Game information recommendation method and device, electronic equipment and medium

Abstract

The invention provides a game information recommendation method and apparatus, an electronic device and a medium. The method comprises the steps of obtaining interaction behavior data generated by a target interaction behavior of a player in a target system in a target game; calculating a matching degree between the interactive behavior data and entity nodes of a knowledge graph in a pre-configured heterogeneous data fusion database, and determining target entity nodes of which the matching degree meets a preset matching degree condition; when the matching degree meets a preset recommendation condition corresponding to a target triggering scene of any system where the player is located currently, obtaining an information recommendation strategy and an interaction strategy corresponding to the target triggering scene; and based on the information recommendation strategy, the interaction strategy and the target entity node, generating and outputting target recommendation information which corresponds to at least one system and is matched with a target triggering scene of the system, thereby simplifying the operation complexity when a player performs information retrieval in different systems by automatically recommending the information.

Description

Technical Field

[0001] This application relates to the field of information processing technology, and more specifically, to a method, apparatus, electronic device, and medium for recommending game information. Background Technology

[0002] With the rapid development of massively multiplayer online role-playing games (MMOs), players' demands for in-game teamwork, real-time strategy information, and accurate product recommendations are increasing.

[0003] However, in current MMO games, players need to manually switch and search through a platform that is independent of the game to obtain information on joining an organization, dungeon guides, or real-time strategies. Moreover, dungeon guides and real-time strategies in MMO games usually require some items and equipment, which also require players to search in the store and perform additional operations. These fragmented data and scenarios increase the complexity of player operations. Summary of the Invention

[0004] In view of this, the purpose of this application is to provide a game information recommendation method, device, electronic device and medium that can simplify the operation complexity of players when searching for information on different systems by automatically recommending information.

[0005] In a first aspect, embodiments of this application provide a game information recommendation method, the method comprising: Acquire interactive behavior data generated from the player's target interactive behavior in the target system of the target game; Calculate the matching degree between the interactive behavior data and the entity nodes of the knowledge graph in the pre-configured heterogeneous data fusion database, and determine the target entity nodes whose matching degree meets the preset matching degree conditions; When the matching degree meets the preset recommendation conditions corresponding to the target triggering scenario of any system in which the player is currently located, the information recommendation strategy and interaction strategy corresponding to the target triggering scenario are obtained. Based on the information recommendation strategy, interaction strategy, and target entity node, generate and output target recommendation information that corresponds to at least one system and matches the target triggering scenario of that system.

[0006] Secondly, embodiments of this application provide a game information recommendation device, the device comprising: The acquisition module is used to acquire interaction behavior data generated by the player's target interaction behavior in the target system of the target game; The first determining module is used to calculate the matching degree between the interactive behavior data and the entity nodes of the knowledge graph in the pre-configured heterogeneous data fusion database, and to determine the target entity nodes whose matching degree meets the preset matching degree conditions. The matching module is used to obtain the information recommendation strategy and interaction strategy corresponding to the target triggering scenario when the matching degree meets the preset recommendation conditions corresponding to the target triggering scenario of any system in which the player is currently located. The generation module is used to generate and output target recommendation information that corresponds to at least one system and matches the target triggering scenario of that system, based on the information recommendation strategy, the interaction strategy, and the target entity node.

[0007] Thirdly, embodiments of this application provide an electronic device, including: a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the electronic device is running, the processor communicates with the memory via the bus. When the machine-readable instructions are executed by the processor, the steps of the game information recommendation method are performed.

[0008] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the steps of the game information recommendation method.

[0009] This application provides a game information recommendation method, device, electronic device, and medium. The method acquires interaction behavior data generated by a player's target interaction behavior in a target system of a target game; calculates the matching degree between the interaction behavior data and entity nodes of a knowledge graph in a pre-configured heterogeneous data fusion database, and determines target entity nodes whose matching degree meets preset matching degree conditions; when the matching degree meets preset recommendation conditions corresponding to the target triggering scenario of any system in which the player is currently located, acquires the information recommendation strategy and interaction strategy corresponding to the target triggering scenario; based on the information recommendation strategy, interaction strategy, and target entity nodes, generates and outputs target recommendation information corresponding to at least one system and matching the target triggering scenario of that system, thereby triggering information recommendations related to that scenario in different scenarios of different systems based on the player's target interaction behavior in any scenario of any system. On the one hand, this eliminates the need for players to perform similar information searches in different scenarios, and on the other hand, it can more accurately determine the recommended information for players in different game scenarios based on the player's interaction behavior, improving the spatiotemporal continuity experience when players switch between different game scenarios. Attached Figure Description

[0010] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0011] Figure 1 A flowchart of the game information recommendation method described in an embodiment of this application is shown; Figure 2 A flowchart of the method for determining target entity nodes whose matching degree meets preset matching degree conditions, as described in an embodiment of this application, is shown. Figure 3 A flowchart of another game information recommendation method according to an embodiment of this application is shown; Figure 4 A flowchart illustrating the method for generating and outputting target recommendation information according to an embodiment of this application is shown; Figure 5 A schematic diagram of the structure of the game information recommendation device according to an embodiment of this application is shown; Figure 6 A schematic diagram of the structure of the electronic device described in an embodiment of this application is shown. Detailed Implementation

[0012] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. It should be understood that the accompanying drawings in this application are for illustrative and descriptive purposes only and are not intended to limit the scope of protection of this application. Furthermore, it should be understood that the schematic drawings are not drawn to scale. The flowcharts used in this application illustrate operations implemented according to some embodiments of this application. It should be understood that the operations in the flowcharts may not be implemented in sequence, and steps without logical contextual relationships may be reversed or implemented simultaneously. In addition, those skilled in the art, guided by the content of this application, may add one or more other operations to the flowcharts, or remove one or more operations from the flowcharts.

[0013] Furthermore, the described embodiments are merely some, not all, of the embodiments of this application. The components of the embodiments of this application described and illustrated herein can typically be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0014] It should be noted that the term "comprising" will be used in the embodiments of this application to indicate the presence of the features declared thereafter, but does not exclude the addition of other features.

[0015] With the rapid development of massively multiplayer online role-playing games (MMOs), players' demands for in-game teamwork, real-time strategy information, and accurate product recommendations are increasing.

[0016] However, in current MMO games, players need to manually switch and search through a platform that is independent of the game to obtain information on joining an organization, dungeon guides, or real-time strategies. Moreover, dungeon guides and real-time strategies in MMO games usually require some items and equipment, which also require players to search in the store and perform additional operations. These fragmented data and scenarios increase the complexity of player operations.

