Modular game development engine system and method

Abstract

The present application relates to the field of intelligent game development, and specifically to a modular game development engine system and method. It uses artificial intelligence technology based on deep learning to perform semantic analysis on game development project requirements, generate game logic text, and call basic component packages at the same time, and randomly arrange and combine each basic component in the basic component package and perform text description global semantic association encoding to generate a diversified basic component package functional feature representation, and then based on the semantic matching between the game logic text and the basic component package functional information, intelligent recommendation of the basic component package is achieved. In this way, the accuracy and efficiency of component selection in the game development process can be effectively improved, development costs can be reduced, and code reuse and game quality can be promoted.

Description

Technical Field

[0001] The present application relates to the field of intelligent game development, and specifically to a modular game development engine system and method. Background Art

[0002] In the current game development process, the traditional development model often relies on a large amount of customized code and is highly dependent on the programming ability of developers, which not only leads to low development efficiency, but also makes the maintenance and update of the game extremely complicated. In addition, in the face of rapidly changing market demands and player preferences, traditional development methods are difficult to quickly adjust game logic and functions, which may lead to project delays or failure to meet market demands in a timely manner.

[0003] With the development of the game industry, the concept of modular game development has gradually gained attention. Modular development allows developers to split various parts of the game, such as characters, environments, logic, etc., into independent components, and build games by combining these components, which greatly improves development efficiency and code reusability.

[0004] However, in game development, although there are many reusable basic components, the reuse efficiency and flexibility of these components are limited due to the lack of effective component management and intelligent matching mechanisms. How to effectively match and select appropriate basic component packages according to project requirements remains an important challenge.

[0005] Therefore, an optimized modular game development engine system and method are desired. Summary of the invention

[0006] This application is made in view of the above problems. One object of this application is to provide a modular game development engine system and method.

[0007] An embodiment of the present application provides a modular game development engine system, which includes:

[0008] Project requirement information acquisition module, used to obtain text description of project requirements;

[0009] A game logic generation module, used to generate game logic based on the text description of the project requirements to obtain a game logic text description that matches the requirements;

[0010] A candidate basic component package information acquisition module is used to acquire the candidate basic component package and extract the text description of each basic component in the candidate basic component package;

[0011] A semantic encoding module, used for semantically encoding the text description of the demand matching game logic and the text description of each basic component in the candidate basic component package to obtain a set of demand matching game logic semantic encoding feature vectors and basic component function semantic encoding feature vectors;

[0012] A component package function semantic coding module, used to extract global semantic association features after performing multiple random permutations and combinations on the set of basic component function semantic coding feature vectors to obtain multiple basic component package global function semantic coding feature vectors;

[0013] A matching analysis module is used to determine whether to return the candidate basic component package based on a one-way interactive matching of the demand matching game logic semantic coding feature vector with respect to the global function semantic coding feature vectors of the multiple basic component packages.

[0014] For example, according to the modular game development engine system of an embodiment of the present application, the game logic generation module is used to:

[0015] The text description of the project requirements is input into a game logic generator based on the AIGC model to obtain a game logic text description matching the requirements.

[0016] For example, according to the modular game development engine system of an embodiment of the present application, the semantic encoding module includes:

[0017] A game logic text description semantic encoding unit, used for inputting the requirement matching game logic text description into a game logic text semantic encoder based on a Bert model to obtain the requirement matching game logic semantic encoding feature vector;

[0018] The basic component function semantic encoding unit is used to use the basic component text description semantic encoder based on the Bert model to semantically encode the text description of each basic component in the alternative basic component package to obtain a set of basic component function semantic encoding feature vectors.

[0019] For example, according to the modular game development engine system of an embodiment of the present application, the component package functional semantic encoding module includes:

[0020] A random permutation and combination unit, used for randomly permuting each basic component function semantic coding feature vector in the set of basic component function semantic coding feature vectors to obtain a random combination of multiple basic component function semantic coding feature vectors;

[0021] The global semantic association encoding unit is used to input the random combination of each basic component function semantic encoding feature vector in the random combination of the multiple basic component function semantic encoding feature vectors into the random basic component package global semantic encoder based on the Bi-LSTM model to obtain the global function semantic encoding feature vectors of the multiple basic component packages.

[0022] For example, according to the modular game development engine system of an embodiment of the present application, the matching analysis module includes:

[0023] A one-way matching analysis unit, used for performing one-way matching analysis on the demand matching game logic semantic coding feature vector and the multiple basic component package global function semantic coding feature vectors to obtain a one-way matching result representation vector;

[0024] An initialization result generating unit is used to input the one-way matching result representation vector into a classifier-based development initialization result generator to obtain an initialization result, wherein the initialization result is used to indicate whether to return the candidate basic component package.

[0025] For example, according to the modular game development engine system of an embodiment of the present application, the one-way matching analysis unit is used to:

[0026] The demand matching game logic semantic coding feature vector is used as a query feature vector, and its one-way attention granular scanning interactive matching result relative to the global functional semantic coding feature vectors of the multiple basic component packages is calculated to obtain the one-way matching result representation vector.

