A game simulation software architecture based on SOA

Abstract

The application discloses a SOA-based game simulation software architecture, and relates to the technical field of game confrontation simulation, and the system comprises a virtual resource layer, a data management layer, a data mining layer and a user interface layer which are connected in sequence; the virtual resource layer is used for abstracting distributed entity game resources into virtual game resources through virtualization technology; the data management layer is used for storing game data generated by the virtual game resources through a network server; the data mining layer is used for performing space-time anomaly detection, co-occurrence pattern mining or correlation model mining on the game data according to demand data, so as to obtain space-time anomaly detection results, co-occurrence pattern mining results or correlation model mining results; and the user interface layer is used for outputting the space-time anomaly detection results through a first service interface, or outputting the co-occurrence pattern mining results through a second service interface, or outputting the correlation model mining results through a third service interface according to user demand, and the application improves resource utilization.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of game confrontation simulation, and particularly relates to a game simulation software architecture based on SOA. BACKGROUND

[0002] The core function of the game system software is intelligent management and reuse of game data, and effective game algorithms depend on effective processing of game data. Traditional game systems store and process game data relatively independently, and cannot realize sharing of game data among different game subjects and effective use of massive game data. SUMMARY

[0003] The present application aims to provide a game simulation software architecture based on SOA, which can improve resource utilization.

[0004] To achieve the above object, the present application provides the following solutions.

[0005] In a first aspect, the present application provides a game simulation software architecture based on SOA, comprising: a virtual resource layer, a data management layer, a data mining layer and a user interface layer connected in sequence.

[0006] The virtual resource layer is configured to abstract distributed entity game resources into virtual game resources through virtualization technology.

[0007] The data management layer is configured to store game data generated by the virtual game resources through a network server.

[0008] The data mining layer is configured to perform spatio-temporal anomaly detection, co-occurrence pattern mining or association model mining on the game data according to demand data, to obtain spatio-temporal anomaly detection results, co-occurrence pattern mining results or association model mining results. The spatio-temporal anomaly detection is configured to detect abnormal behaviors or abnormal events in the time dimension and the space dimension. The abnormal behavior is a behavior that exceeds a preset normal behavior range, and the abnormal event is an event that exceeds a preset normal event range. The co-occurrence pattern mining is configured to detect element combinations in the game data that appear more than a set frequency. The elements in the element combinations are data in the game data. The association model mining is configured to detect elements having an association relationship in the game data. The demand data includes a preset task demand.

[0009] The user interface layer is configured to output the spatio-temporal anomaly detection results through a first service interface, or output the co-occurrence pattern mining results through a second service interface, or output the association model mining results through a third service interface, according to user demand.

[0010] The virtual resource layer, the data management layer, the data mining layer and the user interface layer are all service-based structures.

[0011] Optionally, in terms of storing the game data generated by the virtual game resource, the data management layer is specifically configured to:

[0012] convert the game data of different sources into game data in a set standard format to obtain format-converted game data;

[0013] convert the unstructured or semi-structured data in the format-converted game data into structured data;

[0014] filter the structured data according to a set condition to obtain filtered data;

[0015] clean the multiple pieces of filtered data, and merge the cleaned data according to a set rule to obtain merged processed data;

[0016] classify the multiple pieces of merged processed data according to a set attribute to obtain the categories of the multiple pieces of merged processed data;

[0017] store the merged processed data and the categories thereof.

[0018] Optionally, the virtual game resource includes a self flight simulator and an opponent flight simulator, and the game data includes flight data generated by the self flight simulator and flight data generated by the opponent flight simulator.

[0019] Optionally, the demand data further includes game signals, and the game signals include response signals fed back by the opponent flight simulator when the self flight simulator inputs incentive signals to the opponent flight simulator.

[0020] Optionally, the preset task demand includes a target, a constraint condition, a priority of each subtask, and a resource specification, the target includes a simulated multi-aircraft online strategy game battle process, the constraint condition includes an upper limit value of a response time of each aircraft, and the resource specification includes a display memory size.

[0021] Optionally, the data mining layer includes a spatio-temporal anomaly detection model, the spatio-temporal anomaly detection model is configured to monitor abnormal behaviors and abnormal events of input game data and output spatio-temporal anomaly detection results, the spatio-temporal anomaly detection model is obtained by training a first historical game data set and a first machine learning algorithm, and the first machine learning algorithm is an isolation forest algorithm.

[0022] Optionally, the data mining layer includes a co-occurrence pattern mining unit, and the co-occurrence pattern mining unit is configured to:

[0023] detecting a frequent item set from the game data by using an Apriori algorithm or an FP-growth algorithm;

[0024] extracting an association rule containing the demand data from the frequent item set, and outputting a co-occurrence pattern mining result.