[0017] Based on this, this application provides a game information recommendation method, device, electronic device, and medium. The method involves acquiring interaction behavior data generated by a player's target interaction behavior in a target system within a target game; calculating the matching degree between the interaction behavior data and entity nodes in a pre-configured heterogeneous data fusion database's knowledge graph; determining target entity nodes whose matching degree meets preset matching degree conditions; when the matching degree meets preset recommendation conditions corresponding to the target triggering scenario of any system the player is currently in; acquiring information recommendation strategies and interaction strategies corresponding to the target triggering scenario; and generating and outputting target recommendation information corresponding to at least one system and matching the target triggering scenario of that system, based on the player's target interaction behavior in any scenario of any system, triggering scenario-related information recommendations in different scenarios of different systems. This eliminates the need for players to perform similar information searches in different scenarios and allows for more accurate determination of recommended information for players in different game scenarios based on their interaction behavior, improving the spatiotemporal continuity experience when players switch between different game scenarios.

[0018] Please refer to Figure 1 , Figure 1 A flowchart of the game information recommendation method according to an embodiment of this application is shown; the game information recommendation method includes the following steps S101-S104: S101. Obtain interaction behavior data generated by the player's target interaction behavior in the target system of the target game; S102. Calculate the matching degree between the interactive behavior data and the entity nodes of the knowledge graph in the pre-configured heterogeneous data fusion database, and determine the target entity nodes whose matching degree meets the preset matching degree conditions. S103. When the matching degree meets the preset recommendation conditions corresponding to the target triggering scenario of any system in which the player is currently located, obtain the information recommendation strategy and interaction strategy corresponding to the target triggering scenario. S104. Based on the information recommendation strategy, interaction strategy, and target entity node, generate and output target recommendation information that corresponds to at least one system and matches the target triggering scenario of that system.

[0019] In some embodiments, the target game includes: a community system, a shop system, and a battle system.

[0020] Each system includes at least one trigger scenario, which is determined based on the system interface the user is on and their interaction state.

[0021] For example, the community system can be triggered in the following scenarios: forums / discussion areas, when players post or reply to posts in the community forum, the system will trigger recommendations of related topics, guides, and products; player activity, when players update their personal activity, the system will trigger recommendations of related guides or products.

[0022] The e-commerce system can be triggered in the following scenarios: browsing product pages, where recommending related or similar products when a player browses a particular product; searching product pages, where recommending related products when a player searches for a specific product; and checking out pages, where recommending other products that the player might need when checking out.

[0023] The combat system is triggered in the following scenarios: the game lobby, where the system will recommend strategies or items when players enter; the quest interface, where the system will recommend relevant items or strategies to help players complete quests more effectively when players view them; and the combat interface, where the system will recommend buff items or skill enhancements when players are fighting in the game. The game scenes refer to the specific virtual environments in which players directly interact, perform game quests, explore, or fight during the game. These scenes are the main areas where the core gameplay is carried out, and players interact with game elements (such as monsters, NPCs, and items) within them.

[0024] The game lobby, quest interface, and other transitional interfaces in the game are displayed when players enter, exit, or switch game scenes, providing prompts, feedback, or guidance. The game's user community platform is an online space built around the game, providing a space for players to communicate, share, and interact. This platform can be an official game platform or a game-related group or forum on a third-party social platform. Players can discuss and search for gameplay, strategies, and tips on the user community platform.

[0025] The target interactive behavior refers to the user's interactive behavior in different systems, including information gathering intentions, such as searching for products in an online store, posting a help request in a community, inputting voice or text into an intelligent AI in a game, and so on.

[0026] For example, the interaction behavior data includes: AI dialogue data of interacting with the AI ​​assistant of the target system; Combat operation data, map location data, and item operation data in the combat system; Information posting data, interaction data, and browsing behavior data within the community system; Browsing data, search data, and consumption history data of the e-commerce system.

[0027] Based on this, in some embodiments, before acquiring the interaction behavior data generated by the player's target interaction behavior in the target system of the target game, the method further includes: In response to receiving a launch operation for the AI ​​assistant of the target system, the AI ​​assistant of the target system is activated.

[0028] Activate the AI ​​assistant, which means turning on the AI ​​companion mode for the target game.

[0029] In some embodiments, the AI ​​assistant is a unified service across systems; users can activate a unified AI assistant function with a single operation (such as pressing a shortcut key in the game or launching a standalone application on the desktop). The AI ​​assistant listens to the user's interactive behavior in multiple systems (games, communities, and stores) and provides intelligent recommendations and interactive gameplay.

[0030] Here, for the activation operation of the AI ​​assistant of any system, the AI ​​assistant of the target game in at least one system related to the target game is activated. That is to say, the AI ​​assistant in this system can be activated only, or the AI ​​assistants in multiple systems can be activated.

[0031] Specifically, in some embodiments, users can disable the AI ​​assistant in a system, and the AI ​​assistant will not be enabled in that system; for example, disabling the AI ​​assistant in a game to prevent it from interfering with the game screen.

[0032] When the AI ​​companion mode is turned on, the AI ​​assistant control is displayed on different trigger interfaces of the game. Users can enter information into the AI ​​assistant control to search, such as: how to obtain item A.

[0033] Here, the information input into the AI ​​assistant can be text or voice.

[0034] In step S102, the matching degree between the interactive behavior data and the entity nodes of the knowledge graph in the pre-configured heterogeneous data fusion database is calculated, and the target entity nodes whose matching degree meets the preset matching degree conditions are determined.

[0035] The matching degree is also related to the time decay factor and spatial correlation; the time decay factor represents the real-time nature of the interaction behavior data, and the spatial correlation represents the correlation between the interaction behavior data and the current game scene in which the player's game character is located.

[0036] In some embodiments, please refer to Figure 2 The step of calculating the matching degree between the interaction behavior data and the entity nodes of the knowledge graph in the pre-configured heterogeneous data fusion database, and determining the target entity nodes whose matching degree meets the preset matching degree conditions, includes the following steps S201-S204: S201. Calculate the semantic similarity between the interactive behavior data and the entity nodes of the knowledge graph in the pre-configured heterogeneous data fusion database; S202. Based on the semantic similarity between the interaction behavior data and the entity node, the time decay factor corresponding to the interaction behavior data, and the spatial correlation, calculate the matching degree between the interaction behavior data and the entity node; the time decay factor is used to characterize the real-time nature of the interaction behavior data; the spatial correlation is used to characterize the correlation between the interaction behavior data and the current game scene where the game character controlled by the player is located.

