scheme, the adjustment module is further configured to, for adjacent first and second content units in the initial content unit sequence, if the second difference between the first evaluation difficulty value of the first content unit and the second evaluation difficulty value of the second content unit exceeds the second preset difference, insert a third content unit between the first content unit and the second content unit, wherein the third evaluation difficulty of the third content unit is between the first evaluation difficulty and the second evaluation difficulty, so as to generate the target content unit sequence by combining the third content unit.

[0012] In the above scheme, the adjustment module is further configured to detect content units in the target content unit sequence based on a preset optimization mechanism, and determine the content units to be optimized in the target content unit sequence; for each content unit to be optimized, multiple candidate content units are selected in the content unit library based on the reference difficulty value corresponding to the content unit to be optimized; among the multiple candidate content units, a target content unit with a mechanism type different from the first mechanism type of the content unit to be optimized is determined, and the content unit to be optimized is adjusted based on the target content unit.

[0013] In the above scheme, the adjustment module is further configured to cluster multiple candidate content units based on the mechanism type of the candidate content units to obtain at least one cluster; each cluster includes multiple candidate content units belonging to the same mechanism type; determine a second mechanism type of the content unit at the key node preceding the content unit to be optimized and a third mechanism type of the content unit at the key node following the content unit to be optimized; determine a target cluster in the at least one cluster whose mechanism type is different from the first mechanism type, the second mechanism type and the third mechanism type, and determine the target content unit in the candidate content units of the target cluster.

[0014] This application provides an electronic device, the electronic device comprising: Memory is used to store executable instructions or computer programs. The processor, when executing computer-executable instructions or computer programs stored in the memory, implements the game content generation method provided in the embodiments of this application.

[0015] This application provides a computer-readable storage medium storing a computer program or computer-executable instructions, which, when executed by a processor, implements the game content generation method provided in this application.

[0016] This application provides a computer program product, including a computer program or computer executable instructions. When the computer program or computer executable instructions are executed by a processor, they implement the game content generation method provided in this application.

[0017] The embodiments of this application have the following beneficial effects: Through the above embodiments, when generating game content, a reference difficulty sequence is first constructed based on the initial content unit sequence of the game. The reference difficulty sequence includes the reference difficulty value of each content unit in the initial content unit sequence as it changes with key nodes in the game. Then, the logical difficulty of each content unit in the initial content unit sequence is evaluated to obtain the evaluated difficulty value of each content unit. Based on the difference between the reference difficulty value and the evaluated difficulty value of each content unit, the content units are adjusted to obtain the target content unit sequence. This method, which adjusts the content units based on the reference difficulty sequence, can grasp the overall trend of the difficulty change of the game content and control the overall rhythm of the game. During the generation of the target content unit sequence, only the content units that do not meet the trend need to be adjusted, which reduces the frequency of game adjustments, improves the efficiency of game content generation, and thus improves the system utilization. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the architecture of the game content generation system provided in the embodiments of this application; Figure 2 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application; Figure 3 This is a flowchart illustrating the method for generating game content provided in an embodiment of this application; Figure 4 This is a flowchart illustrating the method for constructing a reference difficulty sequence provided in an embodiment of this application; Figure 5 This is a flowchart illustrating the evaluation method provided in the embodiments of this application. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0020] In the following description, references are made to “some embodiments,” which describe a subset of all possible embodiments. However, it is understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0021] In the following description, the terms "first, second, third" are used merely to distinguish similar objects and do not represent a specific ordering of objects. It is understood that "first, second, third" may be interchanged in a specific order or sequence where permitted, so that the embodiments of this application described herein can be implemented in an order other than that illustrated or described herein.

[0022] In the embodiments of this application, the terms "module" or "unit" refer to a computer program or part of a computer program that has a predetermined function and works with other related parts to achieve a predetermined goal, and can be implemented wholly or partially using software, hardware (such as processing circuitry or memory), or a combination thereof. Similarly, a processor (or multiple processors or memory) can be used to implement one or more modules or units. Furthermore, each module or unit can be part of an overall module or unit that includes the functionality of that module or unit.

[0023] Unless otherwise defined, all technical and scientific terms used in the embodiments of this application have the same meaning as commonly understood by one of ordinary skill in the art. The terminology used in the embodiments of this application is for the purpose of describing the embodiments of this application only and is not intended to limit this application.

[0024] In the implementation of this application, the collection and processing of relevant data should strictly comply with the requirements of relevant laws and regulations, obtain the informed consent or separate consent of the personal information subject, and carry out subsequent data use and processing within the scope of laws and regulations and the authorization of the personal information subject.

[0025] Before providing a further detailed description of the embodiments of this application, the nouns and terms involved in the embodiments of this application will be explained, and the nouns and terms involved in the embodiments of this application shall be interpreted as follows.

[0026] 1) Responding to: used to indicate the conditions or states on which the operation is performed depends. When the conditions or states on which it depends are met, one or more operations can be performed in real time or with a set delay. Unless otherwise specified, there is no restriction on the order in which the multiple operations are performed.

[0027] 2) Client, also known as user terminal, refers to the program that provides local services to users in contrast to the server. Except for some applications that can only run locally, it is generally installed on ordinary client machines and needs to work in conjunction with the server. That is, there needs to be a corresponding server and service program in the network to provide the corresponding services. Thus, a specific communication connection needs to be established between the client and the server to ensure the normal operation of the application, such as a game client.

[0028] This application provides a method, apparatus, electronic device, computer-readable storage medium, and computer program product for generating game content, which can control the overall rhythm of the game, improve the efficiency of game content generation, and thus improve system utilization.

[0029] See Figure 1 , Figure 1 This is a schematic diagram of the architecture of the game content generation system 100 provided in this application embodiment. In order to support the generation of a game content application, the terminal 401 connects to the server 200 through the network 300. The network 300 can be a wide area network or a local area network, or a combination of the two.

[0030] In practical applications, the game content generation method of this application embodiment can be executed collaboratively by terminal 401 and server 200. Specifically, terminal 401 can be equipped with a client for game content generation. The user can determine the initial content unit sequence of the game on terminal 401, and construct a reference difficulty sequence based on the initial content unit sequence. The content unit in the initial content unit sequence is the smallest content carrier with an independent logical structure in the game. The reference difficulty sequence includes the reference difficulty value of each content unit as it changes with the key nodes of the game. Then, terminal 401 sends a game content generation request carrying the initial content unit sequence and the reference difficulty sequence to server 200. In response to the game content generation request, server 200 evaluates the logical difficulty of each content unit in the initial content unit sequence to obtain the evaluated difficulty value of each content unit. It determines the difference between the reference difficulty value and the evaluated difficulty value of each content unit, and adjusts each content unit in the initial content unit sequence based on the difference to obtain the target content unit sequence. The game content is generated based on the target content unit sequence and sent to terminal 401 to display the generated game content to the user.

[0031] In practical applications, the game content generation method of this application embodiment can also be executed by either terminal 401 or server 200 alone. Taking server 200 alone as an example, server 200 responds to the user's game content generation instruction, constructs a reference difficulty sequence based on the initial content unit sequence of the game, where each content unit in the initial content unit sequence is the smallest content carrier with an independent logical structure in the game, and the reference difficulty sequence includes the reference difficulty value of each content unit as the game's key nodes change; evaluates the logical difficulty of each content unit in the initial content unit sequence to obtain the evaluated difficulty value of each content unit; determines the difference between the reference difficulty value and the evaluated difficulty value of each content unit, and adjusts each content unit in the initial content unit sequence based on the difference to obtain the target content unit sequence; and generates the game content based on the target content unit sequence.

[0032] The following describes an electronic device that implements the game content generation method provided in the embodiments of this application. The electronic device implementing the game content generation method of the embodiments of this application can be a terminal, a server, or a combination of both. Therefore, the executing entity of each step will not be repeated below. In some embodiments, the terminal can be implemented as a laptop computer, tablet computer, desktop computer, set-top box, smartphone, smart speaker, smartwatch, smart TV, vehicle terminal, or other types of terminals.

[0033] In some embodiments, the server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms. The terminal and server can be connected directly or indirectly via wired or wireless communication, which is not limited in this embodiment.