[0027] For example, according to the modular game development engine system of an embodiment of the present application, the one-way matching analysis unit includes:

[0028] A semantic metric coefficient calculation subunit, used for calculating the semantic metric coefficient between the query feature vector and each basic component package global function semantic coding feature vector in the multiple basic component package global function semantic coding feature vectors to obtain a sequence of one-way matching semantic metric coefficients;

[0029] A normalization subunit, used for normalizing the sequence of one-way matching semantic metric coefficients through a softmax function to obtain a sequence of one-way matching semantic metric weights;

[0030] A semantic reshaping subunit, used to calculate the weighted sum of the global functional semantic encoding feature vectors of the multiple basic component packages using the sequence of the one-way matching semantic measurement weights as weights to obtain a basic component package functional semantic reshaping representation vector;

[0031] The feature difference calculation subunit is used to calculate the differential feature vector between the query feature vector and the basic component package functional semantic reconstruction representation vector as the one-way matching result representation vector.

[0032] For example, according to the modular game development engine system of an embodiment of the present application, the semantic metric coefficient calculation subunit is used to:

[0033] Calculating the square of each eigenvalue of each basic component package global function semantic coding feature vector in the multiple basic component package global function semantic coding feature vectors to obtain a nonlinear transformation of multiple basic component package global function semantic coding feature vectors;

[0034] The query feature vector is concatenated with each of the nonlinear transformation basic component package global function semantic coding feature vectors in the nonlinear transformation multiple basic component package global function semantic coding feature vectors to obtain a sequence of semantic association feature vectors;

[0035] Using a preset transformation vector to multiply each semantic association feature vector in the sequence of semantic association feature vectors respectively to obtain a sequence of semantic similarities;

[0036] Each semantic similarity in the sequence of semantic similarities is added with a bias parameter and then passed through a sigmoid activation function to obtain a sequence of one-way matching semantic measurement coefficients.

[0037] The embodiment of the present application also provides a modular game development engine method, which includes:

[0038] Get a text description of the project requirements;

[0039] Based on the text description of the project requirements, generate game logic to obtain a text description of the game logic that matches the requirements;

[0040] Obtaining a candidate basic component package, and extracting a text description of each basic component in the candidate basic component package;

[0041] Semantically encoding the demand matching game logic text description and the text description of each basic component in the candidate basic component package to obtain a set of demand matching game logic semantic encoding feature vectors and basic component function semantic encoding feature vectors;

[0042] Performing multiple random permutations and combinations on the set of the basic component functional semantic coding feature vectors, and then extracting the global semantic association features thereof, so as to obtain a plurality of basic component package global functional semantic coding feature vectors;

[0043] Based on the one-way interactive matching of the demand matching game logic semantic coding feature vector with respect to the global function semantic coding feature vectors of the multiple basic component packages, it is determined whether to return the alternative basic component package.

[0044] For example, according to the modular game development engine method of an embodiment of the present application, based on the text description of the project requirements, generating game logic to obtain a text description of the game logic that matches the requirements includes:

[0045] The text description of the project requirements is input into a game logic generator based on the AIGC model to obtain a game logic text description matching the requirements.

[0046] According to the modular game development engine system and method of the embodiments of the present application, it can effectively improve the accuracy and efficiency of component selection during the game development process, reduce development costs, and promote code reuse and improvement of game quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings of the embodiments of the present application are briefly introduced below. Obviously, the drawings described below only relate to some embodiments of the present application, and are not intended to limit the present application.

[0048] Figure 1 A schematic diagram of the application architecture of a modular game development engine system in an embodiment of the present application is shown;

[0049] Figure 2 A schematic diagram of the structure of a modular game development engine system in an embodiment of the present application is shown;

[0050] Figure 3 A schematic diagram showing the structure of a semantic coding module of a modular game development engine system in an embodiment of the present application is shown;

[0051] Figure 4 A schematic diagram showing the structure of a component package function semantic coding module of a modular game development engine system in an embodiment of the present application is shown;

[0052] Figure 5 A schematic diagram of the structure of a matching analysis module of a modular game development engine system in an embodiment of the present application is shown;

[0053] Figure 6 A flowchart showing a modular game development engine method in an embodiment of the present application is shown; and

[0054] Figure 7 An application scenario diagram of a modular game development engine system in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0055] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work also fall within the scope of protection of the present application.

[0056] The terms used in this specification are those common terms currently widely used in the art in consideration of the functions of the present application, but these terms may vary according to the intentions of those skilled in the art, precedents, or new technologies in the art. In addition, specific terms may be selected, and in this case, their detailed meanings will be described in the detailed description of the present application. Therefore, the terms used in the specification should not be understood as simple names, but rather as a general description based on the meaning of the terms and the present application.

[0057] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0058] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. At the same time, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.