[0025] Optionally, the second service interface displays the co-occurrence pattern mining result in a chart.

[0026] Optionally, the data mining layer comprises an association model mining unit configured to input the demand data into an association model to output a predicted behavior, the association model being a prediction model constructed according to a regression analysis, a causal inference or a machine learning model, the machine learning model being obtained by training a second historical game data set and a second machine learning algorithm, the second machine learning algorithm being a deep learning model or a random forest model.

[0027] Optionally, the first service interface, the second service interface and the third service interface all adopt different levels of services, and the different levels of services comprise software as a service, platform as a service and infrastructure as a service.

[0028] According to the specific embodiments provided in the application, the following technical effects are disclosed:

[0029] The application provides a game simulation software architecture based on a service-oriented architecture (SOA), abstracts distributed entity game resources into virtual game resources by using a virtualization technology, stores game data generated by the virtual game resources by using a network server, overcomes the problem that game data between different game subjects cannot be shared and game resources cannot be effectively utilized, and improves resource utilization rate; in addition, a first service interface is used to output a spatio-temporal anomaly detection result, a second service interface is used to output a co-occurrence pattern mining result, or a third service interface is used to output an association model mining result according to user demand, so that directional utilization of massive game data is realized, and functional services for users are realized. BRIEF DESCRIPTION OF DRAWINGS

[0030] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art, the following will briefly introduce the drawings needed in the embodiments. Obviously, the drawings in the following description are only some embodiments of the application, and for those skilled in the art, other drawings can also be obtained from these drawings without creative labor.

[0031] Figure 1A structural schematic diagram of a game simulation software architecture based on SOA is provided for an embodiment of the present application.

[0032] Figure 2 A functional schematic diagram of a game simulation software architecture based on SOA is provided for an embodiment of the present application.

[0033] Figure 3 A remote service game architecture schematic diagram is provided for an embodiment of the present application. DETAILED DESCRIPTION

[0034] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0035] The above purposes, features and advantages of the present application can be more obvious and easy to understand. The present application will be described in further detail below with reference to the drawings and specific embodiments.

[0036] The present application provides a game simulation software architecture based on SOA. The present application can provide services by dividing different functional levels and define corresponding data interfaces, which together constitute the basic framework of the game simulation software architecture based on SOA. Figure 1 As shown in the figure, the game simulation software architecture based on SOA includes a virtual resource layer, a data management layer, a data mining layer and a user interface layer connected in sequence.

[0037] The virtual resource layer is used to abstract distributed entity game resources into virtual game resources through virtualization technology.

[0038] The data management layer is used to store game data generated by virtual game resources through a network server.

[0039] The data mining layer is used to perform spatio-temporal anomaly detection, co-occurrence pattern mining or association model mining on the game data according to demand data, to obtain spatio-temporal anomaly detection results, co-occurrence pattern mining results or association model mining results. The spatio-temporal anomaly detection is used to detect abnormal behaviors or abnormal events in the time dimension and the space dimension. The abnormal behavior is a behavior that exceeds the preset normal behavior range, and the abnormal event is an event that exceeds the preset normal event range. The co-occurrence pattern mining is used to detect element combinations in the game data that appear more than a set frequency. The elements in the element combinations are data in the game data. The association model mining is used to detect elements that have an association relationship in the game data.

[0040] The user interface layer is used to output the spatio-temporal anomaly detection results through the first service interface, or output the co-occurrence pattern mining results through the second service interface, or output the association model mining results through the third service interface according to user needs.

[0041] SOA is a methodology for designing and implementing software systems that emphasizes the communication of different functional units of an application (called services) through well-defined and service protocols. These services can be loosely coupled and can be developed, deployed, and managed independently while being reusable across functional boundaries.

[0042] In the SOA-based game simulation software architecture, each layer (virtual resource layer, data management layer, data mining layer, and user interface layer) embodies the principles of SOA through service-oriented means. SOA emphasizes exposing different functional units of an application through service interfaces and allowing these services to be called by other applications in a uniform manner.

[0043] Therefore, the virtual resource layer, data management layer, data mining layer, and user interface layer in this application expose different functional units of an application through service interfaces through service-oriented means.

[0044] 1. The virtual resource layer is responsible for managing and scheduling physical resources, abstracting them as virtual resources, and providing them to upper layer services. SOA can be implemented in the virtual resource layer in the following ways:

[0045] 1) Service interface: The service interface provided by the virtual resource layer is standardized, such as exposing resource management functions through a Web service interface (SOAP / RESTful API).

[0046] 2) Service registration: Virtual resource services are registered in a service registry center so that other services can discover and call these services.

[0047] 3) Service encapsulation: The management functions of physical resources are encapsulated as independent services, such as "resource allocation service" and "resource monitoring service".