[0034] See Figure 2 , Figure 2 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Figure 2 The illustrated electronic device 400 includes at least one processor 410, a memory 450, at least one network interface 420, and a user interface 430. The various components in the electronic device 400 are coupled together via a bus system 440. It is understood that the bus system 440 is used to implement communication between these components. In addition to a data bus, the bus system 440 also includes a power bus, a control bus, and a status signal bus. However, for clarity, ... Figure 2 The general labeled all buses as Bus System 440.

[0035] Processor 410 can be an integrated circuit chip with signal processing capabilities, such as a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor can be a microprocessor or any conventional processor, etc.

[0036] User interface 430 includes one or more output devices 431 that enable the presentation of media content, including one or more speakers and / or one or more visual displays. User interface 430 also includes one or more input devices 432, including user interface components that facilitate user input, such as a keyboard, mouse, microphone, touch screen display, camera, other input buttons and controls.

[0037] The memory 450 may be removable, non-removable, or a combination thereof. Exemplary hardware devices include solid-state storage, hard disk drives, optical disk drives, etc. The memory 450 may optionally include one or more storage devices physically located away from the processor 410.

[0038] The memory 450 may include volatile memory or non-volatile memory, or both. The non-volatile memory may be read-only memory (ROM), and the volatile memory may be random access memory (RAM). The memory 450 described in this application embodiment is intended to include any suitable type of memory.

[0039] In some embodiments, memory 450 is capable of storing data to support various operations, examples of which include programs, modules, and data structures or subsets or supersets thereof, as illustrated below.

[0040] Operating system 451 includes system programs for handling various basic system services and performing hardware-related tasks, such as the framework layer, core library layer, driver layer, etc., for implementing various basic business functions and handling hardware-based tasks; The network communication module 452 is used to reach other electronic devices via one or more (wired or wireless) network interfaces 420, exemplary network interfaces 420 including: Bluetooth, WiFi, and Universal Serial Bus (USB), etc. Presentation module 453 is configured to enable the presentation of information (e.g., a user interface for operating peripheral devices and displaying content and information) via one or more output devices 431 associated with user interface 430 (e.g., a display screen, a speaker, etc.). The input processing module 454 is used to detect and translate one or more user inputs or interactions from one or more input devices 432.

[0041] In some embodiments, the game content generation apparatus provided in this application can be implemented in software. Figure 2 A device 455 for generating game content stored in memory 450 is shown. This device can be software in the form of programs and plugins, and includes the following software modules: a construction module 4551, an evaluation module 4552, an adjustment module 4553, and a generation module 4554. These modules are logically connected and can therefore be arbitrarily combined or further separated according to their implemented functions. The functions of each module will be described below.

[0042] In other embodiments, the game content generation apparatus provided in this application can be implemented in hardware. As an example, the game content generation apparatus provided in this application can be a processor in the form of a hardware decoding processor, which is programmed to execute the game content generation method provided in this application. For example, the processor in the form of a hardware decoding processor can be one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs), or other electronic components.

[0043] In some embodiments, the terminal or server can implement the game content generation method provided in this application by running various computer-executable instructions or computer programs. For example, computer-executable instructions can be microprogram-level commands, machine instructions, or software instructions. Computer programs can be native programs or software modules in an operating system; they can be native applications (APPs), i.e., programs that need to be installed in the operating system to run, such as game design APPs; or they can be applets that can be embedded in any APP, i.e., programs that only need to be downloaded to a browser environment to run. In summary, the aforementioned computer-executable instructions can be any form of instruction, and the aforementioned computer programs can be any form of application, module, or plugin.

[0044] The method for generating game content provided in this application will be described below with reference to the accompanying drawings. As mentioned earlier, the electronic device 400 that implements the method for generating game content in this application can be a terminal, a server, or a combination of both. Therefore, the executing entity of each step will not be described again below.

[0045] The method for generating game content in this application embodiment is described using a server as an example. See also... Figure 3 , Figure 3 This is a flowchart illustrating the method for generating game content provided in this application embodiment, which will be combined with... Figure 3 The steps shown are explained.

[0046] In step 101, a reference difficulty sequence is constructed based on the initial content unit sequence of the game.

[0047] Among them, the content units in the initial content unit sequence are the smallest content carriers with independent logical structures in the game, and the reference difficulty sequence includes the reference difficulty values ​​of each content unit as the game's key nodes change.

[0048] Here, a content unit is the smallest content carrier in a game that has an independent logical structure and objective difficulty attributes, and can be quantitatively evaluated and flexibly adjusted by the system. The specific form of the content unit can vary flexibly according to the game type. For example, a content unit can be a single level in a puzzle game, or a specific matching task within a level (such as "eliminating 10 sets of 3-in-1 blocks" or "unlocking 2 obstacle blocks"), or a single decision node or a single interactive challenge in a light strategy game. All of these have the core characteristics of being "independently assessable and flexibly adjustable".

[0049] In practical applications, based on the core gameplay, multiple minimal content carriers with independent logical structures are pre-constructed. Each minimal content carrier corresponds to an indivisible independent interactive task in the game (i.e., a single task contains only one complete logical loop, without additional splitting dimensions). These are arranged in the natural progression of the game to form an initial content unit sequence. Alternatively, a preset number of content units can be randomly selected from a content unit library built for the game and combined to obtain the initial content unit sequence. Or, in response to the user's creation operation of the initial content unit sequence, the initial content unit sequence created by the user (including selecting content units and sorting the selected content units) is determined.

[0050] After determining the initial content unit sequence, quantifiable progression nodes (such as level numbers) in the game process are used as key nodes. The order of the key nodes is consistent with the arrangement order of the initial content unit sequence to ensure that each key node corresponds to a content unit in the initial content unit sequence.

[0051] In some embodiments, Figure 4 This is a flowchart illustrating the method for constructing a reference difficulty sequence provided in an embodiment of this application. See also... Figure 4 Step 101 can be achieved through steps 1011 to 1013.

[0052] In step 1011, a difficulty trend map is constructed based on the initial content unit sequence of the game to characterize the difficulty change trend of the game.

[0053] In practical applications, confirm the order of each content unit in the initial content unit sequence (consistent with the natural progression of the game), as well as the key game nodes (such as level numbers) corresponding to each content unit, to ensure that the content units and key nodes correspond one-to-one; and provide fixed anchor points for drawing the difficulty trend chart.

[0054] In practical applications, a difficulty trend chart can be constructed using any of the following methods: Method 1 (Users can draw freely): Users do not need to refer to the objective logical attributes of the content unit. They can freely set the direction of difficulty change (such as increasing, decreasing, fluctuating, random change, etc.) according to their subjective wishes. They only need to ensure that each key node corresponds to a difficulty position in the difficulty trend chart, and that the difficulty trend chart is a continuous curve (without breaks). The core requirement is to "represent the trend of difficulty change".

[0055] Method 2 (Drawing according to the rhythm requirements of the game): First, clarify the core rhythm objectives of the game (such as "low difficulty in the tutorial period → peak in the mid-term → buffer in the late-term", "periodic fluctuation and increase" and other rhythm requirements), and then combine the independent logical structure of each content unit (such as mechanism complexity, target requirements) to set the relative difficulty relationship of each key node (such as "the difficulty of key nodes in the tutorial period gradually increases" and "the difficulty of peak nodes reaches the peak") to ensure that the trend chart fits the game's preset rhythm experience.

[0056] Finally, draw a continuous difficulty trend chart, using the key game nodes as the horizontal axis (arranged in the order of progression) and the difficulty level as the vertical axis (only showing the relative high and low relationship, without quantifying specific values). According to the selected drawing method, connect the difficulty positions corresponding to each key node to form a complete continuous curve. The curve should cover all key nodes, clearly represent the changing trend of game difficulty as key nodes progress, and always maintain a one-to-one correspondence with the key nodes of the initial content unit sequence.

[0057] In step 1012, for each key node in the game, discretization sampling is performed on the difficulty trend map to obtain the difficulty level corresponding to each key node.

[0058] In practical applications, using the horizontal axis of the game's key nodes on the difficulty trend chart as a benchmark, each key node is mapped to a fixed coordinate on the horizontal axis according to the natural progression order of the key nodes (e.g., the first key node corresponds to horizontal axis coordinate "1", the second to "2"), ensuring that the coordinates of each key node are unique and the order is unbiased. For the horizontal axis coordinate of each key node, the vertical axis is extended upwards until it intersects with the continuous curve of the difficulty trend chart, and the vertical axis value of the intersection point is obtained. This vertical axis value is the difficulty height corresponding to the current key node. The sampling process does not depend on the way the trend chart is drawn (whether it is drawn arbitrarily by the user or drawn according to rhythm requirements), but is based solely on the objective position of the curve. According to the progression order of the key nodes, the name, horizontal axis coordinate, and corresponding difficulty height of each node are recorded in sequence, forming a one-to-one correspondence between key nodes and difficulty heights, ensuring no omissions or duplications, and providing complete data support for the subsequent determination of reference difficulty values.