[0037] In some embodiments, the time decay factor is determined based on the timestamp in the interaction behavior data, the current real-time time difference, and a preset decay time constant, wherein the time decay factor of the interaction behavior data that is closer to the current real-time time is larger. The spatial correlation is determined based on the distance between the coordinates of the player's game character in the current game scene and the coordinates corresponding to the interactive behavior data; the closer the distance, the higher the spatial correlation.

[0038] Based on this, in some embodiments, calculating the matching degree between the interaction behavior data and the entity node based on the semantic similarity between the interaction behavior data and the entity node, the time decay factor corresponding to the interaction behavior data, and the spatial correlation degree includes: The time decay factor is determined based on the timestamp in the interaction behavior data, the current real-time time difference, and the preset decay time constant; wherein, the time decay factor of the interaction behavior data that is closer to the current real-time time is larger. The spatial correlation is determined based on the distance between the player's game character's coordinates in the current game scene and the coordinates corresponding to the interactive behavior data. The closer the distance, the higher the spatial correlation. Based on the semantic similarity between the interaction behavior data and the entity node, the time decay factor and spatial correlation degree corresponding to the interaction behavior data, the matching degree between the interaction behavior data and the entity node is calculated.

[0039] In other words, based on the occurrence time of the target interaction behavior and the current real-time time, a time decay factor is determined; based on the spatial correlation between the target interaction behavior and the current game scene where the player's game character is located, semantic similarity is fused to obtain the real-time matching degree corresponding to the target interaction behavior; thus, the degree of influence of target interaction behaviors of different ages on information is characterized.

[0040] Here, information recommendation is made across multiple systems based on the target interaction behavior in one system. In other words, when a user switches to other different systems, information recommendation can still be made based on the target interaction behavior, thereby achieving cross-platform continuous service based on the target interaction behavior.

[0041] For example, if a player asks the AI ​​in the game "How to obtain item A", the target recommendation information will be "Recommended dungeons; Dungeon A"; based on the target retrieval information of "How to obtain item A", the target recommendation information on the transition screen when the player exits the game will be "Team recommendations"; when the player visits the community, the target recommendation information on the community interface will be "Personalized item combinations" or strategy posts for obtaining "item A".

[0042] Specifically, the time decay factor has a higher weight for behaviors closer to the current time; recent behaviors (such as AI dialogue in battle and clicks in the store) have a higher weight to avoid historical data interfering with real-time intent judgment; for example, when a player in "Game A" asks about "epic mounts" in battle, the system prioritizes associating data from the current battle stage (such as dungeon progress) rather than community search records from three days ago.

[0043] The formula for calculating the time decay factor is as follows: time_decay = exp(-(current_time -event_time) / τ).

[0044] Wherein, time_decay represents the time decay factor, current_time represents the current real-time, event_time represents the timestamp in the interaction behavior data, and τ represents the preset decay time constant.

[0045] The spatial relevance is calculated based on the game map coordinates to determine the event density. Its function is to dynamically adjust the spatial relevance of recommended content according to the game scene where the player is located (such as the "CC Mountains" area in "B Game").

[0046] For example, if a player is near the CC Mountains and their intention is more to obtain CC Mountains' specialty CCC products than DD Port's specialty DDD products, then "CCC products" should be recommended instead of "DDD products".

[0047] The formula for calculating the spatial correlation degree is as follows: spatial_weight = gaussian_kernel(event_coords, community_location); Among them, spatial_weight represents spatial correlation, community_location represents the coordinates of the player's game character in the current game scene, and event_coords represents the coordinates corresponding to the interaction behavior data.

[0048] The coordinates corresponding to the interactive behavior data include the coordinates of the player's game character in the game scene when the target interactive behavior occurs, and the location where the game event corresponding to the interactive behavior data occurs.

[0049] The composite intent value = semantic similarity * time decay factor * spatial relevance, and recommendations are triggered only when the composite intent value exceeds a dynamic threshold.

[0050] Please refer to the following formula to calculate the composite intent score: intent_score = NLP_similarity * time_decay * spatial_weight; Wherein, intent_score represents the composite intent value; NLP_similarity represents the semantic similarity corresponding to the interaction behavior data; time_decay represents the time decay factor; and spatial_weight represents the spatial correlation.

[0051] The pre-configured heterogeneous data fusion database is a database built by integrating data from the community system, the store system, and the battle system. It achieves cross-platform data association through federated learning and knowledge graphs, solving the data silo problem of games, communities, and stores.

[0052] The heterogeneous data fusion database updates its knowledge graph through a multimodal pipeline data stream. This multimodal pipeline data stream includes game data streams and community data streams.

[0053] The pre-configured heterogeneous data fusion database is constructed based on the following method: A real-time feature library is built based on game data streams from the combat system; A hot topic trend database is built based on community data streams from the community system; A knowledge graph is constructed based on real-time features in the real-time feature library, community topic features in the hot topic trend library, and the relationship between real-time features, community topic features, and e-commerce SKUs. The knowledge graph includes triples composed of entities from a single system and triples composed of intersections with entities from different systems.

[0054] The game data stream based on the combat system is used to construct a real-time feature library: Obtain the player's combat log in the target game; The combat logs are processed by a behavior encoder to generate real-time features that characterize player behavior. Update the real-time feature library based on the real-time features.

[0055] Features are extracted from battle logs, such as basic features like action type, target type, and location coordinates, as well as derived features like movement speed and skill release frequency. The extracted features are then serialized and vectorized. Specifically, discrete actions are converted into numerical sequences that can be processed by LSTM, enabling real-time feature extraction and storage. This achieves real-time conversion of game data streams from raw logs to high-dimensional behavioral features, providing an operable player profile for cross-platform recommendation systems.

[0056] The LSTM network design and training are as follows: The input layer receives serialized action features; the LSTM layer captures temporal dependencies; feature aggregation: taking the final state or pooling.

[0057] In some alternative embodiments, the weight of key actions can also be enhanced through attention mechanisms.

[0058] For example, a player's combo frequency is mapped to an "operation proficiency" label to determine their willingness to pay.

[0059] The community data stream based on the community system is used to construct a hot topic trend database, including: Obtain player community semantic information within the community system; Process the community semantic information to determine the player's sentiment tags for the target topic; Based on the players' sentiment tags for the target topic, community topic features are generated and the real-time feature library is updated.

[0060] The hot topic trend library focuses on the popularity of topics in the community and identifies frequently discussed topics (such as "artifact drop rate" and "new version bugs").

[0061] The target topic is obtained by processing community post text through a hot topic tag extraction model. Specifically, the input of the hot topic tag extraction model is community post text, keyword frequency, and user interaction data (such as the number of likes and comments), and the output is hot topic tags that are updated in real time (such as {"topic": "Artifact drop rate", "heat_score": 0.92}).