[0048] 4) Service orchestration: Allows collaboration between services, such as dynamically allocating resources according to task requirements.

[0049] 2. The data management layer is responsible for data storage, retrieval, standardization, and processing functions. The implementation of SOA in the data management layer includes:

[0050] 1) Data service: Provides standardized data access interfaces, such as Create, Read, Update, and Delete (CRUD) operations, which are exposed to other services through a Web service interface.

[0051] 2) Data Integration Services: Provide data integration functions such as data cleaning, transformation, and loading (ETL) services.

[0052] 3) Metadata Services: Provide metadata management services to describe the structure, properties, and other information of data.

[0053] 4) Data Security Services: Provide data encryption, access control, and other security services.

[0054] 3、Data Mining Layer: Responsible for extracting useful information and knowledge from data. SOA implementation in the data mining layer includes:

[0055] Mining Services: Provide data mining functions such as anomaly detection, pattern recognition, predictive modeling, and other services.

[0056] Model Management Services: Provide services for model training, validation, and deployment.

[0057] Result Publishing Services: Publish the results of mining through service interfaces for other services to call.

[0058] Analysis Services: Provide data analysis tools and services such as statistical analysis, visual analysis, and other services.

[0059] 4、User Interface Layer: Responsible for interacting with users, providing user-friendly interfaces and services. SOA implementation in the user interface layer includes:

[0060] 1) User Interface Services: Provide services for building and managing user interfaces such as page rendering services.

[0061] 2) User Authentication Services: Provide user identity verification and permission management services.

[0062] 3) Personalization Services: Provide personalized services based on user preferences such as recommendation services.

[0063] 4) Service Integration: Combine underlying services into user-required application services, such as providing customized solutions by combining virtual resources, data management, and data mining services.

[0064] In the above layers, the key features of SOA are reflected in the following aspects:

[0065] 1) Service Reuse: By modularizing functions into independent services, service reuse can be achieved, reducing repeated development.

[0066] 2) Loose Coupling: Services maintain loose coupling between each other, reducing system complexity and improving flexibility.

[0067] 3) Standardized interface: all services expose functions through standardized interfaces, facilitating interoperation between services.

[0068] 4) Service autonomy: each service is self-contained, with its own lifecycle management.

[0069] 5) Service discovery: service registry provides service discovery functions, facilitating lookup and invocation between services.

[0070] By serviceizing the virtual resource layer, data management layer, data mining layer, and user interface layer, a flexible, scalable, and efficient game simulation software architecture can be constructed. This architecture not only better manages resources and data, but also provides customized services to users, improving user experience.

[0071] The present application abstracts distributed entity game resources into virtual game resources through virtualization technology, and stores game data generated by virtual game resources through a network server, overcoming the problem of inability to share game data between different game entities and different effective utilization of game resources, improving resource utilization. In addition, according to user needs, the spatiotemporal anomaly detection results are output through the first service interface, or the co-occurrence pattern mining results are output through the second service interface, or the association model mining results are output through the third service interface, realizing the directional utilization of massive game data and providing functional services to users.

[0072] The requirements of the game simulation software architecture design based on SOA include: (1) data encryption to ensure the information security of game processes and data; (2) convenient and easy to use, in line with user's daily usage habits; (3) stable and reliable, ensuring smooth operation of the game software; (4) upgradeable and expandable, using an open software architecture based on SOA framework, which can match new targets through functional expansion.

[0073] The design idea of the present application is to abstract distributed game resources according to signal types, solving the problem of high coupling between game resources and game tasks in traditional game simulation software architecture based on SOA. The game simulation software architecture based on SOA has the following outstanding features: (1) using a network server to replace the game entity - microprocessor or computer in traditional game simulation software architecture based on SOA; (2) virtualizing game resources according to signal types and connecting them to a shared resource pool, providing services to all users; (3) providing networked services, solving the information silo problem between traditional game simulation software architectures based on SOA, and providing the possibility of separating game entities from data processing. The network server stores massive game data and effectively stores, manages, and processes these game data, realizing various functions of the game software, such asFigure 2 As shown.

[0074] The signal type of the game resource refers to the various attributes or characteristics used to describe the game resource, which can be classified according to different application scenarios and needs. After virtualizing the game resource according to the signal type and accessing the shared resource pool, it means that these resources will be abstracted into a series of signals in order to better manage and allocate to different users. The signal type can include but is not limited to the following aspects:

[0075] 1. Physical signal type includes hardware resources and environmental conditions.

[0076] Hardware resources: including computing resources (such as CPU, GPU), storage resources (such as hard disk capacity, SSD capacity), network resources (such as bandwidth, delay) and so on.