[0059] In step 1013, a reference difficulty value for each key node is determined based on the difficulty level corresponding to each key node, and the reference difficulty values ​​of each key node are combined to obtain a reference difficulty sequence.

[0060] In practical applications, a fixed correspondence standard between difficulty level and reference difficulty value can be pre-defined (e.g., the numerical value of difficulty level is directly equivalent to the reference difficulty value, or the difficulty level is mapped according to the intervals 0-3→1-3, 4-6→4-6, 7-10→7-10). The mapping rule must be consistent throughout and not change with the way the difficulty trend chart is drawn. Then, according to the established mapping rule, the correspondence between key nodes and difficulty levels is recorded for each difficulty level, and converted into a specific, quantifiable reference difficulty value, ensuring that each key node has a unique corresponding reference difficulty value, and that the numerical format is consistent (e.g., all are integers).

[0061] Then, following the natural progression of the game's key nodes, the reference difficulty values ​​of all key nodes are arranged sequentially to form a complete reference difficulty sequence. The reference difficulty sequence must maintain consistency with the initial content unit sequence and the order of key nodes, ensuring that each reference difficulty value can accurately correspond to a content unit in the initial content unit sequence.

[0062] By using the above methods and drawing difficulty trend charts, the overall rhythm of the game can be visually presented through a graphical curve. Whether drawn arbitrarily according to subjective intentions or tailored to the game's preset rhythm requirements, abstract rhythm concepts can be transformed into continuous and clear trends in difficulty changes. This helps users quickly identify core rhythm zones such as the "introduction phase," "peak phase," and "buffer phase," avoiding chaotic rhythm design. Furthermore, the complete closed loop of rhythm visualization → node precision (determining the difficulty level of each key node) → benchmark quantification (determining reference difficulty values) allows users to control the entire process of rhythm from conception to implementation, whether designing the rhythm subjectively or planning it according to game requirements. This ensures that the final reference difficulty sequence accurately reflects the game's overall rhythmic intent, preventing the rhythm from becoming disconnected from the actual content.

[0063] In some embodiments, after obtaining the difficulty height corresponding to each key node in step 1012, the difficulty gradient corresponding to the difficulty height of each key node in the difficulty trend chart can be determined, and the first key node whose difficulty gradient exceeds the preset gradient can be identified among the key nodes of the game; the difficulty height corresponding to the first key node in the difficulty trend chart is adjusted to obtain an adjusted difficulty trend chart; the difficulty gradient corresponding to the first key node after adjustment is lower than the preset gradient; accordingly, the "determining the reference difficulty value of each key node based on the difficulty height corresponding to each key node" in step 1013 can be implemented in the following way: based on the difficulty height corresponding to each key node in the adjusted difficulty trend chart, the reference difficulty value of each key node is determined.

[0064] In practical applications, based on the difficulty heights corresponding to each key node obtained in step 1012, the difference in difficulty height between each key node and its preceding (or following) key node is calculated according to the natural progression order of the key nodes. This difference is the difficulty gradient corresponding to the current key node. Setting a preset gradient and selecting the first key node: A preset gradient (e.g., value 2) is pre-set to regulate the range of difficulty changes. The difficulty gradient of each key node is compared with the preset gradient, and key nodes whose difficulty gradient exceeds the preset gradient are selected as the first key nodes. Adjusting the difficulty height of the first key node: For each first key node, its corresponding difficulty height is adjusted in the difficulty trend graph. The adjustment direction must conform to the overall difficulty change trend (without disrupting the core rhythm), and the adjustment range is based on the standard that "the difficulty gradient of the first key node after adjustment is lower than the preset gradient," ensuring a smooth transition in difficulty between adjacent nodes. Generating the adjusted difficulty trend graph: The difficulty heights of all key nodes (including the adjusted first key node) are reconnected sequentially to form a continuous and smooth adjusted difficulty trend graph.

[0065] Correspondingly, based on the difficulty height corresponding to each key node in the adjusted difficulty trend chart, the reference difficulty value of each key node can be determined according to a unified mapping rule, and then the reference difficulty sequence can be obtained by combining the key nodes in order.

[0066] By employing the methods described above, the difficulty gradient calculation and the selection of the first key node can accurately identify "abrupt nodes" in the difficulty trend chart. This helps users promptly detect rhythmic breaks between key nodes (such as significant increases or decreases in difficulty), preventing unbalanced player experience due to loss of rhythm control. Adjusting the difficulty level of the first key node transforms the overall rhythm of the difficulty trend chart from "potentially abrupt" to "smooth and consistent," ensuring that the difficulty transitions between key nodes conform to the preset rhythm expectations. Users can more clearly grasp the rhythmic flow, avoiding chaotic rhythm design and enhancing the overall rhythm controllability. This allows users to freely design the rhythm while ensuring its reasonable implementation through gradient adjustments, comprehensively improving the precision of their control over the overall game rhythm.

[0067] See also Figure 3 The following explanation follows step 101 above.

[0068] In step 102, the logical difficulty of each content unit in the initial content unit sequence is evaluated to obtain the evaluation difficulty value of each content unit.

[0069] In some embodiments, Figure 5 This is a flowchart illustrating the evaluation method provided in the embodiments of this application. See also... Figure 5 Step 102 can be achieved through steps 1021 to 1023.

[0070] In step 1021, for each content unit in the initial content unit sequence, an evaluation element that can characterize the logical difficulty of the content unit is extracted. The evaluation element includes at least one of the following: mechanism complexity, task chain depth, and fault tolerance space.

[0071] Here, the logical difficulty of a content unit represents the difficulty characteristics determined by its own objective logical attributes as the smallest content carrier with an independent logical structure. This logical difficulty is unrelated to subjective performance data such as the player's skill level and familiarity with the game, and focuses only on the internal logical design of the content unit itself.

[0072] In practical applications, the logical difficulty of a content unit can represent at least three aspects: first, the complexity of the rules and interaction mechanisms contained in the content unit; second, the path characteristics such as the length of the logical process and the number of points required for the player to complete the task objective of the content unit; and third, the tolerance of the content unit for player errors (i.e., fault tolerance conditions). Therefore, based on the three aspects of information that logical difficulty can represent, at least three evaluation elements can be extracted: mechanism complexity, task chain depth, and fault tolerance space.

[0073] Specifically, for each content unit in the initial content unit sequence, the core task objective, internal rules, and interaction logic of the content unit are defined; based on the logical structure of the content unit (i.e., core task objective, internal rules, and interaction logic), at least one of the following is extracted: mechanism complexity, task chain depth, and fault tolerance space of the content unit.

[0074] Specifically, regarding the complexity of the extracted content units, the focus is on the core rules and objective characteristics of the interactive mechanisms contained in the content unit (such as whether it contains basic matching rules, whether it has additional mechanisms such as obstacle unlocking or item triggering); regarding the depth of the extracted task chain, the focus is on the logical process characteristics that players need to follow to complete the task objectives of the content unit (such as whether it is a single process, whether it contains multi-step dependencies or branch selections); regarding the fault tolerance space extracted from the content unit, the focus is on the degree of tolerance of the content unit for player operation errors (such as whether there are time or number of times limits, whether errors can be recovered or retried).

[0075] In step 1022, a difficulty score is assigned to each evaluation element of the content unit to obtain the element score corresponding to each evaluation element.

[0076] In practical applications, objective scoring rules on a scale of 0 to 10 can be formulated for the three evaluation elements: mechanism complexity, task chain depth, and fault tolerance space (the higher the score, the higher the difficulty of the element). The rules are consistent and do not need to be adjusted throughout the process.

[0077] As an example, the scoring rules for mechanism complexity are as follows: 0-3 points = only basic core rules (no additional mechanisms); 4-6 points = basic rules + 1 to 2 simple additional mechanisms; 7-10 points = basic rules + 3 or more additional mechanisms or high-complexity mechanisms. The scoring rules for task chain depth are as follows: 0-3 points = single logical flow (no branches, no step dependencies); 4-6 points = 2 to 3 consecutive logical steps (no branches); 7-10 points = 4 or more logical steps or containing 2 or more branch choices. The scoring rules for fault tolerance are as follows: 0-3 points = extremely small fault tolerance (strict time / attempt limits, failure means instant death); 4-6 points = medium fault tolerance (some limitations, failures can be partially compensated for); 7-10 points = extremely large fault tolerance (no time or attempt limits, repeated attempts are possible).