[0062] The sentiment tags assigned to players for the target topic are determined by the sentiment analysis model (VADER). The sentiment tags analyze the users' sentiment tendencies (positive, neutral, negative) towards the target topic in community posts. The sentiment analysis model analyzes the original text related to the trending topic and inputs a sentiment polarity score (e.g., {"sentiment": -0.85} represents strong negative emotion). The outputs of both are fused through multimodal analysis to form a structured signal that serves the recommendation system, enabling the recommendation system to determine "what solutions the user needs".

[0063] In some embodiments, constructing a knowledge graph based on real-time features in a real-time feature library, community topic features in a hot topic trend library, and the relationship between real-time features, community topic features, and e-commerce SKUs includes: Extract entity relationships from real-time features in the real-time feature library and construct triples corresponding to the combat system; Extract the entity relationships of community topic features from the hot topic trend database and construct the corresponding triples for the community system; The relationships between the entities in the combat system, the community system, and the mall SKUs are explored, and entities from different systems are merged to obtain triplets formed by the intersection of entities from different systems. A knowledge graph is constructed based on the triples corresponding to the combat system, the triples corresponding to the community system, and the triples formed by the intersection of entities in different systems.

[0064] The real-time features in the real-time feature library are triples, and the community topic features in the hot topic trend library are triples; association rules for real-time features, community topic features, and e-commerce SKUs are constructed based on the triple fusion method.

[0065] In some embodiments, entity relationship mining involves extracting “player-dungeon-equipment” triples from combat logs and merging them with “player-problem-solution” triples from community posts.

[0066] For example, player A asks in the community "How to defeat Stormterror" → the knowledge graph links to their in-game record of "missing wind resistance equipment" → recommends resistance potions from the shop.

[0067] In step S103, when the matching degree meets the preset recommendation conditions corresponding to the target triggering scenario of any system in which the player is currently located, the information recommendation strategy and interaction strategy corresponding to the target triggering scenario are obtained.

[0068] Specifically, when the matching degree meets the preset recommendation conditions corresponding to the target triggering scenario of any system in which the player is currently located, that is, when the composite intent value is greater than the preset threshold of the target triggering scenario of the system.

[0069] The preset thresholds for different triggering scenarios can be the same or different.

[0070] This is because users' tolerance for disturbances and their level of concern for user intentions vary greatly in different scenarios.

[0071] In highly immersive scenarios, such as intense battles in games, the threshold θ is set very high (e.g., θ=0.85). It is only triggered when the intent is very clear and urgent (e.g., the player is about to die and is explicitly asking for help), so as not to interrupt the player's flow. In low-immersion scenarios, such as browsing community forums, the threshold θ is set relatively low (e.g., θ=0.5), so even if the intent signal is slightly weak (e.g., the user may just be browsing casually), a recommendation can still be triggered to test the waters.

[0072] The information recommendation strategy corresponding to the target trigger scenario is used to determine the content to recommend to the user under the target trigger scenario; for example, recommending products in an online store, or recommending strategy posts in a community.

[0073] The interaction strategy is used to determine the display method in the interface of the target triggering scenario; the display method includes bubbles, sidebars, etc.

[0074] Please refer to Figure 3 In some embodiments, before obtaining the information recommendation strategy and interaction strategy corresponding to the target triggering scenario, the method further includes the following steps S301-S302: S301. Based on the player's immersion in different types of target triggering scenarios in each system, determine the exposure level of target recommendation information for each target triggering scenario in each system; S302. Based on the exposure level and the interface attributes of the target triggering scene in the system, determine the interaction strategy for the triggering scene in the system.

[0075] In other words, based on the player's immersion in different types of triggering scenarios, a progressive disclosure strategy is designed, and the intensity of recommended intervention is dynamically adjusted according to the player's flow state.

[0076] For example, the player's immersion level for different types of triggering scenarios includes level one, level two, and level three. The immersion level of the game scene, the game's transition interface, and the game's user community platform decreases sequentially, while the exposure level increases sequentially.

[0077] For example, the immersion level of the game scene (e.g., in-game AI dialogue) in the game is level one, and the exposure level of the interaction design is low. Specifically, a non-interrupted interaction design is adopted, such as using bubble prompts.

[0078] The game's transition interfaces, such as the game's homepage or the transition interface when exiting the game, have a medium level of exposure in their interaction design. For example, leaderboards / team recommendations occupy a portion of the graphical user interface for display.

[0079] The game community has an advanced level of exposure for its interactive design. For example, the display format for personalized item combinations is high exposure + CTA button.

[0080] For example, Level 1 trigger: When a player asks the AI ​​in LY Port "How to obtain QWR items" → the bubble displays "Recommended dungeon: HQ Palace (requires SZ materials)"; Level 2 trigger: When a player exits the game → the transition interface displays "Currently online farming teams (including players with double experience cards)"; Level 3 trigger: When a player visits the community → the sidebar recommends "SZ material pack + QWR item enhancement material combination".

[0081] In step 104, based on the information recommendation strategy, the interaction strategy, and the target entity node, target recommendation information corresponding to at least one system and matching the target triggering scenario of that system is generated and output.

[0082] In some embodiments, please refer to Figure 4 , Figure 4 The flowchart of the method for generating and outputting target recommendation information according to an embodiment of this application is shown; specifically, the step of generating and outputting target recommendation information that corresponds to at least one system and matches the target triggering scenario of that system based on the information recommendation strategy, interaction strategy, and target entity node includes the following steps S401-S402: S401. Determine initial recommendation information based on the target entity node, the heterogeneous data fusion database, and the information recommendation strategy; S402. Based on the interaction strategy, render and adapt the initial recommendation information to generate target recommendation information that matches the interface specifications of the target triggering scenario, and output it to the interface of the target triggering scenario.

[0083] In step S401, a heterogeneous data fusion database is used to determine target recommendation information for different systems (combat system, store system, and community system) that matches the target interaction behavior. For example, in a role-playing game, the target interaction behavior determines that the player needs to find a powerful weapon to defeat the BOSS. The heterogeneous data fusion database contains the attributes of various weapons in the store, strategy guides for obtaining the weapon, and information on teams that defeated the BOSS in the game. The information recommendation strategy filters out suitable weapons, strategy guides, and team information from the database based on the system the player is currently logged into.

[0084] After determining the target recommendation information, the display rules defined by the interaction strategy of the target triggering scenario are applied to process the target recommendation information into an encapsulated format that meets the interface requirements of the target triggering scenario in the system, and then push it out. Based on the same inventive concept, this application also provides a game information recommendation device corresponding to the game information recommendation method. Since the principle of the device in this application is similar to the game information recommendation method described above, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be described again.