[0059] Figure 1 A schematic diagram of the application architecture of a modular game development engine system in an embodiment of the present application is shown, including a server 100 and a terminal device 200.

[0060] The terminal device 200 and the server 100 can be connected via the Internet to achieve mutual communication. Optionally, the above-mentioned Internet uses standard communication technology and / or protocol. The Internet is usually the Internet, but it can also be any network, including but not limited to any combination of a local area network (LAN), a metropolitan area network (MAN), a wide area network (WAN), a mobile, wired or wireless network, a dedicated network or a virtual private network. In some embodiments, the data exchanged through the network is represented by technologies and / or formats including Hyper Text Markup Language (HTML), Extensible Markup Language (XML), etc. In addition, conventional encryption technologies such as Secure Socket Layer (SSL), Transport Layer Security (TLS), Virtual Private Network (VPN), Internet Protocol Security (IPsec) can also be used to encrypt all or some links. In other embodiments, customized and / or dedicated data communication technologies can also be used to replace or supplement the above-mentioned data communication technologies.

[0061] The server 100 can provide various network services for the terminal device 200, wherein the server 100 can be a single server, a server cluster consisting of several servers, or a cloud computing center. Specifically, the server 100 may include a processor 110 (Center Processing Unit, CPU), a memory 120, an input device 130, and an output device 140, etc. The input device 130 may include a keyboard, a mouse, a touch screen, etc., and the output device 140 may include a display device, such as a liquid crystal display (Liquid Crystal Display, LCD), a cathode ray tube (Cathode Ray Tube, CRT), etc.

[0062] The memory 120 may include a read-only memory (ROM) and a random access memory (RAM), and provides the processor 110 with program instructions and data stored in the memory 120. In the embodiment of the present application, the memory 120 may be used to store the corresponding method of the modular game development engine system in the embodiment of the present application.

[0063] The processor 110 calls the program instructions stored in the memory 120, and the processor 110 is used to execute the steps of any modular game development engine method in the embodiments of the present application according to the obtained program instructions.

[0064] In addition, the application architecture diagram in the embodiment of the present application is intended to more clearly illustrate the technical solution in the embodiment of the present application, and does not constitute a limitation on the technical solution provided in the embodiment of the present application. Of course, for other application architectures and business applications, the technical solution provided in the embodiment of the present application is also applicable to similar problems.

[0065] The following is a non-restrictive description of the modular game development engine system provided according to at least one embodiment of the present application through several examples or embodiments. As described below, different features in these specific examples or embodiments can be combined with each other without conflicting with each other to obtain new examples or embodiments, and these new examples or embodiments also fall within the scope of protection of the present application.

[0066] In response to the above technical problems, the technical concept of this application is to use artificial intelligence technology based on deep learning to perform semantic analysis on the requirements of game development projects, generate game logic text, and call basic component packages at the same time, and randomly arrange and combine each basic component in the basic component package and perform text description global semantic association encoding to generate a variety of basic component package functional feature representations, and then based on the semantic matching between the game logic text and the basic component package functional information, to achieve intelligent recommendation of the basic component package. In this way, the accuracy and efficiency of component selection in the game development process can be effectively improved, development costs can be reduced, and code reuse and game quality can be promoted.

[0067] Based on this, Figure 2A schematic diagram of the structure of a modular game development engine system 800 in an embodiment of the present application is shown. The modular game development engine system 800 includes: a project requirement information acquisition module 810, which is used to obtain a text description of project requirements; a game logic generation module 820, which is used to generate game logic based on the text description of the project requirements to obtain a requirement matching game logic text description; an alternative basic component package information acquisition module 830, which is used to obtain an alternative basic component package and extract the text description of each basic component in the alternative basic component package; a semantic encoding module 840, which is used to semantically encode the requirement matching game logic text description and the text description of each basic component in the alternative basic component package to obtain a set of requirement matching game logic semantic encoding feature vectors and basic component function semantic encoding feature vectors; a component package function semantic encoding module 850, which is used to extract the global semantic association features of the set of basic component function semantic encoding feature vectors after performing multiple random permutations and combinations to obtain multiple basic component package global function semantic encoding feature vectors; a matching analysis module 860, which is used to determine whether to return the alternative basic component package based on the one-way interactive matching of the requirement matching game logic semantic encoding feature vector with respect to the multiple basic component package global function semantic encoding feature vectors.

[0068] Specifically, in the technical solution of this application, first, a text description of the project requirements is obtained. It should be understood that the text description of the project requirements clarifies the specific goals, functional requirements, style preferences and other key information of the game development to ensure that the entire development process can be carried out around clear goals, which is the starting point of the game development work.

[0069] In the traditional game development process, the design and writing of game logic often requires a lot of manual work, which is time-consuming and prone to errors. In the technical solution of this application, a game logic generator based on the AIGC model is introduced to perform semantic analysis on the text description of the project requirements, automatically generate the corresponding game logic, and thus obtain a text description of the game logic that matches the requirements. Among them, the AIGC model is based on natural language processing technology, and uses deep learning algorithms to parse key information in project requirements, understand and abstract text logic, so that it can automatically generate a logically clear and demand-compliant text description of game logic, greatly improving development efficiency and accuracy.