[0078] Subsequently, in accordance with the above scoring rules, each evaluation element of each content unit is scored based on its objective logical characteristics to obtain the element score corresponding to each evaluation element. Here, subjective performance data of players (such as operational proficiency) is not introduced to ensure that the scoring results are objective and consistent.

[0079] In step 1023, the element scores of each evaluation element are merged to obtain the evaluation difficulty value of the content unit.

[0080] In practical applications, based on pre-defined fusion standards for element scores (such as the weighted average method), the weight of each evaluation element is clearly defined (the sum of weights is 1), ensuring that all content units are fused according to the same rule. Then, the evaluation difficulty value of a single content unit is calculated: according to the unified fusion rule, all element scores of each content unit are substituted into the calculation (e.g., the weighted average formula: evaluation difficulty value = mechanism complexity score × weight 1 + task chain depth score × weight 2 + fault tolerance space score × weight 3). The calculation result retains a unified format (e.g., rounded to the nearest integer) to obtain the evaluation difficulty value of each content unit.

[0081] By employing the above methods, core evaluation elements such as the mechanism complexity, task chain depth, and fault tolerance of content units are extracted in a targeted manner. This precisely focuses on the independent logical structure of each content unit, ensuring that the extracted elements objectively represent the logical difficulty. This provides a targeted foundation for subsequent scoring and integration, preventing the evaluation from deviating from the essence of the content. Furthermore, the scoring process is based solely on the objective logical attributes of the content units, independent of players' subjective performance data. This effectively reduces scoring bias and ensures that the element scores of each content unit are comparable. Consequently, it provides a reliable objective basis for subsequent difficulty deviation correction and target difficulty curve fitting, ensuring that difficulty adjustments always align with the gameplay logic.

[0082] In some embodiments, step 1023 can be implemented as follows: for different player groups, determine the player group's operation preferences based on the player group's operation data, and determine the weight of each evaluation element based on the operation preferences; based on the weight of each evaluation element, perform a weighted summation of the element scores of each evaluation element to obtain the evaluation difficulty value of the content unit. Correspondingly, step 104 can be implemented as follows: for different player groups, generate game content for the player group based on the target content unit sequence corresponding to the player group.

[0083] In practical applications, players can be divided into several independent player groups (such as novice group, intermediate group, and veteran group) based on the characteristics of the core audience of the game (such as operation proficiency, play time, and challenge preference) to ensure that the operation characteristics of players in the same group are similar. Collect player operation data: For each player group, collect their operation data in similar games or during game testing phases (such as mechanic adaptation error rate, task chain completion efficiency, feedback on fault tolerance requirements, etc.). The data can cover scenarios related to mechanic complexity, task chain depth, and fault tolerance space. Analyze and determine player group operation preferences: Based on the collected operation data, extract the operation preferences of each player group. For example, novice players have a high error rate when the mechanics are complex and prefer simple mechanics with high fault tolerance, while experienced players are good at long task chains and low fault tolerance challenges and prefer high-complexity mechanics. Customize evaluation element weights: According to the player group's operation preferences, assign exclusive weights (total weights are 1) to the player group for mechanic complexity, task chain depth, and fault tolerance space, ensuring that the weights are strongly bound to the preferences (e.g., for a group that prefers simple mechanics, the weight of mechanic complexity is higher than other elements). Finally, calculate the evaluation difficulty value by weighted summation: Multiply the element score of each evaluation element in each content unit by the element weight of the corresponding player group and sum them to obtain the evaluation difficulty value of that content unit for that group. The calculation result retains a uniform format (e.g., rounded to the nearest integer).

[0084] Accordingly, this method can determine the evaluation difficulty value sequence corresponding to each player group, and ultimately obtain the target content sequence corresponding to each player group. Therefore, when generating game content in step 104, different game content can be generated for different player groups. Specifically, for each player group, game content exclusive to that player group is generated based on the target content sequence corresponding to that player group. The content units contained in this game content are tailored to the operational preferences of that player group.

[0085] By employing the above methods, operational preferences are determined based on player group operation data, and then the weights of evaluation elements are customized. This makes the evaluation difficulty value more aligned with the actual operational needs of different groups, avoiding a one-size-fits-all difficulty assessment and improving the accuracy and relevance of the evaluation results. The weighted summation fusion method retains the core difficulty characteristics of "mechanism complexity, task chain depth, and fault tolerance," while highlighting differences in operational preferences through group-specific weights. This allows the evaluation difficulty value to accurately reflect the group's perception of the logical difficulty of content units, improving the efficiency of game content generation, system utilization, and operational efficiency. Finally, exclusive game content is generated for different player groups, achieving precise matching between game content and player operational preferences. This prevents new players from giving up due to excessive difficulty and experienced players from leaving due to insufficient difficulty, comprehensively improving the gaming experience for all groups.

[0086] See also Figure 3 The following explanation follows step 102 above.

[0087] In step 103, the difference between the reference difficulty value and the evaluation difficulty value of each content unit is determined, and each content unit in the initial content unit sequence is adjusted based on the difference to obtain the target content unit sequence.

[0088] In some embodiments, the step 103 of "determining the difference between the reference difficulty value and the evaluation difficulty value of each content unit" can be implemented as follows: for each content unit in the initial content unit sequence, the first difference between the reference difficulty value and the evaluation difficulty value of the content unit is determined as the difference of the content unit. Accordingly, the step 103 of "adjusting each content unit in the initial content unit sequence based on the difference" can be implemented as follows: in response to the first difference corresponding to the content unit exceeding a first preset difference, the logical difficulty of the content unit is adjusted; or, a new content unit is inserted before or after the content unit.

[0089] In practical applications, the basic data correspondence is first clarified: the reference difficulty sequence extracted in step 1013 (including the reference difficulty value corresponding to each content unit) and the evaluation difficulty value sequence obtained in step 1023 (including the evaluation difficulty value corresponding to each content unit) are obtained. These two sequences are arranged in the same order as the initial content unit sequence, and the reference difficulty value and evaluation difficulty value of each content unit correspond one-to-one. Then, for each content unit in the initial content unit sequence, the difference between its corresponding reference difficulty value and evaluation difficulty value is calculated (the calculation logic can be unified as "first difference = reference difficulty value - evaluation difficulty value"), and the first difference of the content unit is obtained as an indicator representing the difference between the two. In this way, the first difference of each content unit in the initial content unit sequence is determined.

[0090] A first preset difference (e.g., ±2, the value can be flexibly set according to the needs of the game rhythm) is preset to determine whether adjustment is needed. The judgment criteria for "the first difference exceeds the first preset difference" are clearly defined (i.e., the first difference > the upper limit of the first preset difference or the first difference < the lower limit of the first preset difference). Then, the first difference is judged one by one to determine whether it exceeds the standard. That is, the first difference of each content unit is compared with the first preset difference, and the two cases of "the first difference does not exceed the first preset difference" and "the first difference exceeds the first preset difference" are distinguished. For the content unit whose first difference exceeds the first preset difference, it is adjusted by any one of the following two methods.

[0091] Method 1: In response to the first difference exceeding the first preset difference, adjust the logical difficulty of the content unit. The adjustment direction should be adapted to the characteristics of the first difference (e.g., when the reference difficulty value is greater than the evaluation difficulty value, increase the logical difficulty of the content unit; when the reference difficulty value is less than the evaluation difficulty value, decrease the logical difficulty of the content unit). The adjustment range should aim to ensure that the adjusted first difference does not exceed the first preset difference, so as to ensure that the adjusted logical difficulty still fits the independent logical structure of the content unit.

[0092] Method 2: In response to the first difference exceeding the first preset difference, insert a new content unit before or after the content unit. The new content unit must have an independent logical structure, and its logical difficulty must match the requirement of "making up for the first difference" (e.g., when the reference difficulty value of the original content unit is less than the evaluation difficulty value and the first difference exceeds the standard, insert a new content unit with lower logical difficulty). The insertion position must not disrupt the core progression order of the initial content unit sequence.

