Game simulation software system architecture based on SOA

Abstract

The invention discloses a game simulation software system architecture based on SOA, 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 a 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 mode mining or association model mining on the game data according to the demand data to obtain a space-time anomaly detection result, a co-occurrence mode mining result or an association model mining result; and the user interface layer is used for outputting a space-time anomaly detection result through the first service interface according to a user demand, or outputting a co-occurrence mode mining result through the second service interface or outputting an association model mining result through the third service interface. The resource utilization rate is improved.

Description

Technical Field

[0001] The present application relates to the technical field of game confrontation simulation, and in particular to a game simulation software system architecture based on SOA. Background Art

[0002] The core function of the game system software is the intelligent management and reuse of game data. Effective game algorithms rely on the effective processing of game data. Traditional game systems store and process game data in a relatively isolated manner, which cannot achieve the sharing of game data between different game entities and the effective use of massive game data. Summary of the invention

[0003] The purpose of this application is to provide a game simulation software architecture based on SOA, which can improve resource utilization.

[0004] To achieve the above objectives, this application provides the following solutions:

[0005] In a first aspect, the present application provides a game simulation software system architecture based on SOA, including: 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 used to abstract the distributed physical gaming resources into virtual gaming resources through virtualization technology;

[0007] The data management layer is used to store the gaming data generated by the virtual gaming resources through the network server;

[0008] The data mining layer is used to perform spatiotemporal anomaly detection, co-occurrence pattern mining or association model mining on the game data according to the demand data to 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, 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, and the elements in the element combination are data in the game data; the association model mining is used to detect elements with an associated relationship in the game data; the demand data includes preset task requirements;

[0009] The user interface layer is used to output the spatiotemporal anomaly detection result through the first service interface, or output the co-occurrence pattern mining result through the second service interface, or output the association model mining result through the third service interface according to user needs;

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

[0011] Optionally, in terms of storing the gaming data generated by the virtual gaming resources, the data management layer is specifically used for:

[0012] The gaming data from different sources are converted into gaming data in a set standard format to obtain the gaming data after format conversion;

[0013] Convert the unstructured or semi-structured data in the game data after format conversion into structured data;

[0014] Filtering the structured data according to set conditions to obtain filtered data;

[0015] Cleaning the multiple filtered data, and merging the cleaned data according to the set rules to obtain merged data;

[0016] Classify the multiple merged data according to the set attributes to obtain the category of each merged data;

[0017] The merged processed data and their categories are stored.

[0018] Optionally, the virtual game resources include our flight simulator and the opponent's flight simulator, and the game data include flight data generated by our flight simulator and flight data generated by the opponent's flight simulator.

[0019] Optionally, 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.

[0020] Optionally, the preset task requirements include objectives, constraints, priorities of each subtask and resource specifications, the objectives include simulating a combat process of a multi-aircraft online strategy game, the constraints include an upper limit on the response time of each aircraft, and the resource specifications include video memory size.

[0021] Optionally, 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.

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

[0023] Using Apriori algorithm or FP-growth algorithm to detect frequent itemsets from the game data;

[0024] Extract association rules containing the demand data from the frequent item set, and output co-occurrence pattern mining results.

[0025] Optionally, the second service interface uses a chart to display the co-occurrence pattern mining results.

[0026] Optionally, 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, and 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.

[0027] Optionally, 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.

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

[0029] The present application provides a game simulation software system architecture based on a service-oriented architecture (Service-Oriented Architecture, SOA), which abstracts distributed physical game resources into virtual game resources through virtualization technology, and stores game data generated by virtual game resources through a network server, thereby overcoming the problems of the inability to share game data between different game entities and the different effective utilization of game resources, and 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, thereby realizing the targeted utilization of massive game data and realizing functional services for users. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0031] Figure 1A schematic diagram of the structure of a game simulation software system architecture based on SOA provided in one embodiment of the present application;

[0032] Figure 2 A functional schematic diagram of a game simulation software architecture based on SOA provided in one embodiment of the present application;

[0033] Figure 3 A schematic diagram of the remote service gaming system architecture provided in one embodiment of the present application. DETAILED DESCRIPTION

[0034] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0035] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0036] This application provides a game simulation software system architecture based on SOA. This application divides the services that can be provided by different functional levels and defines corresponding data interfaces, which together constitute the basic framework of the game simulation software system architecture based on SOA; Figure 1 As shown, 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.