[0070] Correspondingly, the game logic generation module 820 is used to: input the text description of the project requirements into the game logic generator based on the AIGC model to obtain the game logic text description matching the requirements.

[0071] Next, in order to achieve deep semantic mining of the demand matching game logic text description, the demand matching game logic text description is further semantically encoded to understand its inherent semantic structural characteristics. In a specific example of the present application, considering the powerful capabilities of the Bert model in semantic understanding and representation learning, the present application uses a game logic text semantic encoder based on the Bert model to encode the demand matching game logic text description to fully capture the implicit meaning and contextual relationship in the game logic text, understand its logical structure, and thus generate a demand matching game logic semantic encoding feature vector.

[0072] Accordingly, if Figure 3 As shown, the semantic encoding module 840 includes: a game logic text description semantic encoding unit 841, which is used to input the requirement matching game logic text description into a game logic text semantic encoder based on the Bert model to obtain the requirement matching game logic semantic encoding feature vector; a basic component function semantic encoding unit 842, which is used to use the basic component text description semantic encoder based on the Bert model to semantically encode the text description of each basic component in the alternative basic component package respectively to obtain a set of the basic component function semantic encoding feature vectors.

[0073] At the same time, the candidate basic component package is obtained, and the text description of each basic component in the candidate basic component package is extracted. It should be understood that the candidate basic component package is a collection of reusable modules in the game development process, wherein the text description of each basic component refers to the textual description or introduction of the basic unit constituting the game or software project, including the component name, function description, interface description, usage restrictions and dependencies, etc., which is an important basis for evaluating whether the component meets the project requirements.

[0074] In the technical solution of the present application, a pre-trained Bert model is also used to semantically encode the text descriptions of each basic component in the alternative basic component package, so as to capture the deep semantic information of the text, understand the core functions and characteristics of each basic component, and convert it from complex text information into a structured vector form, unify the data format, so as to facilitate matching analysis with the game logic semantic coding feature vector that matches the requirements, thereby obtaining a set of basic component function semantic coding feature vectors.

[0075] Furthermore, in order to fully explore the functional effects that may be brought about by different arrangements and combinations of the basic components in the alternative basic component package, so as to improve the comprehensiveness and flexibility of component matching, in the technical solution of the present application, first, the set of functional semantic coding feature vectors of the basic components are randomly arranged and combined, for example, 12345678 is a group, and 13245678 is a group, and random arrangements and combinations are performed in this way to form a random combination of multiple basic component functional semantic coding feature vectors.

[0076] Next, the random combination of the functional semantic coding feature vectors of each basic component in the random combination of the functional semantic coding feature vectors of the multiple basic components is subjected to contextual semantic association coding to understand the semantic association and logical structure of each basic component after combination. The present application adopts a global semantic encoder of a random basic component package based on the Bi-LSTM model to implement this encoding process. Based on the bidirectional information flow control capability, the Bi-LSTM model can effectively capture the global semantic dependencies in each random combination, fully understand the interactions and logical relationships between components in different permutations and combinations, and thus obtain the global functional semantic coding feature vectors of the basic component package corresponding to each random combination, so as to reflect the diverse functional effects of each basic component under different permutations and combinations.

[0077] Accordingly, if Figure 4 As shown, the component package functional semantic encoding module 850 includes: a random permutation and combination unit 851, which is used to randomly permute each basic component functional semantic encoding feature vector in the set of basic component functional semantic encoding feature vectors to obtain a random combination of multiple basic component functional semantic encoding feature vectors; a global semantic association encoding unit 852, which is used to input the random combination of each basic component functional semantic encoding feature vector in the random combination of the multiple basic component functional semantic encoding feature vectors into a random basic component package global semantic encoder based on a Bi-LSTM model to obtain the multiple basic component package global functional semantic encoding feature vectors.

[0078] Furthermore, the demand matching game logic semantic coding feature vector is used as the query feature vector, and its one-way attention granular scanning interactive matching result relative to the global functional semantic coding feature vectors of the multiple basic component packages is calculated to obtain a one-way matching result representation vector. That is, the correlation between the alternative basic component package and the project requirements is revealed through matching analysis technology. Specifically, based on the one-way attention mechanism and the granular scanning strategy, the attention weight is generated by granularly calculating the semantic similarity between the demand matching game logic semantic coding feature vector and the global functional semantic coding feature vectors of each basic component package, and the global functional semantic coding feature vectors of the multiple basic component packages are weighted and fused to obtain a comprehensive representation of the functional semantic features of the basic component packages related to the game logic, and then the difference between it and the game logic semantic features is calculated to reveal the difference or mismatch between the two in the semantic feature space, thereby obtaining a one-way matching result representation vector, thereby characterizing the adaptability of the alternative basic component package in meeting the game logic requirements.