[0093] Finally, after adjusting all content units whose first difference exceeds the first preset difference (adjusting the logic difficulty or inserting new content units), organize all content units (including content units that have not been adjusted, adjusted content units, and newly added content units) to form the target content unit sequence.

[0094] Through the above methods, two flexible adjustment methods are provided for the excessive difference: "adjusting the logical difficulty" and "inserting new content units". The appropriate solution can be selected according to the logical characteristics of the content unit and the needs of the game rhythm. This ensures that the difficulty deviation correction effect is achieved without destroying the independent logical structure of the content unit. Moreover, the adjustment process always revolves around the core objective of "the first difference does not exceed the first preset difference", ensuring that the trend of the change of the reference difficulty value of each content unit in the final target content unit sequence is highly consistent with the trend of the change of the evaluation difficulty value, accurately carrying the overall rhythm intention of the game.

[0095] In some embodiments, before step 103 is executed, the initial content unit sequence can be adjusted in the following way: for adjacent first content units and second content units in the initial content unit sequence, if the second difference between the first evaluation difficulty value of the first content unit and the second evaluation difficulty value of the second content unit exceeds the second preset difference, a third content unit is inserted between the first content unit and the second content unit, and the third evaluation difficulty of the third content unit is between the first evaluation difficulty and the second evaluation difficulty, so as to combine the third content unit to generate the target content unit sequence.

[0096] In practical applications, the initial content unit sequence is traversed, and each pair of adjacent content units (denoted as the first content unit and the second content unit) is identified one by one. The first evaluation difficulty value corresponding to the first content unit and the second evaluation difficulty value corresponding to the second content unit are extracted to ensure that the evaluation difficulty value corresponds one-to-one with the content unit. For each pair of adjacent first and second content units, the second difference is calculated (the calculation logic is uniformly "second difference = |first evaluation difficulty value - second evaluation difficulty value|"). At the same time, a second preset difference is preset (the value is set according to the game rhythm requirements and is used to determine whether the difficulty transition between adjacent units is abrupt). Then, the second difference of each pair of adjacent first and second content units is compared with the second preset difference. If the second difference exceeds the second preset difference, it is determined that a third content unit needs to be inserted; if it does not exceed, the original order of adjacent units is retained without adjustment.

[0097] For adjacent unit groups that are judged to be out of standard (i.e., the second difference exceeds the second preset difference), a third content unit with an independent logical structure is obtained from the content unit library, ensuring that the third evaluation difficulty value of the third content unit is between the first evaluation difficulty value and the second evaluation difficulty value (i.e., greater than the smaller of the two and less than the larger of the two); the third content unit is inserted between the first content unit and the second content unit, and the core progression order of the initial sequence remains unchanged after insertion; after completing the insertion operation of the third content unit for all out-of-standard adjacent unit groups, the initial content unit sequence (including the inserted third content unit) is rearranged according to the natural progression order of the game process to obtain the initial content unit sequence after the first rearrangement, and then the target content unit sequence is generated through step 103.

[0098] Here, for all inserted third content units, a reference difficulty value corresponding to the third content unit is inserted at the node position corresponding to the third content unit in the reference difficulty sequence. This reference difficulty value is located between the preceding and following reference difficulty values.

[0099] By focusing on the difference in difficulty values ​​between adjacent content units, the system accurately identifies abrupt transitions in difficulty within the initial sequence, preventing gaps in player experience due to large differences in difficulty between adjacent units and ensuring the continuity of the game's rhythm. The inserted third content unit has an evaluation difficulty value between the evaluation difficulty values ​​of the two adjacent units, effectively buffering the difficulty jumps between adjacent units and making the difficulty change trend of the target content unit sequence smoother, conforming to players' perception habits of difficulty progression. Furthermore, the judgment criteria for the second difference (the second preset difference) can be flexibly set to adapt to the rhythm design needs of different types of games, improving the versatility and adaptability of the adjustment method.

[0100] In some embodiments, after obtaining the target content unit sequence, the target content unit sequence can be optimized and adjusted in the following ways: the content units in the target content unit sequence are detected based on a preset optimization mechanism to determine the content units to be optimized in the target content unit sequence; for each content unit to be optimized, multiple candidate content units are selected in the content unit library based on the reference difficulty value corresponding to the content unit to be optimized; among the multiple candidate content units, a target content unit with a mechanism type different from the first mechanism type of the content unit to be optimized is determined, and the content unit to be optimized is adjusted based on the target content unit.

[0101] In practical applications, optimization mechanisms are pre-defined (e.g., two or more consecutive content units have the same mechanism type, the proportion of a single content unit's mechanism type exceeds 50% of the total number of types in the sequence, and the content unit's logical difficulty deviates from the reference difficulty value twice). All content units in the target content unit sequence are traversed, and each is tested according to the pre-defined optimization mechanism. Content units that do not meet the optimization mechanism's requirements are selected and identified as content units to be optimized. The testing process must ensure coverage of all content units in the target sequence without omissions. For each content unit to be optimized, its corresponding reference difficulty value is extracted. Using this reference difficulty value as the core benchmark (e.g., allowing a fluctuation range of ±1), multiple candidate content units whose evaluation difficulty values ​​meet the benchmark requirements are selected from a pre-defined content unit library. The content unit library must contain content units with different mechanism types and independent logical structures, and each unit is labeled with its evaluation difficulty value and mechanism type to ensure diversity in the selection results. Next, the first mechanism type (i.e., its core interaction / rule mechanism type) of the content unit to be optimized is extracted. From the multiple candidate content units selected, content units with mechanism types different from the first mechanism type are selected and identified as target content units for adjustment. For the content unit to be optimized, adjustments are made by "replacing" or "supplementing" (replacing: directly replacing the content unit to be optimized with the target content unit; supplementing: adding target content units before and after the content unit to be optimized), to obtain the final target content unit sequence. The overall trend of the reference difficulty value of the final target content unit sequence remains unchanged.

[0102] In some embodiments, the above-mentioned "determining a target content unit whose mechanism type is different from the first mechanism type of the content unit to be optimized" can be implemented in the following way: based on the mechanism type of the candidate content units, clustering multiple candidate content units to obtain at least one cluster; each cluster includes multiple candidate content units belonging to the same mechanism type; determining the second mechanism type of the content unit at the previous key node of the content unit to be optimized and the third mechanism type of the content unit at the next key node of the content unit to be optimized; determining a target cluster in the at least one cluster whose mechanism type is different from the first mechanism type, the second mechanism type and the third mechanism type, and determining the target content unit in the candidate content units of the target cluster.

[0103] In practical applications, the mechanism types of all selected candidate content units are extracted. Using "mechanism type" as the clustering dimension, multiple candidate content units are clustered using a merging method to obtain at least one cluster. Each cluster contains only multiple candidate content units belonging to the same mechanism type, and different clusters correspond to different mechanism types, ensuring no overlap in mechanism types between clusters. The key node positions of the content unit to be optimized within the target content unit sequence are located, and the mechanism types of the content units at the key node preceding that key node (denoted as the second mechanism type) and the content units at the key node following that key node (denoted as the third mechanism type) are extracted.

[0104] Iterate through all clusters, verify the mechanism type of each cluster, and select clusters whose mechanism type is different from both the first mechanism type, the second mechanism type, and the third mechanism type of the content unit to be optimized. These clusters are then identified as target clusters. If multiple target clusters meet the criteria, prioritize the cluster with the largest number of candidate content units and the smallest deviation between the reference difficulty value and the unit to be optimized. Finally, among the candidate content units contained in the target clusters, select the candidate content unit with the highest good fit between its reference difficulty value and the reference difficulty value corresponding to the content unit to be optimized (e.g., difference ≤ 1). This candidate content unit is then identified as the target content unit for adjusting the content unit to be optimized. If multiple candidate units with consistent fit exist within the target cluster, they can be selected randomly or according to a preset priority (e.g., mechanism complexity adaptability).