[0077] Environmental conditions: including temperature, humidity and other environmental parameters, which are very important for some sensitive game environments.

[0078] 2. Logical signal type includes software environment and data set.

[0079] Software environment: including operating system type, programming language version, middleware configuration, etc.

[0080] Data set: including data format, size, storage location, etc.

[0081] 3. Functional signal type includes service type and computing task.

[0082] Service type: including API service, message queue service, database service, etc.

[0083] Computing task: including machine learning task, image processing task, etc.

[0084] 4. State signal type includes device state and resource usage.

[0085] Device state: including the online state, health status of the device, etc.

[0086] Resource usage: such as CPU utilization, memory occupancy, etc.

[0087] 5. Behavior signal type includes user behavior and player behavior.

[0088] User behavior: including user login frequency, operation habit, etc.

[0089] Player behavior: including movement path, attack mode in the game, etc.

[0090] 6. Control signal type includes instruction set and configuration parameter.

[0091] Instruction Set: Includes instructions for controlling devices or services.

[0092] Configuration Parameters: Includes threshold settings, priority settings, etc.

[0093] 7. Interaction Signal Types include communication protocols and interface specifications.

[0094] Communication Protocols: Includes network communication protocols, messaging protocols, etc.

[0095] Interface Specifications: Includes RESTful APIs, SOAP services, etc.

[0096] The SOA-based game simulation software architecture also includes Runtime Service located in the user interface layer and the data mining layer, which belongs to the core layer of the architecture.

[0097] The main role of the Runtime Service is to coordinate and manage various services and resources during the running of the game simulation software, and to monitor and optimize the system running state to ensure that the software can run efficiently and stably. Specifically, the role of the Runtime Service includes but is not limited to:

[0098] 1) Resource scheduling and management: dynamically allocate and schedule resources according to the current resource state and task demand.

[0099] 2) Service coordination: coordinate the interaction between different services to ensure data exchange and control signal transmission between services.

[0100] 3) State monitoring: monitor the running state of the entire system, including resource usage, service health status, etc.

[0101] 4) Fault recovery: when a fault or exception occurs, it can quickly locate the problem and take measures to restore the service.

[0102] 5) Performance optimization: adjust resource allocation strategies according to system running conditions to optimize performance.

[0103] The Runtime Service is used to coordinate the interaction between the resource mapping service and the game confrontation service.

[0104] The resource mapping service is used to map the task requirements of the game task to specific resources. The Runtime Service will call the resource mapping service to select the optimal resource configuration scheme according to the complexity of the task requirements and the state of the resources.

[0105] The game confrontation service is used to provide online game confrontation capabilities. The Runtime Service interacts with the game confrontation service to ensure real-time processing and feedback of data during the game process.

[0106] The runtime service acts as an intermediary layer to isolate the user interface layer from the data mining layer, reducing direct dependencies between the two layers and improving system flexibility and maintainability.

[0107] The runtime service is responsible for coordinating key components such as resource mapping services and game confrontation services, ensuring the correct flow of data and control signals.

[0108] The runtime service can monitor system status and perform resource optimization and fault recovery as needed, improving system stability and performance.

[0109] The user interface layer is responsible for interacting with users and providing customized services.

[0110] The data mining layer is responsible for extracting useful information from game data to provide decision support for the runtime service.

[0111] This layered design helps to clearly separate different functions and services, making the software more modular, easy to extend and maintain.

[0112] The SOA-based game simulation software architecture calls the runtime service state by inputting game resource descriptions (resource descriptions) and task requirement descriptions (task descriptions). First, the game resource management service is performed, which mainly manages the game resources of the game equipment, standardizes the storage and processing of game information, and calls the resource mapping service according to the task requirements to map the optimized game resources for the corresponding game signals; secondly, game data management is performed, the core function of which is to realize online game confrontation service; finally, according to the complexity of the game task, the runtime service interacts with the resource mapping service and the game confrontation service in real time to realize dynamic processing of the game process.

[0113] The task description includes a signal set and a game sequence, where the game sequence refers to a sequence of game tasks. The signal set refers to a set of game signals.

[0114] Among them, the game resource description refers to the information description of various physical or logical resources that can be used in the game process. These game resources can include hardware devices (such as computing nodes, network bandwidth), software environments (such as operating systems, middleware), and data sets, etc.

[0115] Hardware resources include server CPU model, GPU quantity, memory size, and storage space size, etc. Software environment includes supported operating system version and installed software library version, etc. Data set includes the format, size, and storage location of specific types of data sets that can be used for training or testing, etc.