[0079] The one-way matching result representation vector is input into the classifier-based development initialization result generator to obtain an initialization result, and the initialization result is used to indicate whether to return the alternative basic component package. In the technical solution of the present application, the development initialization result generator uses a pre-trained classifier model to perform feature learning and analysis on the one-way matching result representation vector, identify whether the alternative basic component package meets the game logic requirements, and generate an initialization result through classification mapping to indicate whether to return the alternative basic component package as a component selection for game development. In this way, the applicability of the basic component package can be quickly and accurately evaluated to ensure that it can form a good synergy with the game logic, thereby avoiding invalid development attempts and saving time and resources.

[0080] Accordingly, if Figure 5 As shown, the matching analysis module 860 includes: a one-way matching analysis unit 861, which is used to perform a one-way matching analysis on the demand matching game logic semantic coding feature vector and the global function semantic coding feature vectors of the multiple basic component packages to obtain a one-way matching result representation vector; an initialization result generation unit 862, which is used to input the one-way matching result representation vector into a classifier-based development initialization result generator to obtain an initialization result, and the initialization result is used to indicate whether to return the alternative basic component package.

[0081] Among them, the one-way matching analysis unit 861 is used to: use the demand matching game logic semantic coding feature vector as the query feature vector, calculate its one-way attention granular scanning interaction matching result relative to the global functional semantic coding feature vectors of the multiple basic component packages to obtain the one-way matching result representation vector.

[0082] Specifically, the one-way matching analysis unit 861 includes: a semantic metric coefficient calculation subunit, which is used to calculate the semantic metric coefficient between the query feature vector and each of the global functional semantic coding feature vectors of the multiple basic component packages to obtain a sequence of one-way matching semantic metric coefficients; a normalization subunit, which is used to normalize the sequence of one-way matching semantic metric coefficients through a softmax function to obtain a sequence of one-way matching semantic metric weights; a semantic reshaping subunit, which is used to calculate the weighted sum of the global functional semantic coding feature vectors of the multiple basic component packages using the sequence of one-way matching semantic metric weights as weights to obtain a basic component package functional semantic reshaping representation vector; a feature difference calculation subunit, which is used to calculate the differential feature vector between the query feature vector and the basic component package functional semantic reshaping representation vector as the one-way matching result representation vector.

[0083] More specifically, the semantic metric coefficient calculation subunit is used to: calculate the square of each eigenvalue of each basic component package global function semantic coding feature vector in the multiple basic component package global function semantic coding feature vectors to obtain multiple non-linear transformation basic component package global function semantic coding feature vectors; cascade the query feature vector with each non-linear transformation basic component package global function semantic coding feature vector in the multiple non-linear transformation basic component package global function semantic coding feature vectors to obtain a sequence of semantic association feature vectors; use a preset transformation vector to multiply each semantic association feature vector in the sequence of semantic association feature vectors to obtain a sequence of semantic similarities; add a bias parameter to each semantic similarity in the sequence of semantic similarities and then pass it through a sigmoid activation function to obtain a sequence of one-way matching semantic metric coefficients.

[0084] In a specific example, the one-way matching analysis unit 861 is used to process the query feature vector and the global functional semantic encoding feature vectors of the multiple basic component packages using the following one-way interactive matching formula to obtain the one-way matching result representation vector, wherein the one-way interactive matching formula is:

[0085]

[0086] Where c1 represents the query feature vector, x j represents the jth basic component package global function semantic encoding feature vector among the multiple basic component package global function semantic encoding feature vectors, [·,·] represents the cascade operation, w is the preset transformation vector, b is the bias parameter, σ represents the Simgoid function, r 1jrepresents the one-way matching semantic metric coefficient between the query feature vector and the global functional semantic coding feature vector of the jth basic component package, m is the number of global functional semantic coding feature vectors of the basic component package, exp(·) represents the exponential function operation with e as the base, e 1j represents the one-way matching semantic metric weight between the query feature vector and the global functional semantic encoding feature vector of the jth basic component package, x 1p The basic components package functions and semantics to reshape the representation vector. Indicates subtraction by position, x m Indicates that the one-way matching result represents a vector.

[0087] Preferably, the initialization result generating unit 862 includes: an optimization subunit, which is used to optimize the one-way matching result representation vector to obtain a corrected one-way matching result representation vector; a classification subunit, which is used to input the corrected one-way matching result representation vector into the classifier-based development initialization result generator to obtain the initialization result. It should be understood that the role of the classifier is to use given categories and known training data to learn classification rules and classifiers, and then classify (or predict) unknown data. Logistic regression, SVM, etc. are often used to solve binary classification problems. For multi-class classification problems, logistic regression or SVM can also be used, but multiple binary classifications are required to form multi-classification, but this is prone to errors and inefficient. Commonly used multi-classification methods include the Softmax classification function.