[0105] By clustering candidate units based on mechanism type, the above methods enable the classification and management of candidate units, allowing for the rapid differentiation of candidate units with different mechanism types and improving the efficiency and accuracy of target cluster selection. When selecting target clusters, the method avoids the mechanism types of the unit to be optimized itself, its preceding nodes, and its succeeding nodes, effectively preventing the continuous occurrence of the same mechanism type in the target content unit sequence, significantly increasing the diversity of mechanism types in the sequence, and reducing player fatigue caused by repetitive mechanisms. The selection logic of target clusters takes into account the mechanism types of the nodes before and after the unit to be optimized, ensuring that the adjusted content unit is different from its own mechanism type and also differentiated from the mechanism types of adjacent nodes, maintaining the continuity and richness of mechanism changes in the target content unit sequence. The candidate unit with the highest fit to the reference difficulty value is selected from the target clusters, ensuring the differentiation of mechanism types while not deviating from the reference difficulty value requirements of the unit to be optimized, ensuring that the difficulty rhythm of the target content unit sequence is not affected by mechanism adjustments.

[0106] See also Figure 3 The following explanation follows step 103 above.

[0107] In step 104, game content is generated based on the target content unit sequence.

[0108] In practical applications, each content unit within the target content unit can be transformed into a concrete, executable game content module. Each module corresponds to a complete piece of game interaction content, and must meet the requirements of being "independently runnable and fitting the attributes of the corresponding content unit," while also adapting to the game's basic gameplay framework to ensure that the module and the overall gameplay have no logical conflicts. Following the sequence of the target content units, the game content modules corresponding to each content unit are sequentially linked and integrated. During the linking process, the transition logic between modules is connected to ensure smooth progress between modules, resulting in the specific game content.

[0109] Through the above embodiments, when generating game content, a reference difficulty sequence is first constructed based on the initial content unit sequence of the game. The reference difficulty sequence includes the reference difficulty value of each content unit in the initial content unit sequence as it changes with key nodes in the game. Then, the logical difficulty of each content unit in the initial content unit sequence is evaluated to obtain the evaluated difficulty value of each content unit. Based on the difference between the reference difficulty value and the evaluated difficulty value of each content unit, the content units are adjusted to obtain the target content unit sequence. This method, which adjusts the content units based on the reference difficulty sequence, can grasp the overall trend of the difficulty change of the game content and control the overall rhythm of the game. During the generation of the target content unit sequence, only the content units that do not meet the trend need to be adjusted, which reduces the frequency of game adjustments, improves the efficiency of game content generation, and thus improves the system utilization.

[0110] The following will describe an exemplary application of the embodiments of this application in a real-world application scenario. First, the terms used in the embodiments of this application will be explained, including: 1. Machine Learning (ML) is a technique that learns patterns from historical data to predict or classify future data. In this application, machine learning is used to train a difficulty feature model, such as learning indicators like player failure rate, completion time, and cognitive load level under different combinations of mechanisms, in order to optimize the parameters of the fitting algorithm and improve the accuracy and adaptability of the generated content in matching the expected difficulty curve.

[0111] 2. Game Difficulty Curve (GDC) is a graph that describes the trend of challenge intensity changing over time or at different stages of the game. It is usually represented by a continuous function curve. The difficulty curve can be set as a target reference curve (such as an increasing curve, bell curve, or periodic oscillation curve) to guide the system in dynamically generating or adjusting content to achieve game pacing control.

[0112] 3. Difficulty Oscillation refers to the non-linear, periodic, or reactive patterns of difficulty variation that occur between different points in time or between different tasks during gameplay. For example, a series of relatively easy game tasks might be followed by a sudden, high-pressure battle to create a "rhythm climax," which is then tempered by a "relief phase." Difficulty oscillation is considered an important design mechanism for maintaining player attention, enhancing immersion, and reinforcing memorable challenges.

[0113] In existing technologies, dynamic difficulty adjustment typically relies on player performance metrics (such as win rate or score) as feedback, using preset functions for linear or piecewise adjustments. However, these approaches often overlook the structural complexity of the game content itself, making it difficult to achieve precise control over the overall game rhythm. To address this, some game systems introduce difficulty fluctuations through parameter randomization or mechanism layering. While this creates non-linear changes in the experience, it lacks a systematic modeling and control mechanism for difficulty trends, failing to meet the need for fitting specific difficulty rhythms. Therefore, these technologies suffer from the following drawbacks: One drawback is the inability to precisely control the pace and rhythm of the difficulty.

[0114] Related technical solutions typically arrange content using fixed difficulty labels or simple linear progression models, making it difficult to support complex rhythmic structures such as "gradual increase followed by slowness," "periodic fluctuations," and "phased releases." This approach lacks flexibility in expressing the designer's intent and is prone to problems such as uneven distribution of challenge intensity and discontinuous player experience.

[0115] The second drawback is the lack of ability to evaluate the logical structure of the game's content.

[0116] Most methods in related technologies rely on player performance data (such as win rate, time consumption, and health loss) to infer game difficulty, ignoring the inherent challenges of the game content itself in terms of mechanic combinations, task flow, and failure paths. This type of external data is not only lagging and highly affected by player status, but also cannot be predicted or optimized in advance during the content generation stage.

[0117] The third drawback is that the difficulty fluctuates uncontrollably and the rhythm is inconsistent.

[0118] In game systems using related technologies, difficulty fluctuations often rely on the experience of content designers or generate a "non-linear" experience through random parameters. However, this approach often results in unpredictable fluctuations, disjointed difficulty changes, or even continuous low or high pressure, affecting player immersion and challenge experience.

[0119] Therefore, the game content generation method proposed by the inventors in this application has the following characteristics: 1. Curve fitting mechanism: Construct various types of target difficulty curves (GDC, i.e., difficulty trend graphs), such as rhythm progression type, peak cycle type, and compression release type, as reference templates for the overall rhythm. During the game content generation stage, the target difficulty curve is used as the evaluation benchmark to guide the logical difficulty layout of levels or tasks, so that the generated content as a whole fits the expected rhythm.

[0120] 2. Logical difficulty modeling: An internally defined difficulty description language is introduced to quantify the logic of a level (a form of content unit) from dimensions such as mechanism complexity, task path length, and number of failure paths, generating an objective logical difficulty score (i.e., an evaluation difficulty value) to accurately control the challenge intensity of the game content, independent of player performance.

[0121] 3. Difficulty fluctuation control algorithm: Through feedback mechanism, it analyzes the deviation between the game content and the target difficulty curve in local and overall aspects, dynamically inserts high-pressure or slow-release segments, maintains the structural trend of the curve while retaining the rhythm fluctuation characteristics, and enhances the sense of tension and relaxation and the player's immersive experience.

[0122] 4. Intelligent fitting and diversity constraints: By combining machine learning models to perform feature clustering and classification training on the game's content units, a balance is achieved between fitting and controlling the difficulty curve and content diversity, avoiding excessive homogenization and improving content freshness and gameplay variability.

[0123] In this application, game content generation algorithms can be embedded into game editors or content building tools, providing the product with visualized difficulty control and content generation capabilities, thereby improving design efficiency and rhythm management accuracy. The main process is as follows: Developers can intuitively select or draw target difficulty curve templates through the game development interface, such as linear ascending, segmented step, bell curve, periodic sawtooth, and compression-release types. The game development system supports fine-tuning based on standard templates (e.g., dragging nodes, setting peak / trough positions) to precisely express the designer's intent. After obtaining the target difficulty curve, the game development system automatically performs logical difficulty analysis on the current level sequence (i.e., the initial content unit sequence) or task flow. Combining this with the built-in difficulty description language rule system, it quantitatively evaluates the mechanism complexity, failure path depth, and resource allocation of each level node (i.e., key node), draws the actual difficulty curve, and compares it with the target difficulty curve.

[0124] This application starts from the overall logic and is divided into three main stages: target difficulty fitting-driven, logical difficulty modeling and regulation, and generation control and diversity optimization. It achieves precise regulation and intelligent generation of content rhythm through a series of modular strategies.

[0125] First, the target difficulty fitting process involves constructing a target difficulty curve to drive content generation. The game development system allows developers to set or plot the desired trend of game difficulty change, i.e., the target difficulty curve (GDC). This curve can take the form of linear, bell-shaped, sawtooth, or segmented staircase, representing the pace of challenge the designer wants players to experience during gameplay. The game development system discretizes the GDC, transforming it into a series of target difficulty values ​​arranged in time or level order, serving as a fitting benchmark when generating content.