[0037] The virtual resource layer is used to abstract the distributed physical gaming resources into virtual gaming resources through virtualization technology.

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

[0039] The data mining layer is used to perform spatiotemporal anomaly detection, co-occurrence pattern mining or association model mining on the game data according to the demand data to 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, the abnormal behavior is the behavior that exceeds the preset normal behavior range, and the abnormal event is the 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, and the elements in the element combination are data in the game data; the association model mining is used to detect elements with association relationships in the game data.

[0040] 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.

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

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

[0043] Therefore, in the present application, the virtual resource layer, the data management layer, the data mining layer and the user interface layer all expose different functional units of the application program through service interfaces in a service-oriented manner.

[0044] 1. The virtual resource layer is responsible for managing and scheduling physical resources, abstracting them into 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-oriented interface: The service interface provided by the virtual resource layer is standardized, for example, resource management functions are exposed through Web service interfaces (SOAP / RESTful API).

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

[0047] 3) Service encapsulation: Encapsulate the management functions of physical resources into independent services, such as "resource allocation service", "resource monitoring service", etc.

[0048] 4) Service orchestration: allows collaboration between services, such as dynamically allocating resources based on task requirements.

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

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

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

[0052] 3) Metadata service: Provides metadata management services to describe data structure, attributes and other information.

[0053] 4) Data security services: provide security services such as data encryption and access control.

[0054] 3. The data mining layer is responsible for extracting useful information and knowledge from the data. SOA's implementation of 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 service: provides services for model training, validation, and deployment.

[0057] Result publishing service: Publish the mining results through the service interface for other services to call.

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

[0059] 4. The user interface layer is responsible for interaction with users and provides user-friendly interfaces and services. SOA implementation at the user interface layer includes:

[0060] 1) User interface service: Provides user interface construction and management services, such as page rendering services.

[0061] 2) User authentication service: Provide user identity authentication and authority management services.

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

[0063] 4) Service integration: Combining underlying services into application services required by users, such as providing customized solutions by combining virtual resources, data management and data mining services.

[0064] Among 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 and duplicate development can be reduced.

[0066] 2) Loose coupling: Services are loosely coupled to reduce system complexity and improve flexibility.

[0067] 3) Standardized interfaces: All services expose their functions through standardized interfaces to facilitate interoperability between services.

[0068] 4) Service autonomy: Each service is self-contained and has its own lifecycle management.

[0069] 5) Service discovery: The service registry provides service discovery functions to facilitate the search and calling of services.

[0070] By servitizing 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 can not only better manage resources and data, but also provide users with customized services and improve user experience.

[0071] The present application abstracts distributed physical game resources into virtual game resources through virtualization technology, and stores the game data generated by the virtual game resources through a network server, thereby overcoming the problems of the inability to share game data between different game entities and the different effective utilization of game resources, and 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, so as to realize the targeted utilization of massive game data and provide functional services to users.

[0072] The requirements for the design of the game simulation software system architecture based on SOA in this application include: (1) data encryption to ensure the information security of the game process and data; (2) convenience and ease of use, in line with the daily usage habits of users; (3) stability and reliability to ensure the smooth operation of the game software; (4) upgradeability and expansion, using an open software architecture based on the SOA framework, and matching new goals through functional expansion.