[0088] Among them, the optimization subunit is used to: cascade the global function semantic coding feature vectors of the multiple basic component packages into a cascade feature vector of the global function semantic coding of the basic component package, and convert the demand matching game logic semantic coding feature vector into a demand matching game logic semantic coding interpolation feature vector of the same length as the cascade feature vector of the global function semantic coding of the basic component package through linear interpolation; calculate the first norm and second norm of the mean vector of the demand matching game logic semantic coding interpolation feature vector and the cascade feature vector of the global function semantic coding of the basic component package; calculate the weighted sum of the inverse of the square root of the second norm and the first norm, and add it to the weighted sum of the inverse of the square root of the second norm and the first norm, and add it to the weighted sum of the inverse of the square root of the second norm and the first norm. The dot-sum vector of the cascade feature vectors is dot-multiplied to obtain a first one-way matching result correction sub-vector; the dot-multiplied product vector of the demand matching game logic semantic encoding interpolation feature vector and the basic component package global function semantic encoding cascade feature vector is dot-multiplied with the square root of the length of the order semantics-feed inventory interactive response representation vector to obtain a second one-way matching result correction sub-vector; the weighted sum of the first one-way matching result correction sub-vector and the second one-way matching result correction sub-vector is calculated to obtain a one-way matching result correction vector; after linearly transforming the one-way matching result correction vector to the same length as the one-way matching result representation vector, the one-way matching result representation vector is dot-multiplied with the one-way matching result representation vector to obtain the corrected one-way matching result representation vector.

[0089] In a specific example, the optimization subunit is used to calculate the one-way matching result correction vector using the following correction formula; wherein the correction formula is expressed as:

[0090]

[0091] V1 and V2 are the interpolation feature vector of the semantic encoding of the demand matching game logic and the cascade feature vector of the semantic encoding of the global function of the basic component package, respectively. μ is the mean vector of the interpolation feature vector of the demand matching game logic semantic coding and the cascade feature vector of the global functional semantic coding of the basic component package, ||·||1 represents the first norm of the vector, ||·||2 represents the second norm of the vector, L is the length of the cascade feature vector of the global functional semantic coding of the basic component package, and α and β are weighted sum weights as hyperparameters, represents point addition by position, ⊙ represents point multiplication by position, and V' represents the one-way matching result correction vector; after the one-way matching result correction vector is linearly transformed to the same length as the one-way matching result representation vector, point multiplication is performed with the one-way matching result representation vector to obtain the corrected one-way matching result representation vector.

[0092] Here, considering that the demand matching game logic semantic coding feature vector and the global function semantic coding feature vectors of the multiple basic component packages respectively represent the coding semantic features of the demand matching game logic text description and the text description of each basic component in the alternative basic component package, when performing one-way attention granular scanning interactive matching, it is expected to further compensate for the imbalance of matching domain offset caused by the imbalance of correspondence ratio under the fine-grained distribution of the sequence, so as to improve the expression effect of the one-way matching result representation vector, thereby improving the accuracy of the classification result.

[0093] Therefore, based on the constrained representation of the structured foreground and background distinction of the supermanifold of the demand-matching game logic semantic encoding feature vector and the global function semantic encoding feature vectors of the multiple basic component packages based on the mean vector norm between the demand-matching game logic semantic encoding feature vector and the global function semantic encoding feature vectors of the multiple basic component packages, the feature-level key correspondence between the feature vectors of the demand-matching game logic semantic encoding feature vector and the global function semantic encoding feature vectors of the multiple basic component packages is modeled, and the global association relationship between the corresponding features is adjusted, so as to make positive fine-grained correspondence suggestions through the unbalanced proportion control between the corresponding eigenvalues ​​of the demand-matching game logic semantic encoding feature vector and the global function semantic encoding feature vectors of the multiple basic component packages, so as to avoid the imbalance of matching domain offset between the feature vectors of the demand-matching game logic semantic encoding feature vector and the global function semantic encoding feature vectors of the multiple basic component packages by focusing on the focus.

[0094] In this way, by correcting the one-way matching result representation vector through the one-way matching result correction vector, the expression effect of the one-way matching result representation vector can be improved, thereby improving the accuracy of the initialization result obtained by inputting the classifier-based development initialization result generator.

[0095] Based on the above embodiments, see Figure 6 FIG. 1 is a flowchart of a modular game development engine method in an embodiment of the present application. For example, the modular game development engine method can be executed by a server, which can be Figure 1 The server 100 shown in FIG. Figure 6As shown, according to the modular game development engine method of the embodiment of the present application, the steps include: S510, obtaining a text description of the project requirements; S520, generating a game logic based on the text description of the project requirements to obtain a text description of the game logic that matches the requirements; S530, obtaining an alternative basic component package, and extracting a text description of each basic component in the alternative basic component package; S540, semantically encoding the text description of the game logic that matches the requirements and the text description of each basic component in the alternative basic component package to obtain a set of semantic encoding feature vectors for the semantic encoding of the requirements matching game logic and basic component function semantic encoding feature vectors; S550, extracting the global semantic association features of the set of basic component function semantic encoding feature vectors after performing multiple random permutations and combinations to obtain global function semantic encoding feature vectors for multiple basic component packages; S560, determining whether to return the alternative basic component package based on the one-way interactive matching of the semantic encoding feature vector of the game logic that matches the requirements relative to the global function semantic encoding feature vectors of the multiple basic component packages.