[0126] Secondly, logical difficulty modeling and control includes introducing a logical difficulty assessment mechanism to achieve content evaluation and control. The game development system, based on its internally defined difficulty description language (DDL) rule system, performs static logical difficulty modeling on existing or to-be-generated content units. This modeling process considers multiple dimensions of the task structure: the complexity of the mechanism layering; the depth of paths and failure conditions; player tolerance space and feedback intensity, etc. Specifically, the game development system, based on its internally defined difficulty description language (DDL) rule system, performs static logical difficulty modeling on existing or to-be-generated game content. This modeling approach does not rely on player behavior data but instead starts from the content itself, comprehensively evaluating three core dimensions: First, the complexity of the mechanic layering, referring to the number of combinations and interactions of different game mechanics (such as jumping, puzzle-solving, and combat) in the level. The system calculates the mechanic complexity score using a mechanic combination graph and a collaboration matrix. Second, the depth of paths and failure conditions, assessing the critical path length, failure node density, and retry cost of the task objective chain. This is modeled using a task flowchart and generates a path depth score. Third, the player's fault tolerance space and feedback intensity, measuring the timeliness of remedial mechanisms (such as respawn points and resource recovery) and system feedback responses provided in the task. The system outputs a fault tolerance score based on this. Through comprehensive calculation, the game development system can output a logical difficulty score (i.e., an evaluated difficulty value) for each content unit, and then compare it with the target difficulty sequence (i.e., a reference difficulty sequence) to determine whether adjustments, replacements, or insertions of task segments are needed. Simultaneously, to avoid abrupt or disjointed difficulty changes, a fluctuation control algorithm is introduced to constrain the amplitude, cycle, and buffer segments of the rhythm, creating a more natural rhythm. In terms of implementation, the first step is to perform local differential analysis on the target difficulty curve to extract the gradient of change between peaks and troughs, and set a reasonable "fluctuation threshold range" to avoid the accumulation of consecutive high-pressure or low-pressure segments. Subsequently, when generating content or sorting task nodes, the difficulty difference between each content unit and the nodes before and after it can be detected based on the logical difficulty score of each content unit. If the difference exceeds the set threshold, a "transition task" is inserted or the task order is adjusted to buffer the change in pace. At the same time, the fluctuation cycle of the entire process is modeled (e.g., allowing a peak every N tasks) to ensure that the overall structure has rhythmic highlights without being overly tense or loose, achieving organic fluctuations in the challenge and enhancing immersion.

[0127] Finally, generation control and diversity optimization are achieved by combining machine learning models to optimize fitting accuracy and content diversity. During the content generation or reconstruction phase, a trained classification model or sequence generation model is used to identify content segments or mechanism combinations that match the current target difficulty level. This model can be used to fit the target difficulty trend and ensure sufficient diversity in the generated results across dimensions such as structure, theme, and interaction methods. After identifying candidate segments, the system determines whether to insert, replace, or rearrange them based on their fit to the target difficulty value and their relationship to the rhythm of preceding and following content. A diversity constraint mechanism prevents the repetition of similar mechanisms or styles. This process ensures that the generated content closely follows the GDC rhythm requirements while maintaining continuous freshness and gameplay variability, effectively supporting the dual goals of content generation stability and innovation. Furthermore, the system can dynamically detect issues such as content redundancy and structural monotony, introducing a penalty mechanism to encourage differentiated construction, thereby balancing the dual needs of "rhythm control" and "gameplay diversity."

[0128] By providing a clear fitting mechanism, developers can design and control the overall difficulty curve, enhancing the game's rhythm and improving difficulty control. Judging difficulty through logical structure rather than player performance avoids unfair experiences caused by skill bias or luck, enhancing the fairness and objectivity of the challenge. The system can automatically judge and optimize generated content, saving time spent manually adjusting level order and difficulty, and improving game design efficiency. Fine-tuning the pacing through controlled fluctuations allows players to be more naturally immersed in the challenge, preventing premature fatigue or boredom and increasing the rhythm and immersion of the player experience.

[0129] The following description continues to illustrate the exemplary structure of the game content generation device 455 provided in the embodiments of this application as a software module. In some embodiments, such as... Figure 2 As shown, the software module in the game content generation device 455 stored in the memory 450 may include: Module 4551 is used to construct a reference difficulty sequence based on the initial content unit sequence of the game. The content units in the initial content unit sequence are the smallest content carriers with independent logical structures in the game. The reference difficulty sequence includes the reference difficulty value of each content unit as the game's key nodes change.

[0130] Evaluation module 4552 is used to evaluate the logical difficulty of each content unit in the initial content unit sequence and obtain the evaluation difficulty value of each content unit.

[0131] The adjustment module 4553 is used to determine the difference between the reference difficulty value and the evaluation difficulty value of each content unit, and adjust each content unit in the initial content unit sequence based on the difference to obtain the target content unit sequence.

[0132] The generation module 4554 is used to generate game content based on the target content unit sequence.

[0133] In some embodiments, the construction module 4551 is further configured to construct a difficulty trend map to characterize the difficulty change trend of the game based on the initial content unit sequence of the game; for each key node in the game, discretize sampling is performed on the difficulty trend map to obtain the difficulty height corresponding to each key node; determine the reference difficulty value of each key node based on the difficulty height corresponding to each key node, and combine the reference difficulty values ​​of each key node to obtain a reference difficulty sequence.

[0134] In some embodiments, the construction module 4551 is further configured to determine the difficulty gradient corresponding to the difficulty height of each key node in the difficulty trend graph, and determine the first key node in the key nodes of the game whose difficulty gradient exceeds the preset gradient; adjust the difficulty height corresponding to the first key node in the difficulty trend graph to obtain an adjusted difficulty trend graph; the difficulty gradient corresponding to the adjusted first key node is lower than the preset gradient; and determine the reference difficulty value of each key node based on the difficulty height corresponding to each key node in the adjusted difficulty trend graph.

[0135] In some embodiments, the evaluation module 4552 is further configured to extract evaluation elements that can characterize the logical difficulty of the content unit for each content unit in the initial content unit sequence, wherein the evaluation elements include at least one of the following: mechanism complexity, task chain depth, and fault tolerance space; to score the difficulty of each evaluation element of the content unit to obtain the element score corresponding to each evaluation element; and to fuse the element scores of each evaluation element to obtain the evaluation difficulty value of the content unit.

[0136] In some embodiments, the evaluation module 4552 is further configured to determine the operation preferences of different player groups based on the operation data of the player groups, and determine the weight of each evaluation element based on the operation preferences; and to perform a weighted summation of the element scores of each evaluation element based on the weight of each evaluation element to obtain the evaluation difficulty value of the content unit.

[0137] In some embodiments, the evaluation module 4552 is further configured to determine the first difference between the reference difficulty value and the evaluation difficulty value of each content unit in the initial content unit sequence as the difference of the content unit; the adjustment module 4553 is further configured to adjust the logical difficulty of the content unit in response to the first difference corresponding to the content unit exceeding a first preset difference; or, insert a new content unit before or after the content unit.

[0138] In some embodiments, the adjustment module 4553 is further configured to, for adjacent first content units and second content units in the initial content unit sequence, if the second difference between the first evaluation difficulty value of the first content unit and the second evaluation difficulty value of the second content unit exceeds the second preset difference, insert a third content unit between the first content unit and the second content unit, wherein the third evaluation difficulty of the third content unit is between the first evaluation difficulty and the second evaluation difficulty, so as to combine the third content unit to generate a target content unit sequence.

[0139] In some embodiments, the adjustment module 4553 is further configured to detect content units in the target content unit sequence based on a preset optimization mechanism, determine the content units to be optimized in the target content unit sequence; for each content unit to be optimized, select multiple candidate content units in the content unit library based on the reference difficulty value corresponding to the content unit to be optimized; determine the target content unit with a mechanism type different from the first mechanism type of the content unit to be optimized among the multiple candidate content units, and adjust the content unit to be optimized based on the target content unit.

[0140] In some embodiments, the adjustment module 4553 is further configured to cluster multiple candidate content units based on the mechanism type of the candidate content units to obtain at least one cluster; each cluster includes multiple candidate content units belonging to the same mechanism type; determine the second mechanism type of the content unit at the key node before the content unit to be optimized and the third mechanism type of the content unit at the key node after the content unit to be optimized; determine the target cluster in the at least one cluster whose mechanism type is different from the first mechanism type, the second mechanism type and the third mechanism type, and determine the target content unit in the candidate content units of the target cluster.

[0141] This application provides a computer program product, which includes a computer program or computer-executable instructions. When the computer-executable instructions or the computer program are executed by a processor, the processor will execute the game content generation method provided in this application embodiment, for example, such as... Figure 3 The method for generating game content is illustrated. The processor of an electronic device reads the computer program or computer-executable instructions from a computer-readable storage medium, and executes the computer program or computer-executable instructions, causing the electronic device to perform the game content generation method described in the embodiments of this application.