[0073] The design concept of this application is to abstractly describe the distributed game resources according to the signal type, so as to solve the problem of high coupling between game resources and game tasks in the traditional SOA-based game simulation software architecture. This makes the SOA-based game simulation software architecture system have the following outstanding features: (1) The network server is used to replace the game subject - microprocessor or computer in the traditional automatic SOA-based game simulation software architecture; (2) Game resources are virtualized according to signal type and connected to the shared resource pool, which is provided to all users as a service; (3) Providing networked services solves the problem of information islands between traditional SOA-based game simulation software architectures and provides the possibility of separating game entities from data processing. The network server stores massive amounts of game data, and effectively stores, manages and processes the game data to realize various functions of the game software, such as Figure 2 shown.

[0074] The signal type of gaming resources refers to the various attributes or characteristics used to describe gaming resources, which can be classified according to different application scenarios and requirements. After gaming resources are classified and virtualized according to signal types and connected to the shared resource pool, these resources will be abstracted into a series of signals for better management and allocation to different users. Signal types may include but are not limited to the following aspects:

[0075] 1. Physical signal types include 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, latency), etc.

[0077] Environmental conditions: including environmental parameters such as temperature and humidity, which are very important for certain sensitive gaming environments.

[0078] 2. Logical signal types include software environment and data set.

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

[0080] Data set: includes the format, size, storage location, etc. of the data.

[0081] 3. Functional signal types include service types and computing tasks.

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

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

[0084] 4. Status signal types include device status and resource usage.

[0085] Device status: including the device's online status, health status, etc.

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

[0087] 5. Behavioral signal types include user behavior and player behavior.

[0088] User behavior: including user login frequency, operating habits, etc.

[0089] Player behavior: including movement paths, attack patterns, etc. in the game.

[0090] 6. Control signal types include instruction sets and configuration parameters.

[0091] Instruction set: Contains instructions for controlling a device or service.

[0092] Configuration parameters: including threshold settings, priority settings, etc.

[0093] 7. Interaction signal types include communication protocols and interface specifications.

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

[0095] Interface specifications: including RESTful API, SOAP services, etc.

[0096] The SOA-based game simulation software architecture also includes runtime services located at the user interface layer and the data mining layer. The runtime services belong to the core layer of the architecture.

[0097] The main function of the runtime service is to coordinate and manage various services and resources during the operation of the game simulation software, as well as to monitor and optimize the system operation status to ensure that the software can run efficiently and stably. Specifically, the functions of the runtime service include but are not limited to:

[0098] 1) Resource scheduling and management: Dynamically allocate and schedule resources based on current resource status and task requirements.

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

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

[0101] 4) Fault recovery: When a fault or anomaly occurs, the problem can be quickly located and measures can be taken to restore service.

[0102] 5) Performance optimization: Adjust resource allocation strategy according to system operation status 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 solution based on the complexity of the task requirements and the status of the resources.

[0105] The game confrontation service is used to provide the ability of online game confrontation. The runtime service will interact with the game confrontation service to ensure the real-time processing and feedback of data during the game.

[0106] The runtime service acts as an intermediary layer to isolate the user interface layer from the data mining layer, reduce the direct dependency between the user interface layer and the data mining layer, and improve the flexibility and maintainability of the system.

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

[0108] Runtime services can monitor system status and perform resource optimization and fault recovery as needed to improve 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 and providing decision support for runtime services.

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

[0112] The SOA-based game simulation software architecture calls the runtime service status by inputting the game resource description (resource description) and the task requirement description (task description). First, the game resource management service is performed, whose main task is to manage the game resources of the game equipment, store and process the game information in a standardized manner, and call the resource mapping service according to the task requirements to optimize the selected game resources for the corresponding game signal mapping; secondly, the game data management is performed, and its core function is to realize the 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 the dynamic processing of the game process.

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

[0114] 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.

[0115] Hardware resources include server CPU model, number of GPUs, memory size, storage space size, etc. Software environment includes supported operating system versions and installed software library versions, etc. Datasets include the format, size, and storage location of specific types of datasets that can be used for training or testing.