[0096] In one example, based on the text description of the project requirements, generating game logic to obtain the requirements matching game logic text description includes: inputting the text description of the project requirements into a game logic generator based on the AIGC model to obtain the requirements matching game logic text description.

[0097] Here, those skilled in the art will appreciate that the specific operations of each step in the above modular game development engine method have been described in detail above. Figures 2 to 5 The modular game development engine system 800 is described in detail in the description of the modular game development engine system 800, and therefore, its repeated description will be omitted.

[0098] Figure 7 FIG. 1 is an application scenario diagram of a modular game development engine system according to an embodiment of the present application. Figure 7 As shown, in this application scenario, first, a text description of the project requirements is obtained (for example, Figure 7 D1 as shown in the figure) and alternative basic component packages (e.g. Figure 7 Then, the text description of the project requirements and the candidate basic component package are input to a server (e.g., Figure 7 In S) as shown in , the server is able to use the modular game development engine algorithm to process the text description of the project requirements and the alternative basic component package to determine whether to return the alternative basic component package.

[0099] Based on the above embodiments, another exemplary embodiment of an electronic device is also provided in the embodiments of the present application. In some possible implementations, the electronic device in the embodiments of the present application may include a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor can implement the steps of the modular game development engine method in the above embodiments when executing the program.

[0100] For example, in the case of electronic equipment Figure 1 Taking the server 100 in the example as an example, the processor in the electronic device is the processor 110 in the server 100, and the memory in the electronic device is the memory 120 in the server 100.

[0101] The embodiment of the present application also provides a computer-readable storage medium, on which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, the modular game development engine method according to the embodiment of the present application described with reference to the above figures can be executed. The computer-readable storage medium includes, but is not limited to, for example, volatile memory and / or non-volatile memory. The volatile memory may, for example, include a random access memory (RAM) and / or a cache memory (cache), etc. The non-volatile memory may, for example, include a read-only memory (ROM), a hard disk, a flash memory, etc.

[0102] The embodiment of the present application also provides a computer program product or a computer program, which includes computer executable instructions, and the computer executable instructions are stored in a computer readable storage medium. The processor of the computer device reads the computer executable instructions from the computer readable storage medium, and the processor executes the computer executable instructions, so that the computer device executes the modular game development engine method according to the embodiment of the present application.

[0103] Those skilled in the art will appreciate that the contents disclosed in this application may be subject to various variations and improvements. For example, the various devices or components described above may be implemented by hardware, or by software, firmware, or a combination of some or all of the three.

[0104] In addition, although the present application makes various references to certain units in the system according to embodiments of the present application, any number of different units can be used and run on the client and / or server. The units are only illustrative, and different aspects of the system and method can use different units.

[0105] Those skilled in the art will appreciate that all or part of the steps in the above method can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium, such as a read-only memory, a disk or an optical disk. Optionally, all or part of the steps in the above embodiment can also be implemented using one or more integrated circuits. Accordingly, each module / unit in the above embodiment can be implemented in the form of hardware or in the form of software function modules. The present application is not limited to any particular form of combination of hardware and software.

[0106] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art to which this application belongs. It should also be understood that terms such as those defined in common dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an idealized or extremely formal sense, unless explicitly defined as such herein.

[0107] The above is an explanation of the present application and should not be considered as limiting thereof. Although several exemplary embodiments of the present application have been described, those skilled in the art will readily appreciate that many modifications may be made to the exemplary embodiments without departing from the novel teachings and advantages of the present application.

Claims

1. A modular game development engine system, characterized in that: include: Project requirement information acquisition module, used to obtain text description of project requirements; A game logic generation module, used to generate game logic based on the text description of the project requirements to obtain a game logic text description that matches the requirements; A candidate basic component package information acquisition module is used to acquire the candidate basic component package and extract the text description of each basic component in the candidate basic component package; A semantic encoding module, used for semantically encoding the text description of the demand matching game logic and the text description of each basic component in the candidate basic component package to obtain a set of demand matching game logic semantic encoding feature vectors and basic component function semantic encoding feature vectors; A component package function semantic coding module, used to extract global semantic association features after performing multiple random permutations and combinations on the set of basic component function semantic coding feature vectors to obtain multiple basic component package global function semantic coding feature vectors; A matching analysis module, used to determine whether to return the candidate basic component package based on a one-way interactive matching of the demand matching game logic semantic coding feature vector with respect to the global function semantic coding feature vectors of the multiple basic component packages; Wherein, the matching analysis module includes: A one-way matching analysis unit, used for performing one-way matching analysis on the demand matching game logic semantic coding feature vector and the multiple basic component package global function semantic coding feature vectors to obtain a one-way matching result representation vector; An initialization result generating unit is used to input the one-way matching result representation vector into a classifier-based development initialization result generator to obtain an initialization result, wherein the initialization result is used to indicate whether to return the candidate basic component package.