[0116] The preset task requirements (task requirement description) include objectives, constraints, subtask priorities, and resource specifications. The objectives include simulating the combat process of a multi-aircraft online strategy game, constraints include the upper limit of each aircraft's response time, and resource specifications include video memory size. For example, the resource specification requires at least four GPU nodes to participate in the calculation, each with at least 32GB of video memory. The priority of each subtask is used to determine the order in which they are completed.

[0117] Game signals refer to various status or control information generated during the game. This information can be used to guide the game's progress or adjust game strategies. For example, game signals include status information, control information, and environmental feedback. Status information includes the player's current position, health points, and remaining ammunition; control information includes movement and attack commands issued by the player; and environmental feedback includes changes in the game environment, such as terrain and weather.

[0118] In this SOA-based game simulation software architecture, game resource descriptions and task requirement descriptions provide the fundamental information and support necessary for gaming. Game signals are key data generated during the game and used to adjust the progress. Through the interaction between this information and services, the software enables the effective management and simulation of complex gaming scenarios.

[0119] In the SOA-based game simulation software architecture, the virtual resource layer separates game resources from the data management layer through the network. The SOA-based game simulation software architecture builds a virtual resource pool through the virtual resource layer, which can access standardized devices from different regions, users, and manufacturers, and centrally manage and utilize them through virtualization technology. Standardized devices are abstracted into virtual game resources. When new devices are needed, the virtual resource layer abstracts the device entity into a virtual game resource. The abstracted information includes the device entity, the number of connections, the connectivity relationship, and the exchange protocol.

[0120] Game data has diverse data structures, diverse sampling methods, and a lot of repeated and redundant information, which brings many difficulties to the management of game data. Therefore, the data management layer is required to process and store game data.

[0121] In an exemplary embodiment, in terms of storing gaming data generated by virtual gaming resources, the data management layer is specifically used to:

[0122] 1) Standardization: that is, all gaming data from different sources are converted into gaming data in a set standard format to obtain gaming data after format conversion; standardization is to ensure that all data follows a unified standard or format, which can make the data more consistent and facilitate subsequent processing and analysis.

[0123] The standardization process includes first defining a standard format, such as using JSON or XML to represent data, and then converting data from different sources into a unified format. For numerical data, scaling or normalization may be required. For example, if there are different scoring systems in the game data, they can be converted to a 0-100 score range.

[0124] 2) Structuring: converting unstructured or semi-structured data in the formatted game data into structured data.

[0125] The structuring process includes storing data in a relational database or NoSQL database, designing appropriate table structures or document models, and indexing common fields to improve query efficiency. For example, for player behavior records, these data can be structured by creating a table containing player ID, behavior type, timestamp, etc.

[0126] 3) Filtering: filtering the structured data according to the set conditions to obtain filtered data.

[0127] Filtering is to select game data that meets the set conditions from a large amount of game data.

[0128] The filtering process includes setting filtering rules (set conditions) according to requirements, such as retaining only the last week's data, using SQL or other query languages to extract the required data subset, and removing duplicate data entries. For example, in a multiplayer online game match, only the top ten players' detailed records may be of interest, i.e. filtering out the game data of the top ten players.

[0129] 4) Integration: the process of merging game data from multiple data sources into a unified view, including data cleaning, data fusion, and data mapping. Data cleaning is used to solve data inconsistency problems, such as unit inconsistency, spelling errors, etc. Data fusion combines related game data to form a more complete information set, i.e. cleaning the filtered data and merging the cleaned data according to the set rules to obtain the merged data. Data mapping is used to ensure that game data fields from different sources can be correctly mapped. For example, if game data comes from different virtual sensors, these data need to be integrated to form a unified game state description. Virtual sensors are virtual sensors on a flight simulator, and the data from each virtual sensor is mapped to form unified flight data, which represents the game state of the flight simulator.

[0130] 5) Classification: The multiple pieces of merged processed data are classified according to the set attributes, obtaining the categories of each piece of merged processed data.

[0131] The classification process includes applying clustering algorithms or classification algorithms to automatically identify the categories of the data, or classifying by manual marking. For example, for the skill level of a player, it can be divided into several levels such as beginner, intermediate, advanced, etc.

[0132] The merged processed data and its categories are stored.

[0133] Storage refers to saving data on a persistent storage medium for long-term storage and ready access. According to the category of game data (merged processed data), choose a relational database, file system or cloud storage, etc. Regularly back up the game data and implement data encryption and access control mechanisms. Game data can be stored in high-performance relational databases to support fast queries and transaction processing.

[0134] Through the above processing of game data, the quality and usability of data can be greatly improved, thereby better supporting the functional requirements of game simulation software.

[0135] In an exemplary embodiment, the virtual game resources include a self flight simulator and an opponent flight simulator, and the game data includes flight data generated by the self flight simulator and flight data generated by the opponent flight simulator.