[0116] The preset task requirements (task requirement description) include goals, constraints, priorities of each subtask and resource specifications. The goals include simulating the combat process of multi-aircraft online strategy games, the constraints include the upper limit of the response time of each aircraft, and the resource specifications include the size of video memory. For example, the resource specification is that at least 4 GPU nodes are required to participate in the calculation, and each node has at least 32GB of video memory. The priority of each subtask is used to set the order of completion of each subtask.

[0117] Game signals refer to various states or control information generated during the game process, which can be used to guide the development of the game process or adjust the game strategy. For example, game signals include state information, control information and environmental feedback. State information includes the current player's position, health value, remaining ammunition, etc.; control information includes movement commands and attack commands issued by the player; environmental feedback includes changes in the game environment, such as terrain changes and weather changes.

[0118] In the architecture of this SOA-based game simulation software architecture, game resource description and task requirement description provide the basic information and support necessary for gaming, while game signals are key data generated during gaming and used to adjust the gaming process. Through the interaction between this information and services, the software can effectively manage and simulate complex gaming scenarios.

[0119] In the SOA-based game simulation software architecture, the virtual resource layer separates the game resources from the data management layer through the network. The SOA-based game simulation software architecture can access standardized devices from different regions, different users, and different manufacturers through the virtual resource pool built by the virtual resource layer, and manage and use them in a unified manner based on virtualization technology. Standardized devices will be abstracted into virtual game resources. When new devices are needed, the device entity is abstracted into a virtual game resource through the virtual resource layer. The abstract information includes the device entity, the number of connections of the device entity, the conduction 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 the gaming data generated by the 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 follow a unified standard or format, which can make the data more consistent and facilitate subsequent processing and analysis.

[0123] The process of standardization 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 according to the set standard format. For numerical data, scaling or normalization may be required. For example, if there are different scoring systems in the gaming data, they can be uniformly converted to a score range of 0 to 100.

[0124] 2) Structuring: Convert the unstructured or semi-structured data in the game data into structured data after format conversion.

[0125] The structuring process includes: storing data in a relational database or NoSQL database; designing a suitable data table structure or document model; and indexing commonly used fields to improve query efficiency. For example, for player behavior records, you can structure these data by creating a table containing fields such as player ID, behavior type, and timestamp.

[0126] 3) Screening: Screening the structured data according to set conditions to obtain screened data.

[0127] The purpose of screening is to select gaming data that meets the set conditions from a large amount of gaming data.

[0128] The screening process includes: setting filtering rules (setting conditions) according to needs, such as retaining only the data of the most recent week, using SQL or other query languages ​​to extract the required data subset, and removing duplicate data entries. For example, in a multiplayer online gaming competition, you may only be interested in the detailed records of the top ten players, that is, to filter out the gaming data of the top ten players.

[0129] 4) Integration: The process of merging game data from multiple data sources to form a unified view, specifically including: data cleaning, data fusion and data mapping. Data cleaning is used to solve data inconsistency problems, such as inconsistent units, spelling errors, etc. Data fusion merges related game data together to form a more complete information set, that is, multiple filtered data are cleaned, and the cleaned data are merged according to the set rules to obtain the merged data; data mapping is used to ensure that the game data fields from different sources can correspond correctly. For example, if the game data comes from different virtual sensors, then these data need to be integrated to form a unified game state description. Virtual sensors are virtual sensors on flight simulators. The data of each virtual sensor forms unified flight data through data mapping. Flight data is used to represent the game state of the flight simulator.

[0130] 5) Classification: Classify the multiple merged data according to the set attributes to obtain the category of each merged data.

[0131] The classification process involves applying clustering or classification algorithms to automatically identify the categories of data, or by manual labeling. For example, the skill level of players can be divided into novice, intermediate, advanced, and so on.

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

[0133] Storage refers to saving data on a persistent storage medium for long-term storage and access at any time. According to the category of game data (merged and processed data), a relational database, file system or cloud storage is selected, game data is backed up regularly, and data encryption and access control mechanisms are implemented. Game data can be stored in a high-performance relational database to support fast query and transaction processing.