2. The modular game development engine system according to claim 1, characterized in that: The game logic generation module is used to: The text description of the project requirements is input into a game logic generator based on the AIGC model to obtain a game logic text description matching the requirements.

3. The modular game development engine system according to claim 2, characterized in that: The semantic encoding module comprises: A game logic text description semantic encoding unit, used for inputting the requirement matching game logic text description into a game logic text semantic encoder based on a Bert model to obtain the requirement matching game logic semantic encoding feature vector; The basic component function semantic encoding unit is used to use the basic component text description semantic encoder based on the Bert model to semantically encode the text description of each basic component in the alternative basic component package to obtain a set of basic component function semantic encoding feature vectors.

4. The modular game development engine system according to claim 3, characterized in that: The component package functional semantic coding module includes: A random permutation and combination unit, used for randomly permuting each basic component function semantic coding feature vector in the set of basic component function semantic coding feature vectors to obtain a random combination of multiple basic component function semantic coding feature vectors; The global semantic association encoding unit is used to input the random combination of each basic component function semantic encoding feature vector in the random combination of the multiple basic component function semantic encoding feature vectors into the random basic component package global semantic encoder based on the Bi-LSTM model to obtain the global function semantic encoding feature vectors of the multiple basic component packages.

5. The modular game development engine system according to claim 4, characterized in that: The one-way matching analysis unit is used to: The demand matching game logic semantic coding feature vector is used as a query feature vector, and its one-way attention granular scanning interactive matching result relative to the global functional semantic coding feature vectors of the multiple basic component packages is calculated to obtain the one-way matching result representation vector.

6. The modular game development engine system according to claim 5, characterized in that: The one-way matching analysis unit comprises: A semantic metric coefficient calculation subunit, used for calculating the semantic metric coefficient between the query feature vector and each basic component package global function semantic coding feature vector in the multiple basic component package global function semantic coding feature vectors to obtain a sequence of one-way matching semantic metric coefficients; A normalization subunit, used for normalizing the sequence of one-way matching semantic measurement coefficients through a softmax function to obtain a sequence of one-way matching semantic measurement weights; A semantic reshaping subunit, used to calculate the weighted sum of the global functional semantic encoding feature vectors of the multiple basic component packages using the sequence of the one-way matching semantic measurement weights as weights to obtain a basic component package functional semantic reshaping representation vector; The feature difference calculation subunit is used to calculate the differential feature vector between the query feature vector and the basic component package functional semantic reconstruction representation vector as the one-way matching result representation vector.

7. The modular game development engine system according to claim 6, characterized in that: The semantic metric coefficient calculation subunit is used to: Calculating the square of each eigenvalue of each basic component package global function semantic coding feature vector in the multiple basic component package global function semantic coding feature vectors to obtain a nonlinear transformation of multiple basic component package global function semantic coding feature vectors; The query feature vector is concatenated with each of the nonlinear transformation basic component package global function semantic coding feature vectors in the nonlinear transformation multiple basic component package global function semantic coding feature vectors to obtain a sequence of semantic association feature vectors; Using a preset transformation vector to multiply each semantic association feature vector in the sequence of semantic association feature vectors respectively to obtain a sequence of semantic similarities; Each semantic similarity in the sequence of semantic similarities is added with a bias parameter and then passed through a sigmoid activation function to obtain a sequence of one-way matching semantic measurement coefficients.

8. A modular game development engine method, using the modular game development engine system of claim 1, characterized in that: include: Get a text description of the project requirements; Based on the text description of the project requirements, generate game logic to obtain a text description of the game logic that matches the requirements; Obtaining a candidate basic component package, and extracting a text description of each basic component in the candidate basic component package; Semantically encoding the demand matching game logic text description and the text description of each basic component in the candidate basic component package to obtain a set of demand matching game logic semantic encoding feature vectors and basic component function semantic encoding feature vectors; Performing multiple random permutations and combinations on the set of the basic component functional semantic coding feature vectors, and then extracting the global semantic association features thereof, so as to obtain a plurality of basic component package global functional semantic coding feature vectors; Based on the one-way interactive matching of the demand matching game logic semantic coding feature vector with respect to the global function semantic coding feature vectors of the multiple basic component packages, it is determined whether to return the alternative basic component package.

9. The modular game development engine method according to claim 8, characterized in that: Based on the text description of the project requirements, generate game logic to obtain a text description of game logic that matches the requirements, including: The text description of the project requirements is input into a game logic generator based on the AIGC model to obtain a game logic text description matching the requirements.

Images