[0085] Please refer to Figure 5 , Figure 5 A schematic diagram of the structure of the game information recommendation device according to an embodiment of this application is shown; as follows: Figure 5 As shown, the device includes: Module 501 is used to acquire interaction behavior data generated by the player's target interaction behavior in the target system of the target game. The first determining module 502 is used to calculate the matching degree between the interactive behavior data and the entity nodes of the knowledge graph in the pre-configured heterogeneous data fusion database, and to determine the target entity nodes whose matching degree meets the preset matching degree conditions. The matching module 503 is used to obtain the information recommendation strategy and interaction strategy corresponding to the target triggering scenario when the matching degree meets the preset recommendation conditions corresponding to the target triggering scenario of any system in which the player is currently located. The generation module 504 is used to generate and output target recommendation information that corresponds to at least one system and matches the target triggering scenario of that system, based on the information recommendation strategy, the interaction strategy, and the target entity node.

[0086] In some embodiments, the target game in the game information recommendation device includes: a community system, a store system, and a battle system.

[0087] In some embodiments, in the game information recommendation device, the first determining module, when calculating the matching degree between the interactive behavior data and the entity nodes of the knowledge graph in the pre-configured heterogeneous data fusion database, and determining the target entity node whose matching degree meets the preset matching degree condition, is specifically used for: Calculate the semantic similarity between the interactive behavior data and the entity nodes of the knowledge graph in the pre-configured heterogeneous data fusion database; Based on the semantic similarity between the interactive behavior data and the entity node, the time decay factor corresponding to the interactive behavior data, and the spatial correlation, the matching degree between the interactive behavior data and the entity node is calculated; the time decay factor is used to characterize the real-time nature of the interactive behavior data; the spatial correlation is used to characterize the correlation between the interactive behavior data and the current game scene where the player-controlled game character is located.

[0088] In some embodiments, in the game information recommendation device, the first determining module, when calculating the matching degree between the interaction behavior data and the entity node based on the semantic similarity between the interaction behavior data and the entity node, the time decay factor corresponding to the interaction behavior data, and the spatial correlation, is specifically used for: The time decay factor is determined based on the timestamp in the interaction behavior data, the current real-time time difference, and the preset decay time constant; wherein, the time decay factor of the interaction behavior data that is closer to the current real-time time is larger. The spatial correlation is determined based on the distance between the player's game character's coordinates in the current game scene and the coordinates corresponding to the interactive behavior data. The closer the distance, the higher the spatial correlation. Based on the semantic similarity between the interaction behavior data and the entity node, the time decay factor and spatial correlation degree corresponding to the interaction behavior data, the matching degree between the interaction behavior data and the entity node is calculated.

[0089] In some embodiments, the game information recommendation device further includes: Build modules are used to construct a real-time feature library based on the game data stream of the combat system; A hot topic trend database is built based on community data streams from the community system; A knowledge graph is constructed based on real-time features in the real-time feature library, community topic features in the hot topic trend library, and the relationship between real-time features, community topic features, and e-commerce SKUs. The knowledge graph includes triples composed of entities from a single system and triples composed of intersections with entities from different systems.

[0090] In some embodiments, the construction module in the game information recommendation device is specifically used to: When constructing a real-time feature library based on the game data stream of the combat system, Obtain the player's combat log in the target game; The combat logs are processed by a behavior encoder to generate real-time features that characterize player behavior. Update the real-time feature library based on the real-time features.

[0091] In some embodiments, the construction module in the game information recommendation device is specifically used to: When constructing a hot topic trend database based on community data streams from a community system, Obtain player community semantic information within the community system; Process the community semantic information to determine the player's sentiment tags for the target topic; Based on the players' sentiment tags for the target topic, community topic features are generated, and the hot topic trend database is updated.

[0092] In some embodiments, the construction module in the game information recommendation device, when constructing a knowledge graph based on real-time features in the real-time feature library, community topic features in the hot topic trend library, and the relationship between real-time features, community topic features, and store SKUs, is specifically used for: Extract entity relationships from real-time features in the real-time feature library and construct triples corresponding to the combat system; Extract the entity relationships of community topic features from the hot topic trend database and construct the corresponding triples for the community system; The relationships between the entities in the combat system, the community system, and the mall SKUs are explored, and entities from different systems are merged to obtain triplets formed by the intersection of entities from different systems. A knowledge graph is constructed based on the triples corresponding to the combat system, the triples corresponding to the community system, and the triples formed by the intersection of entities in different systems.

[0093] In some embodiments, the game information recommendation device further includes: The second determining module is used to determine the exposure level of target recommendation information for each target triggering scenario in each system based on the player's immersion in different types of target triggering scenarios in each system before obtaining the information recommendation strategy and interaction strategy corresponding to the target triggering scenario. Based on the exposure level and the interface attributes of the target triggering scene in the system, an interaction strategy is determined for the triggering scene in the system.

[0094] In some embodiments, the target triggering scenario in the game information recommendation device includes game scenes in the game, game transition interfaces, game user community platforms, and game store interfaces. The exposure levels of the game scenes, game transition interfaces, game user community platform, and game store interface increase sequentially.

[0095] In some embodiments, in the game information recommendation device, the interactive behavior data includes: AI dialogue data of interacting with the AI ​​assistant of the target system; Combat operation data, map location data, and item operation data in the combat system; Information posting data, interaction data, and browsing behavior data within the community system; Browsing data, search data, and consumption history data of the e-commerce system.

[0096] In some embodiments, the game information recommendation device further includes: The activation module is used to activate the AI ​​assistant of the target system in response to receiving a activation operation for the AI ​​assistant of the target system before acquiring the interaction behavior data generated by the player's target interaction behavior in the target system of the target game.

[0097] In some embodiments, the generation module in the game information recommendation device, when generating and outputting target recommendation information that corresponds to at least one system and matches the target triggering scenario of that system based on the information recommendation strategy, interaction strategy, and target entity node, is specifically used for: Based on the target entity node, the heterogeneous data fusion database, and the information recommendation strategy, initial recommendation information is determined; Based on the interaction strategy, the initial recommendation information is rendered and adapted to generate target recommendation information that matches the interface specifications of the target triggering scenario, and then output to the interface of the target triggering scenario.

[0098] Based on the same inventive concept, this application also provides an electronic device corresponding to the game information recommendation method. Since the principle of the electronic device in this application is similar to the game information recommendation method described above, the implementation of the electronic device can refer to the implementation of the method, and the repeated parts will not be described again.