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

[0135] In an exemplary embodiment, the virtual game resources include our flight simulator and the opponent's flight simulator, and the game data include flight data generated by the our flight simulator and flight data generated by the opponent's flight simulator.

[0136] The demand data also includes a game signal, and the game signal includes a response signal fed back by the other flight simulator when the own flight simulator inputs an excitation signal to the other flight simulator.

[0137] The data mining layer mines useful information from the game data based on demand, realizes polymorphic modeling and self-update of resource status, realizes targeted utilization of massive game data, and realizes functional services for users. Resource status includes available, unavailable, etc.

[0138] The data mining layer includes a spatiotemporal anomaly detection model, which is used to monitor abnormal behaviors and abnormal events of 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.

[0139] Spatiotemporal anomaly detection aims to identify abnormal behaviors or events in both time and space dimensions. This can help discover behaviors that do not conform to normal patterns, thereby revealing potential problems or opportunities. The construction of a spatiotemporal anomaly detection model specifically includes: using historical game data to establish a baseline model of normal behavior; monitoring real-time data streams and comparing with the baseline model. Use statistical methods (such as Z-score, IQR, etc.) to quantify the degree of anomaly. Use supervised learning or unsupervised learning techniques 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 is used to refer to the continuous monitoring of data streams, while statistical methods are used to evaluate whether these data streams deviate from expected behavior.

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

[0142] The data mining layer includes a co-occurrence pattern mining unit, which is used to: detect frequent item sets from the game data using the Apriori algorithm or the FP-growth algorithm; extract meaningful association rules from the frequent item sets, and output co-occurrence pattern mining results. Meaningful association rules are those that can 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, thereby providing valuable insights for decision support, optimization of resource allocation, etc. Association rules are relationships between a set of items discovered from 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 you find that Player A and Player B often appear in the same game at the same time, and their win rate increases significantly when they play together, this is a co-occurrence pattern.

[0145] Association model mining aims to discover dependencies between variables, which can help predict future behaviors or outcomes.

[0146] The data mining layer includes an association model mining unit, which is used to input the demand data into an association model to output a predicted behavior, wherein the association model is a prediction model constructed based on regression analysis, causal inference or a machine learning model, and the machine learning model is obtained by training 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 the player's historical behavior data, a model can be established to predict the actions that the player may take in a specific situation.

[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 prediction model according to different data and contexts. Self-updating refers to the ability of the spatiotemporal anomaly detection model and prediction model to automatically adjust themselves according to new data to adapt to the changing environment.

[0148] The data mining layer provides targeted utilization and functional services, with the purpose of using the information mined from the gaming data to provide targeted services, including:

[0149] Personalized recommendations: Provide personalized gaming suggestions based on player preferences and behavior patterns.

[0150] Intelligent decision support: Provide players with strategic suggestions based on data analysis.

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

[0152] User experience improvement: Improve 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 to improve their gaming experience.

[0154] This application can extract valuable information from gaming data and use this information 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 use services at different levels, 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, the utilization rate of gaming data is improved, and the usage requirements of different users are met.

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

[0157] The data mining layer is responsible for processing and analyzing raw data to extract useful information and knowledge from it. This information and knowledge is usually highly abstracted and processed data, which 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 the quality and consistency of data. 2) Data mining: using various algorithms and techniques to discover patterns, trends and associations in data. 3) Model construction: building models for prediction, classification, clustering and other purposes. 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. 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: ensuring that only authenticated users can access the corresponding services. 2) User interface design: providing a friendly user interface so that users can easily use the functions provided by the system. 3) Service customization: providing customized services based on user needs and preferences. 4) Result presentation: presenting the results generated by the data mining layer in a user-friendly form.

[0159] In this application, the data logic 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.) will be passed to the user interface layer as input, and the user interface layer will provide customized services to users based on this information. 2) Demand matching: The useful information mined by the data mining layer needs to match the needs of users. For example, if the user needs to understand the trend analysis of player behavior, the data mining layer needs to provide 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 user needs. For example, SaaS may focus on providing high-level analysis results, while PaaS may require lower-level data processing capabilities and API interfaces.