[0142] This application provides a computer-readable storage medium storing computer-executable instructions or a computer program. When the computer-executable instructions or the computer program are executed by a processor, the processor will execute the game content generation method provided in this application. For example, ... Figure 3 The method for generating game content is shown.

[0143] In some embodiments, the computer-readable storage medium may be a memory such as RAM, ROM, flash memory, magnetic surface memory, optical disk, or CD-ROM; or it may be a variety of devices including one or any combination of the above-mentioned memories.

[0144] In some embodiments, computer-executable instructions may take the form of programs, software, software modules, scripts, or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and may be deployed in any form, including as stand-alone programs or as modules, components, subroutines, or other units suitable for use in a computing environment.

[0145] As an example, computer-executable instructions may, but do not necessarily, correspond to files in a file system. They may be stored as part of a file that holds other programs or data, for example, in one or more scripts in a Hyper Text Markup Language (HTML) document, in a single file dedicated to the program in question, or in multiple co-located files (e.g., files that store one or more modules, subroutines, or code sections).

[0146] As an example, computer-executable instructions can be deployed to execute on a single electronic device, or on multiple electronic devices located at one location, or on multiple electronic devices distributed across multiple locations and interconnected via a communication network.

[0147] In summary, through the above embodiments, when generating game content, a reference difficulty sequence is first constructed based on the initial content unit sequence of the game. The reference difficulty sequence includes the reference difficulty value of each content unit in the initial content unit sequence as it changes with key nodes in the game. Then, the logical difficulty of each content unit in the initial content unit sequence is evaluated to obtain the evaluated difficulty value of each content unit. Based on the difference between the reference difficulty value and the evaluated difficulty value of each content unit, the content units are adjusted to obtain the target content unit sequence. This method, which adjusts the content units based on the reference difficulty sequence, can grasp the overall trend of the difficulty change of the game content and control the overall rhythm of the game. During the generation of the target content unit sequence, only content units that do not meet the trend need to be adjusted, reducing the frequency of game adjustments, improving the efficiency of game content generation, and thus improving system utilization.

[0148] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, and improvements made within the spirit and scope of this application are included within the scope of protection of this application.

Claims

1. A method for generating game content, characterized in that, The method includes: A reference difficulty sequence is constructed based on the initial content unit sequence of the game. The content unit in the initial content unit sequence is the smallest content carrier with an independent logical structure in the game. The reference difficulty sequence includes the reference difficulty value of each content unit as the key nodes of the game change. For each content unit in the initial content unit sequence, an evaluation element that can characterize the logical difficulty of the content unit is extracted, and the evaluation element includes at least one of the following: mechanism complexity, task chain depth, and fault tolerance space. The mechanism complexity represents the objective characteristics of the core rules and interaction mechanisms contained in the content unit; the task chain depth represents the logical process required to complete the task objective of the content unit; and the fault tolerance space represents the degree to which the content unit tolerates player operation errors. Difficulty scores are assigned to each evaluation element of the content unit to obtain the element score corresponding to each evaluation element, and the element scores of each evaluation element are merged to obtain the evaluation difficulty value of the content unit. Determine the difference between the reference difficulty value and the evaluation difficulty value of each content unit, and adjust each content unit in the initial content unit sequence based on the difference to obtain the target content unit sequence; The game content of the game is generated based on the target content unit sequence.

2. The method according to claim 1, characterized in that, The construction of the reference difficulty sequence based on the initial content unit sequence of the game includes: Based on the initial content unit sequence of the game, a difficulty trend graph is constructed to characterize the difficulty change trend of the game; For each key node in the game, discretization sampling is performed on the difficulty trend map to obtain the difficulty level corresponding to each key node; Based on the difficulty level corresponding to each key node, a reference difficulty value for each key node is determined, and the reference difficulty values ​​for each key node are combined to obtain the reference difficulty sequence.

3. The method according to claim 2, characterized in that, After obtaining the difficulty level corresponding to each key node, the method further includes: Determine the difficulty gradient corresponding to the difficulty height of each key node in the difficulty trend graph, and identify the first key node in the game whose difficulty gradient exceeds a preset gradient; The difficulty height corresponding to the first key node in the difficulty trend graph is adjusted to obtain an adjusted difficulty trend graph; the difficulty gradient corresponding to the first key node in the adjusted graph is lower than the preset gradient. The process of determining a reference difficulty value for each key node based on the difficulty level corresponding to each key node includes: Based on the difficulty height corresponding to each key node in the adjusted difficulty trend graph, a reference difficulty value for each key node is determined.

4. The method according to claim 1, characterized in that, The process of fusing the element scores of each of the evaluation elements to obtain the evaluation difficulty value of the content unit includes: For different player groups, the operation preferences of the player groups are determined based on the operation data of the player groups, and the weight of each of the evaluation elements is determined based on the operation preferences; Based on the weight of each evaluation element, the element scores of each evaluation element are weighted and summed to obtain the evaluation difficulty value of the content unit.

5. The method according to claim 1, characterized in that, Determining the difference between the reference difficulty value and the assessed difficulty value for each content unit includes: For each content unit in the initial content unit sequence, a first difference between the reference difficulty value and the evaluation difficulty value of the content unit is determined as the difference of the content unit; The adjustment of each content unit in the initial content unit sequence based on the difference includes: In response to the first difference corresponding to the content unit exceeding a first preset difference, the logical difficulty of the content unit is adjusted; or, a new content unit is inserted before or after the content unit.

6. The method according to claim 5, characterized in that, The method further includes: For adjacent first and second content units in the initial content unit sequence, if the second difference between the first evaluation difficulty value of the first content unit and the second evaluation difficulty value of the second content unit exceeds the second preset difference, a third content unit is inserted between the first content unit and the second content unit. The third evaluation difficulty value of the third content unit is between the first evaluation difficulty value and the second evaluation difficulty value, so as to generate the target content unit sequence by combining the third content unit.

7. The method according to claim 1, characterized in that, After obtaining the target content unit sequence, the method further includes: Based on a preset optimization mechanism, the content units in the target content unit sequence are detected to determine the content units to be optimized in the target content unit sequence; For each of the content units to be optimized, multiple candidate content units are selected from the content unit library based on the reference difficulty value corresponding to the content unit to be optimized; Among the plurality of candidate content units, a target content unit with a mechanism type different from the first mechanism type of the content unit to be optimized is determined, and the content unit to be optimized is adjusted based on the target content unit.

8. The method according to claim 7, characterized in that, The step of determining the target content unit whose mechanism type differs from the first mechanism type of the content unit to be optimized from the plurality of candidate content units includes: Based on the mechanism type of the candidate content units, multiple candidate content units are clustered to obtain at least one cluster; each cluster includes multiple candidate content units belonging to the same mechanism type. Determine the second mechanism type of the content unit at the key node preceding the content unit to be optimized and the third mechanism type of the content unit at the key node following the content unit to be optimized; In the at least one cluster, a target cluster with a mechanism type different from the first mechanism type, the second mechanism type, and the third mechanism type is determined, and the target content unit is determined from the candidate content units of the target cluster.

9. A device for generating game content, characterized in that, The device includes: A construction module is used to construct a reference difficulty sequence based on the initial content unit sequence of the game. The content unit in the initial content unit sequence is the smallest content carrier with an independent logical structure in the game. The reference difficulty sequence includes the reference difficulty value of each content unit as the game's key nodes change. An evaluation module is used to extract evaluation elements that characterize the logical difficulty of each content unit in the initial content unit sequence. These evaluation elements include at least one of the following: mechanism complexity, task chain depth, and fault tolerance space. The mechanism complexity characterizes the objective features of the core rules and interaction mechanisms contained in the content unit; the task chain depth characterizes the logical process required to complete the task objective of the content unit; and the fault tolerance space characterizes the tolerance of the content unit for player errors. A difficulty score is assigned to each evaluation element of the content unit to obtain a corresponding element score, and the element scores of each evaluation element are merged to obtain the evaluation difficulty value of the content unit. An adjustment module is used to determine the difference between the reference difficulty value and the evaluation difficulty value of each content unit, and to adjust each content unit in the initial content unit sequence based on the difference to obtain a target content unit sequence. The generation module is used to generate game content for the game based on the target content unit sequence.

Images