[0099] Please refer to Figure 6 , Figure 6 A schematic diagram of the structure of the electronic device described in an embodiment of this application is shown; as follows: Figure 6As shown, the electronic device 600 includes a processor 602, a memory 601, and a bus. The memory 601 stores machine-readable instructions executable by the processor 602. When the electronic device 600 is running, the processor 602 communicates with the memory 601 via the bus. When the machine-readable instructions are executed by the processor 602, the processor performs the steps of the game information recommendation method as follows: Acquire interactive behavior data generated from the player's target interactive behavior in the target system of the target game; Calculate the matching degree between the interactive behavior data and the entity nodes of the knowledge graph in the pre-configured heterogeneous data fusion database, and determine the target entity nodes whose matching degree meets the preset matching degree conditions; When the matching degree meets the preset recommendation conditions corresponding to the target triggering scenario of any system in which the player is currently located, the information recommendation strategy and interaction strategy corresponding to the target triggering scenario are obtained. Based on the information recommendation strategy, interaction strategy, and target entity node, generate and output target recommendation information that corresponds to at least one system and matches the target triggering scenario of that system.

[0100] In some embodiments, the target game includes: a community system, a shop system, and a battle system.

[0101] In some embodiments, when the processor performs the step of calculating the matching degree between the interaction behavior data and the entity nodes of the knowledge graph in the pre-configured heterogeneous data fusion database, and determining the target entity node whose matching degree meets the preset matching degree conditions, it specifically performs the following steps: Calculate the semantic similarity between the interactive behavior data and the entity nodes of the knowledge graph in the pre-configured heterogeneous data fusion database; Based on the semantic similarity between the interactive behavior data and the entity node, the time decay factor corresponding to the interactive behavior data, and the spatial correlation, the matching degree between the interactive behavior data and the entity node is calculated; the time decay factor is used to characterize the real-time nature of the interactive behavior data; the spatial correlation is used to characterize the correlation between the interactive behavior data and the current game scene where the player-controlled game character is located.

[0102] In some embodiments, when the processor performs the step of calculating the matching degree between the interaction behavior data and the entity node based on the semantic similarity between the interaction behavior data and the entity node, the time decay factor corresponding to the interaction behavior data, and the spatial correlation, it specifically performs the following steps: The time decay factor is determined based on the timestamp in the interaction behavior data, the current real-time time difference, and the preset decay time constant; wherein, the time decay factor of the interaction behavior data that is closer to the current real-time time is larger. The spatial correlation is determined based on the distance between the player's game character's coordinates in the current game scene and the coordinates corresponding to the interactive behavior data. The closer the distance, the higher the spatial correlation. Based on the semantic similarity between the interaction behavior data and the entity node, the time decay factor and spatial correlation degree corresponding to the interaction behavior data, the matching degree between the interaction behavior data and the entity node is calculated.

[0103] In some embodiments, the processor further performs the following steps: A real-time feature library is built based on game data streams from the combat system; A hot topic trend database is built based on community data streams from the community system; A knowledge graph is constructed based on real-time features in the real-time feature library, community topic features in the hot topic trend library, and the relationship between real-time features, community topic features, and e-commerce SKUs. The knowledge graph includes triples composed of entities from a single system and triples composed of intersections with entities from different systems.

[0104] In some embodiments, when the processor performs the step of constructing a real-time feature library based on the game data stream of the combat system, it specifically performs the following steps: Obtain the player's combat log in the target game; The combat logs are processed by a behavior encoder to generate real-time features that characterize player behavior. Update the real-time feature library based on the real-time features.

[0105] In some embodiments, when the processor executes the step of constructing a hotspot trend database based on community data streams from the community system, it specifically performs the following steps: Obtain player community semantic information within the community system; Process the community semantic information to determine the player's sentiment tags for the target topic; Based on the players' sentiment tags for the target topic, community topic features are generated, and the hot topic trend database is updated.

[0106] In some embodiments, when the processor performs the step of constructing a knowledge graph based on real-time features in a real-time feature library, community topic features in a hot topic trend library, and the association between real-time features, community topic features, and e-commerce SKUs, the specific steps are as follows: Extract entity relationships from real-time features in the real-time feature library and construct triples corresponding to the combat system; Extract the entity relationships of community topic features from the hot topic trend database and construct the corresponding triples for the community system; The relationships between the entities in the combat system, the community system, and the mall SKUs are explored, and entities from different systems are merged to obtain triplets formed by the intersection of entities from different systems. A knowledge graph is constructed based on the triples corresponding to the combat system, the triples corresponding to the community system, and the triples formed by the intersection of entities in different systems.

[0107] In some embodiments, before executing the step of obtaining the information recommendation strategy and interaction strategy corresponding to the target triggering scenario, the processor further performs the following steps: Based on the player's immersion in different types of target-triggered scenarios in each system, determine the exposure level of target recommendation information for each target-triggered scenario in each system; Based on the exposure level and the interface attributes of the target triggering scene in the system, an interaction strategy is determined for the triggering scene in the system.

[0108] In some embodiments, the target triggering scenario includes game scenes in the game, game transition interfaces, game user community platforms, and game store interfaces; The exposure levels of the game scenes, game transition interfaces, game user community platform, and game store interface increase sequentially.

[0109] In some embodiments, the interaction behavior data includes: AI dialogue data of interacting with the AI ​​assistant of the target system; Combat operation data, map location data, and item operation data in the combat system; Information posting data, interaction data, and browsing behavior data within the community system; Browsing data, search data, and consumption history data of the e-commerce system.

[0110] In some embodiments, before performing the step of acquiring interaction behavior data generated from the player's target interaction behavior in the target system of the target game, the processor further performs the following steps: In response to receiving a launch operation for the AI ​​assistant of the target system, the AI ​​assistant of the target system is activated.

[0111] In some embodiments, when the processor executes the step of generating and outputting target recommendation information that corresponds to at least one system and matches the target triggering scenario of that system based on the information recommendation strategy, interaction strategy, and target entity node, it specifically performs the following steps: Based on the target entity node, the heterogeneous data fusion database, and the information recommendation strategy, initial recommendation information is determined; Based on the interaction strategy, the initial recommendation information is rendered and adapted to generate target recommendation information that matches the interface specifications of the target triggering scenario, and then output to the interface of the target triggering scenario.

[0112] Based on the same inventive concept, this application also provides a computer-readable storage medium corresponding to the game information recommendation method. Since the principle of the computer-readable storage medium in this application is similar to the game information recommendation method described above, the implementation of the computer-readable storage medium can refer to the implementation of the method, and the repeated parts will not be described again.