[0160] The relationship between useful information and user needs 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 specific conditions, the data mining layer can provide relevant co-occurrence patterns and association models, and the user interface layer presents this information to the user in the form of graphs or reports.

[0162] Indirect relationships: In some cases, the information generated by the data mining layer may need further processing to meet the needs of the user. 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 logic relationship between the data mining layer and the user interface layer. The data mining layer provides valuable information to users, while the user interface layer converts this information into user-operable services. The two 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) Through game resource allocation, game resources and game objects are connected, responses are obtained through input stimuli, and game signals are obtained; (2) Through virtualization technology, distributed physical game resources are abstracted into virtual game resources, and game resources and game signals are described through standard XML; (3) The game data obtained from the virtual resource pool is processed and multi-dimensional game data management is performed; (4) Different data mining methods are selected for different game needs; (5) Game data mining results are called according to different service interface requirements; (6) Services with different authority levels are provided according to user needs, and corresponding service interfaces are accessed.

[0165] The object of the game refers to the other party's flight simulator.

[0166] This application integrates the services of different levels provided by the virtual resource layer, data management layer, data mining layer and user interface layer to form a game simulation software system architecture based on SOA with game data as the core. This system architecture can effectively improve the storage and management of game data, laying the foundation for more efficient and reliable online game confrontation. In addition, due to the openness of this system architecture, it supports continuous extension and expansion in the later stage, making the game simulation software system architecture based on SOA based on SOA framework proposed in this application have good application prospects.

[0167] The SOA-based game simulation software architecture needs to provide users with a friendly human-computer interface and complete game function services, allowing users to flexibly view and manage game data, and be able to manage game resources and play online games. The SOA-based game simulation software architecture of this application is a web service structure, which is developed using the Browser / Server model. It is based on the Visual C++6.0 environment and can effectively adapt to different terminal devices. It also implements the isolation of terminal devices and game data, further ensuring data security, and facilitating future upgrades and modifications of the SOA-based game simulation software architecture and software updates and iterations. Its SOA-based game simulation software architecture program, operating environment, and database are all deployed on the network server of this type of out-of-position detection equipment. The overall implementation architecture of the SOA-based game simulation software architecture is as follows: Figure 3 shown.

[0168] In the architecture of the SOA-based game simulation software architecture, remote users access the software based on the SOA-based game simulation software architecture on the Web browser through the network and connect to the network server through the TCP / IP protocol; control of the remote operation interface and data operations is achieved through ActiveX controls and ADO controls on the network server; the database service connects the network server and the embedded controller to achieve data storage services; in the embedded controller, the game data is packaged, processed and transmitted based on the relevant signal characteristics of the game resources and the local program, and the network connection is achieved by a switch, so that users can remotely access the SOA-based game simulation software architecture through the web page.

[0169] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0170] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A game simulation software architecture based on SOA, characterized in that: The game simulation software system 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; The virtual resource layer is used to abstract the 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 the 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 the spatiotemporal anomaly detection result through the first service interface, or output the co-occurrence pattern mining result through the second service interface, or output the association model mining result 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, characterized in that: In terms of storing the gaming data generated by the virtual gaming resources, the data management layer is specifically used for: The gaming data from different sources are converted into gaming data in a set standard format to obtain the gaming data after format conversion; Convert the unstructured or semi-structured data in the game data after format conversion into structured data; Filtering the structured data according to set conditions to obtain filtered data; Cleaning the multiple filtered data, and merging the cleaned data according to the set rules to obtain merged data; Classify the 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 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 other flight simulator when the own flight simulator inputs an excitation signal to the other 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 of 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, and the co-occurrence pattern mining unit is used to: Using Apriori algorithm or FP-growth algorithm to detect frequent itemsets from the game data; Extract association rules containing the demand data from the frequent item set, 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 a 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 use 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