[0136] The demand data further includes game signals, and the game signals include response signals fed back by the opponent flight simulator when the self flight simulator inputs incentive signals to the opponent flight simulator.

[0137] The data mining layer mines useful information from the game data according to the demand, realizes multi-state modeling and self-updating of resource state, realizes directional utilization of massive game data, and realizes functional services for users. The resource state includes available, unavailable, etc.

[0138] The data mining layer includes a spatiotemporal anomaly detection model, which is used for monitoring abnormal behavior and abnormal events of input game data, and outputs spatiotemporal anomaly detection results; the spatiotemporal anomaly detection model is obtained by training a first historical game data set and a first machine learning algorithm; the first machine learning algorithm is an isolation forest algorithm.

[0139] Temporal-spatial anomaly detection aims to identify abnormal behavior or events in both time and space dimensions. This can help uncover behaviors that deviate from regular patterns, revealing potential problems or opportunities. The construction of a temporal-spatial anomaly detection model specifically includes: using historical game data to establish a baseline model of normal behavior; monitoring real-time data streams and comparing them with the baseline model. Statistical methods such as Z-score, IQR, etc. are used to quantify the degree of abnormality. Supervised learning or unsupervised learning techniques are used to identify anomalies. For example, in a multiplayer online game, if a player achieves an abnormally high score in a short period of time, this may be identified as abnormal behavior.

[0140] Real-time monitoring refers to continuously monitoring data streams, while statistical methods are used to assess whether these data streams deviate from expected behavior.

[0141] Co-occurrence pattern mining focuses on combinations of elements that frequently appear in game data. This is useful for understanding the interactions between different factors.

[0142] The data mining layer includes a co-occurrence pattern mining unit that uses the Apriori algorithm or the FP-growth algorithm to detect frequent itemsets from the game data; extracts meaningful association rules from the frequent itemsets, and outputs the co-occurrence pattern mining results. Meaningful association rules are those that reveal useful or interesting relationships in the data set. In the co-occurrence pattern mining of game data, meaningful association rules can help us understand the potential connections between different elements, providing valuable insights for decision support, optimizing resource allocation, etc. Association rules are relationships between groups of items discovered in a data set. A typical association rule can be expressed as "if A occurs, then B is likely to occur", where A and B are items in the data set.

[0143] The second service interface uses a chart to display the co-occurrence pattern mining results.

[0144] For example, if it is found that players A and B often appear in the same game at the same time, and when they play together, the win rate significantly increases, this is a co-occurrence pattern.

[0145] Association model mining aims to discover dependencies between variables, which helps to predict future behavior or outcomes.

[0146] The data mining layer includes a correlation model mining unit that inputs the demand data into a correlation model to output predicted behavior. The correlation model is a prediction model constructed based on regression analysis, causal inference, or a machine learning model trained based on a second historical game data set and a second machine learning algorithm. The second machine learning algorithm is a deep learning model or a random forest model. For example, by analyzing historical behavior data of players, a model can be established to predict the actions that players may take in specific situations.

[0147] The data mining layer also has the functions of polymorphic modeling and self-updating. Polymorphic modeling refers to dynamically adjusting the structure and parameters of the spatiotemporal anomaly detection model and the prediction model according to different data and contexts. Self-updating refers to the ability of the anomaly detection model and the prediction model to automatically adjust themselves according to new data to adapt to changing environments.

[0148] The data mining layer provides targeted utilization and functional services, aiming to use the information mined from game data to provide targeted services, including:

[0149] Personalized recommendations: providing personalized game recommendations based on player preferences and behavior patterns.

[0150] Intelligent decision support: providing players with strategy suggestions based on data analysis.

[0151] Performance optimization: optimizing system configuration and resource allocation based on model prediction results.

[0152] User experience improvement: improving game design and interface layout by analyzing user feedback.

[0153] For example, based on players' historical performance and preferences, the SOA-based game simulation software architecture can recommend the most suitable game mode or opponent for them to improve their gaming experience.

[0154] This application can extract valuable information from game data and use it to improve decision-making processes, optimize resource utilization, and enhance user experience.

[0155] The first service interface, the second service interface and the third service interface all adopt different levels of services, including Software as a Service (SaaS), Platform as a Service (PaaS) and Infrastructure as a Service (IaaS). Through layered services, customized functional services are provided for different users, improving the utilization rate of game data and meeting the use requirements of different users.

[0156] There is a clear data logic relationship between the data mining layer and the user interface layer. This relationship ensures that useful information obtained from the data mining layer can be effectively passed to the user interface layer, thereby meeting the needs of different users.