[0113] A computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the game information recommendation method, as follows: Acquire interactive behavior data generated from the player's target interactive behavior in the target system of the target game; Calculate the matching degree between the interactive behavior data and the entity nodes of the knowledge graph in the pre-configured heterogeneous data fusion database, and determine the target entity nodes whose matching degree meets the preset matching degree conditions; When the matching degree meets the preset recommendation conditions corresponding to the target triggering scenario of any system in which the player is currently located, the information recommendation strategy and interaction strategy corresponding to the target triggering scenario are obtained. Based on the information recommendation strategy, interaction strategy, and target entity node, generate and output target recommendation information that corresponds to at least one system and matches the target triggering scenario of that system.

[0114] In some embodiments, the target game includes: a community system, a shop system, and a battle system.

[0115] In some embodiments, when the processor performs the step of calculating the matching degree between the interaction behavior data and the entity nodes of the knowledge graph in the pre-configured heterogeneous data fusion database, and determining the target entity node whose matching degree meets the preset matching degree conditions, it specifically performs the following steps: Calculate the semantic similarity between the interactive behavior data and the entity nodes of the knowledge graph in the pre-configured heterogeneous data fusion database; Based on the semantic similarity between the interactive behavior data and the entity node, the time decay factor corresponding to the interactive behavior data, and the spatial correlation, the matching degree between the interactive behavior data and the entity node is calculated; the time decay factor is used to characterize the real-time nature of the interactive behavior data; the spatial correlation is used to characterize the correlation between the interactive behavior data and the current game scene where the player-controlled game character is located.

[0116] In some embodiments, when the processor performs the step of calculating the matching degree between the interaction behavior data and the entity node based on the semantic similarity between the interaction behavior data and the entity node, the time decay factor corresponding to the interaction behavior data, and the spatial correlation, it specifically performs the following steps: The time decay factor is determined based on the timestamp in the interaction behavior data, the current real-time time difference, and the preset decay time constant; wherein, the time decay factor of the interaction behavior data that is closer to the current real-time time is larger. The spatial correlation is determined based on the distance between the player's game character's coordinates in the current game scene and the coordinates corresponding to the interactive behavior data. The closer the distance, the higher the spatial correlation. Based on the semantic similarity between the interaction behavior data and the entity node, the time decay factor and spatial correlation degree corresponding to the interaction behavior data, the matching degree between the interaction behavior data and the entity node is calculated.

[0117] In some embodiments, the processor further performs the following steps: A real-time feature library is built based on game data streams from the combat system; A hot topic trend database is built based on community data streams from the community system; A knowledge graph is constructed based on real-time features in the real-time feature library, community topic features in the hot topic trend library, and the relationship between real-time features, community topic features, and e-commerce SKUs. The knowledge graph includes triples composed of entities from a single system and triples composed of intersections with entities from different systems.

[0118] In some embodiments, when the processor performs the step of constructing a real-time feature library based on the game data stream of the combat system, it specifically performs the following steps: Obtain the player's combat log in the target game; The combat logs are processed by a behavior encoder to generate real-time features that characterize player behavior. Update the real-time feature library based on the real-time features.

[0119] In some embodiments, when the processor executes the step of constructing a hotspot trend database based on community data streams from the community system, it specifically performs the following steps: Obtain player community semantic information within the community system; Process the community semantic information to determine the player's sentiment tags for the target topic; Based on the players' sentiment tags for the target topic, community topic features are generated, and the hot topic trend database is updated.

[0120] In some embodiments, when the processor performs the step of constructing a knowledge graph based on real-time features in a real-time feature library, community topic features in a hot topic trend library, and the association between real-time features, community topic features, and e-commerce SKUs, the specific steps are as follows: Extract entity relationships from real-time features in the real-time feature library and construct triples corresponding to the combat system; Extract the entity relationships of community topic features from the hot topic trend database and construct the corresponding triples for the community system; The relationships between the entities in the combat system, the community system, and the mall SKUs are explored, and entities from different systems are merged to obtain triplets formed by the intersection of entities from different systems. A knowledge graph is constructed based on the triples corresponding to the combat system, the triples corresponding to the community system, and the triples formed by the intersection of entities in different systems.

[0121] In some embodiments, before executing the step of obtaining the information recommendation strategy and interaction strategy corresponding to the target triggering scenario, the processor further performs the following steps: Based on the player's immersion in different types of target-triggered scenarios in each system, determine the exposure level of target recommendation information for each target-triggered scenario in each system; Based on the exposure level and the interface attributes of the target triggering scene in the system, an interaction strategy is determined for the triggering scene in the system.

[0122] In some embodiments, the target triggering scenario includes game scenes in the game, game transition interfaces, game user community platforms, and game store interfaces; The exposure levels of the game scenes, game transition interfaces, game user community platform, and game store interface increase sequentially.

[0123] In some embodiments, the interaction behavior data includes: AI dialogue data of interacting with the AI ​​assistant of the target system; Combat operation data, map location data, and item operation data in the combat system; Information posting data, interaction data, and browsing behavior data within the community system; Browsing data, search data, and consumption history data of the e-commerce system.

[0124] In some embodiments, before performing the step of acquiring interaction behavior data generated from the player's target interaction behavior in the target system of the target game, the processor further performs the following steps: In response to receiving a launch operation for the AI ​​assistant of the target system, the AI ​​assistant of the target system is activated.

[0125] In some embodiments, when the processor executes the step of generating and outputting target recommendation information that corresponds to at least one system and matches the target triggering scenario of that system based on the information recommendation strategy, interaction strategy, and target entity node, it specifically performs the following steps: Based on the target entity node, the heterogeneous data fusion database, and the information recommendation strategy, initial recommendation information is determined; Based on the interaction strategy, the initial recommendation information is rendered and adapted to generate target recommendation information that matches the interface specifications of the target triggering scenario, and then output to the interface of the target triggering scenario.

[0126] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems and devices described above can be referred to the corresponding processes in the method embodiments, and will not be repeated here. In the several embodiments provided in this application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed mutual coupling or direct coupling or communication connection can be through some communication interfaces; the indirect coupling or communication connection of devices or modules can be electrical, mechanical, or other forms.

[0127] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0128] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0129] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a processor-executable, non-volatile, computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, a platform server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, ROM, RAM, magnetic disks, or optical disks.