[0157] The data mining layer is responsible for processing and analyzing raw data to extract useful information and knowledge. This information and knowledge is usually highly abstracted and processed data that can be used to support decision-making, optimize processes or provide personalized services. The main tasks of the data mining layer include but are not limited to: 1) data preprocessing: including data cleaning, data conversion, data integration, etc., to ensure data quality and consistency. 2) Data mining: using various algorithms and techniques to discover patterns, trends and associations in data. 3) Model building: building models for prediction, classification, clustering, etc. 4) Result interpretation: interpreting the mined information to make it easy to understand and apply.

[0158] The user interface layer is the interaction point between the user and the system, and it is responsible for receiving user requests and presenting the processed results to the user. The main responsibilities of the user interface layer include: 1) User authentication and authorization: Ensure that only authenticated users can access corresponding services. 2) User interface design: Provide a friendly user interface so that users can easily use the functions provided by the system. 3) Service customization: Provide customized services according to user needs and preferences. 4) Result display: Display the results produced by the data mining layer in a user-friendly manner.

[0159] In this application, the data logical relationship between the data mining layer and the user interface layer includes: 1) information flow: the data results processed by the data mining layer (such as anomaly detection results, co-occurrence patterns, association models, etc.) are passed as input to the user interface layer, which provides customized services for users according to these information. 2) demand matching: the useful information mined by the data mining layer needs to be matched with the user's demand. For example, if the user needs to understand the trend analysis of player behavior, the data mining layer needs to provide the corresponding time series analysis results. 3) service level: different levels of services (SaaS, PaaS, IaaS) will obtain different types of information from the data mining layer according to the user's demand. For example, SaaS may focus on providing high-level analysis results, while PaaS may need more underlying data processing capabilities and API interfaces.

[0160] The relationship between useful information and user demand includes direct and indirect relationships.

[0161] Direct relationship: the useful information mined by the data mining layer directly supports the services provided by the user interface layer. For example, if the user needs to understand the behavior patterns of players under certain conditions, the data mining layer can provide relevant co-occurrence patterns and association models, and the user interface layer will display these information to the user in the form of graphs or reports.

[0162] Indirect relationship: in some cases, the information generated by the data mining layer may need to be further processed to meet the user's demand. For example, the data mining layer may identify a new player behavior pattern, but the user interface layer needs to compare this pattern with the user's actual behavior to provide more specific suggestions or recommendations.

[0163] There is a close data logical relationship between the data mining layer and the user interface layer, the data mining layer provides valuable information to users, and the user interface layer converts these information into user-operable services, both of which work together to meet the different needs of users.

[0164] The information flow of the entire SOA-based game simulation software architecture can be described as follows: (1) connect game resources and game objects through game resource allocation, get responses through input incentives, and obtain game signals; (2) abstract distributed entity game resources into virtual game resources through virtualization technology, and describe game resources and game signals through standard XML; (3) manage multi-dimensional game data through data processing of game data obtained from the virtual resource pool; (4) select different data mining methods for different game needs; (5) call game data mining results according to different service interface needs; (6) provide services with different permission levels according to user needs, and access corresponding service interfaces.

[0165] The game object refers to the opponent flight simulator.

[0166] The different levels of services provided by the virtual resource layer, the data management layer, the data mining layer and the user interface layer are integrated by the present application to form the SOA-based game simulation software architecture with game data as the core. The architecture can effectively improve the storage and management of game data, and lay a foundation for realizing more efficient and reliable online game confrontation. In addition, due to the openness of the architecture, it supports continuous extension and expansion in the later stage, so that the SOA-based game simulation software architecture based on the SOA framework has good application prospects.

[0167] The SOA-based game simulation software architecture needs to provide a friendly man-machine interface and perfect game function services for users, allow users to flexibly view and manage game data, and enable game resource management and online game. The SOA-based game simulation software architecture of the present application is a web service structure, developed using the Browser / Server mode, based on the Visual C++ 6.0 environment, can effectively adapt to different terminal devices, and realizes the isolation of terminal devices and game data, further guarantees the data security, and is convenient for future upgrading and software updating iteration of the SOA-based game simulation software architecture. The SOA-based game simulation software architecture program, the running environment and the database are all deployed on the network server of the off-site detection device, and the overall implementation architecture of the SOA-based game simulation software architecture is as shown in Figure 3

[0168] In the architecture of the SOA-based game simulation software architecture, the remote user accesses the SOA-based game simulation software architecture software on the Web browser through the network, connects with the network server through the TCP / IP protocol, realizes the control of the remote operation interface and data operation through the ActiveX control and the ADO control on the network server, connects the network server and the embedded controller through the database service to realize the data storage service, and realizes the packaging and transmission of the game data based on the local program through the related signal characteristics of the game resources in the embedded controller. The network connection is realized by the switch, so that the user can remotely access the SOA-based game simulation software architecture through the webpage.

[0169] The technical features of the above embodiments can be combined in any way. To make the description concise, not all possible combinations of the technical features in the above embodiments are described, but as long as the combinations of the technical features do not exist contradictions, they should be considered as the scope of the present application.

[0170] ​The principles and implementation manners of the present application are described herein by using specific examples, and the above examples are only used to help understand the method of the present application and its core idea; meanwhile, for those skilled in the art, according to the idea of the present application, the specific implementation manners and application ranges will have changes. In conclusion, the content of the specification should not be understood as a limitation of the present application.

Claims

1. A game simulation software architecture based on SOA, characterized by: The SOA-based game simulation software system architecture includes a virtual resource layer, a data management layer, a data mining layer and a user interface layer connected in sequence; The virtual resource layer is used to abstract distributed physical gaming resources into virtual gaming resources through virtualization technology; The data management layer is used to store the gaming data generated by the virtual gaming resources through the network server; The data mining layer is used to perform spatiotemporal anomaly detection, co-occurrence pattern mining, or association model mining on the gaming data according to the demand data, and obtain spatiotemporal anomaly detection results, co-occurrence pattern mining results, or association model mining results; the spatiotemporal anomaly detection is used to detect abnormal behaviors or abnormal events in the time dimension and space dimension, wherein the abnormal behaviors are behaviors that exceed the preset normal behavior range, and the abnormal events are events that exceed the preset normal event range; The co-occurrence pattern mining is used to detect element combinations in the game data that appear more than a set frequency, where the elements in the element combinations are data in the game data; The association model mining is used to detect elements with association relationships in the game data; The demand data includes preset task requirements; The user interface layer is used to output spatiotemporal anomaly detection results through the first service interface, or output co-occurrence pattern mining results through the second service interface, or output association model mining results through the third service interface according to user needs; The virtual resource layer, the data management layer, the data mining layer and the user interface layer are all service-oriented structures.

2. The SOA-based game simulation software architecture according to claim 1 is characterized in that: In terms of storing the gaming data generated by the virtual gaming resources, the data management layer is specifically used to: The gaming data from different sources are converted into gaming data in a set standard format to obtain the gaming data after format conversion; Converting unstructured or semi-structured data in the gaming data after format conversion into structured data; Filtering the structured data according to set conditions to obtain filtered data; Clean the multiple filtered data and merge the cleaned data according to the set rules to obtain the merged data; Classify multiple merged data according to the set attributes to obtain the category of each merged data; The merged processed data and their categories are stored.

3. The SOA-based game simulation software architecture according to claim 1, characterized in that: The virtual game resources include our flight simulator and the opponent's flight simulator, and the game data include the flight data generated by the our flight simulator and the flight data generated by the opponent's flight simulator.

4. The SOA-based game simulation software architecture according to claim 3 is characterized in that: The demand data also includes a game signal, and the game signal includes a response signal fed back by the opponent flight simulator when the own flight simulator inputs an excitation signal to the opponent flight simulator.

5. The SOA-based game simulation software architecture according to claim 1, characterized in that: The preset task requirements include objectives, constraints, priorities of each subtask and resource specifications. The objectives include simulating the combat process of a multi-aircraft online strategy game, the constraints include the upper limit of the response time of each aircraft, and the resource specifications include the size of the video memory.

6. The SOA-based game simulation software architecture according to claim 1, characterized in that: The data mining layer includes a spatiotemporal anomaly detection model, which is used to monitor abnormal behaviors and abnormal events in the input game data and output spatiotemporal anomaly detection results; the spatiotemporal anomaly detection model is trained using a first historical game data set and a first machine learning algorithm; the first machine learning algorithm is an isolation forest algorithm.

7. The SOA-based game simulation software architecture according to claim 1, characterized in that: The data mining layer includes a co-occurrence pattern mining unit, which is used to: Detecting frequent itemsets from the game data using an Apriori algorithm or an FP-growth algorithm; Extract association rules containing the demand data from the frequent item sets, and output co-occurrence pattern mining results.

8. The SOA-based game simulation software architecture according to claim 7, characterized in that: The second service interface uses a chart to display the co-occurrence pattern mining results.

9. The SOA-based game simulation software architecture according to claim 1, characterized in that: The data mining layer includes an association model mining unit, which is used to input the demand data into an association model to output predicted behavior. The association model is a prediction model constructed based on regression analysis, causal inference or a machine learning model. The machine learning model is trained based on a second historical game data set and a second machine learning algorithm; the second machine learning algorithm is a deep learning model or a random forest model.

10. The SOA-based game simulation software architecture according to claim 1, characterized in that: The first service interface, the second service interface and the third service interface all adopt services at different levels, and the services at different levels include software as a service, platform as a service and infrastructure as a service.

Images