[0130] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for recommending game information, characterized in that, The method includes: Acquire interactive behavior data generated from the player's target interactive behavior in the target system of the target game; Calculate the matching degree between the interactive behavior data and the entity nodes of the knowledge graph in the pre-configured heterogeneous data fusion database, and determine the target entity nodes whose matching degree meets the preset matching degree conditions; When the matching degree meets the preset recommendation conditions corresponding to the target triggering scenario of any system in which the player is currently located, the information recommendation strategy and interaction strategy corresponding to the target triggering scenario are obtained. Based on the information recommendation strategy, interaction strategy, and target entity node, generate and output target recommendation information that corresponds to at least one system and matches the target triggering scenario of that system.

2. The method according to claim 1, characterized in that, The target game includes: a community system, a shop system, and a combat system.

3. The method according to claim 1, characterized in that, The step of calculating the matching degree between the interactive behavior data and the entity nodes of the knowledge graph in the pre-configured heterogeneous data fusion database, and determining the target entity nodes whose matching degree meets the preset matching degree conditions, includes: Calculate the semantic similarity between the interactive behavior data and the entity nodes of the knowledge graph in the pre-configured heterogeneous data fusion database; Based on the semantic similarity between the interactive behavior data and the entity node, the time decay factor corresponding to the interactive behavior data, and the spatial correlation, the matching degree between the interactive behavior data and the entity node is calculated; the time decay factor is used to characterize the real-time nature of the interactive behavior data; the spatial correlation is used to characterize the correlation between the interactive behavior data and the current game scene where the player-controlled game character is located.

4. The method according to claim 3, characterized in that, The calculation of the matching degree between the interaction behavior data and the entity nodes based on the semantic similarity between the interaction behavior data and the entity nodes, the time decay factor corresponding to the interaction behavior data, and the spatial correlation degree includes: The time decay factor is determined based on the timestamp in the interaction behavior data, the current real-time time difference, and the preset decay time constant; wherein, the time decay factor of the interaction behavior data that is closer to the current real-time time is larger. The spatial correlation is determined based on the distance between the player's game character's coordinates in the current game scene and the coordinates corresponding to the interactive behavior data. The closer the distance, the higher the spatial correlation. Based on the semantic similarity between the interaction behavior data and the entity node, the time decay factor and spatial correlation degree corresponding to the interaction behavior data, the matching degree between the interaction behavior data and the entity node is calculated.

5. The method according to claim 2, characterized in that, The pre-configured heterogeneous data fusion database is constructed based on the following method: A real-time feature library is built based on game data streams from the combat system; A hot topic trend database is built based on community data streams from the community system; A knowledge graph is constructed based on real-time features in the real-time feature library, community topic features in the hot topic trend library, and the relationship between real-time features, community topic features, and e-commerce SKUs. The knowledge graph includes triples composed of entities from a single system and triples composed of intersections with entities from different systems.

6. The method according to claim 5, characterized in that, The construction of a real-time feature library based on the game data stream of the combat system includes: Obtain the player's combat log in the target game; The combat logs are processed by a behavior encoder to generate real-time features that characterize player behavior. Update the real-time feature library based on the real-time features.

7. The method according to claim 5, characterized in that, The community data stream based on the community system is used to construct a hot topic trend database, including: Obtain player community semantic information within the community system; Process the community semantic information to determine the player's sentiment tags for the target topic; Based on the players' sentiment tags for the target topic, community topic features are generated, and the hot topic trend database is updated.

8. The method according to claim 5, characterized in that, The knowledge graph is constructed based on real-time features in the real-time feature library, community topic features in the hot topic trend library, and the relationship between real-time features, community topic features, and e-commerce SKUs, including: Extract entity relationships from real-time features in the real-time feature library and construct triples corresponding to the combat system; Extract the entity relationships of community topic features from the hot topic trend database and construct the corresponding triples for the community system; The relationships between the entities in the combat system, the community system, and the mall SKUs are explored, and entities from different systems are merged to obtain triplets formed by the intersection of entities from different systems. A knowledge graph is constructed based on the triples corresponding to the combat system, the triples corresponding to the community system, and the triples formed by the intersection of entities in different systems.

9. The method according to claim 1, characterized in that, Before obtaining the information recommendation strategy and interaction strategy corresponding to the target triggering scenario, the method further includes: Based on the player's immersion in different types of target-triggered scenarios in each system, determine the exposure level of target recommendation information for each target-triggered scenario in each system; Based on the exposure level and the interface attributes of the target triggering scene in the system, an interaction strategy is determined for the triggering scene in the system.

10. The method according to claim 9, characterized in that, The target triggering scenarios include game scenes, game transition interfaces, game user community platforms, and game store interfaces. The exposure levels of the game scenes, game transition interfaces, game user community platform, and game store interface increase sequentially.

11. The method according to claim 1, characterized in that, The interactive behavior data includes: AI dialogue data of interaction with the AI ​​assistant of the target system; Combat operation data, map location data, and item operation data in the combat system; Information posting data, interaction data, and browsing behavior data within the community system; Browsing data, search data, and consumption history data of the e-commerce system.

12. The method according to claim 11, characterized in that, Before acquiring the interaction behavior data generated by the player's target interaction behavior in the target system of the target game, the method further includes: In response to receiving a launch operation for the AI ​​assistant of the target system, the AI ​​assistant of the target system is activated.

13. The method according to claim 1, characterized in that, The step of generating and outputting target recommendation information that corresponds to at least one system and matches the target triggering scenario of that system, based on the information recommendation strategy, interaction strategy, and target entity node, includes: Based on the target entity node, the heterogeneous data fusion database, and the information recommendation strategy, initial recommendation information is determined; Based on the interaction strategy, the initial recommendation information is rendered and adapted to generate target recommendation information that matches the interface specifications of the target triggering scenario, and then output to the interface of the target triggering scenario.

14. A game information recommendation device, characterized in that, The device includes: The acquisition module is used to acquire interaction behavior data generated by the player's target interaction behavior in the target system of the target game; The first determining module is used to calculate the matching degree between the interactive behavior data and the entity nodes of the knowledge graph in the pre-configured heterogeneous data fusion database, and to determine the target entity nodes whose matching degree meets the preset matching degree conditions. The matching module is used to obtain the information recommendation strategy and interaction strategy corresponding to the target triggering scenario when the matching degree meets the preset recommendation conditions corresponding to the target triggering scenario of any system in which the player is currently located. The generation module is used to generate and output target recommendation information that corresponds to at least one system and matches the target triggering scenario of that system, based on the information recommendation strategy, the interaction strategy, and the target entity node.

15. An electronic device, characterized in that, include: The device includes a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the electronic device is running, the processor communicates with the memory via the bus. When the machine-readable instructions are executed by the processor, they perform the steps of the game information recommendation method as described in any one of claims 1 to 13.

16. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, performs the steps of the game information recommendation method as described in any one of claims 1 to 13.

Images