Information processing method and apparatus, electronic device, and medium

Abstract

The application relates to an information processing method and device, electronic equipment and medium. The method comprises the following steps: acquiring a state data item set of at least one scene unit in a virtual scene; determining a candidate recommendation item corresponding to each scene unit from a preset target recommendation rule library according to the state data item set of each scene unit; determining a target recommendation item matched with role state information of an operating role in a game according to a matching relationship between the role state information and the recommendation item and the candidate recommendation item corresponding to each scene unit, and outputting a recommendation result corresponding to the target recommendation item. By adopting the technical scheme, the maintenance cost can be effectively reduced, and the performance controllability in the information processing process can be ensured.

Description

Technical Field

[0001] This disclosure relates to the field of computer technology, specifically to an information processing method, apparatus, electronic device, and medium. Background Technology

[0002] In farm simulation games, the virtual scene consists of a large number of logical grids, each with multiple operable states, such as: empty land waiting to be tilled, tilled land waiting to be sown, sown land waiting to be watered, crops waiting to be harvested, and ripe land waiting to be picked. When the player character moves around the farm, recommended actions are usually triggered based on the character's currently held items and the state of the surrounding grids to enhance the game's playability.

[0003] In related technologies, each type of grid component in farm simulation games typically uses a branching tree-style triggering logic. By judging the grid type, crop type, growth stage, watering status, and other dimensions one by one, recommendation behavior is triggered when specific combination conditions are met. This makes the existing recommendation method costly to maintain and its performance uncontrollable. Summary of the Invention

[0004] This application provides an information processing method, apparatus, electronic device, and medium that can effectively reduce maintenance costs and ensure performance controllability during information processing.

[0005] In a first aspect, embodiments of this application provide an information processing method, including: Obtain a state data item set of at least one scene unit in a virtual scene, wherein the state data item set is a combination of data characterizing the state of the scene unit; Based on the state data item set of each scene unit, candidate recommendation items corresponding to each scene unit are determined from a preset target recommendation rule library, wherein the target recommendation rule library includes a mapping relationship between state data item sets and recommendation items; Based on the matching relationship between the character's status information in the game and the recommended items, as well as the candidate recommended items corresponding to each scene unit, the target recommended item that matches the character's status information is determined, and the recommendation result corresponding to the target recommended item is output.

[0006] Secondly, embodiments of this application provide an information processing apparatus, including: A state data acquisition module is used to acquire a state data item set of at least one scene unit in a virtual scene, wherein the state data item set is a combination of data characterizing the state of the scene unit; The candidate recommendation acquisition module is used to determine the candidate recommendation items corresponding to each scene unit from a preset target recommendation rule library based on the state data item set of each scene unit. The target recommendation rule library includes the mapping relationship between the state data item set and the recommendation items. The recommendation module is used to determine the target recommendation item that matches the character's state information based on the matching relationship between the character's state information and the recommended items, as well as the candidate recommended items corresponding to each scene unit, and output the recommendation result corresponding to the target recommendation item.

[0007] Thirdly, embodiments of this application also provide an electronic device, including a memory storing multiple instructions; a processor loads instructions from the memory to execute the steps of any of the information processing methods provided in embodiments of this application.

[0008] Fourthly, embodiments of this application also provide a computer-readable storage medium storing a plurality of instructions adapted for loading by a processor to execute the steps of any of the information processing methods provided in embodiments of this application.

[0009] Fifthly, embodiments of this application also provide a computer program product, including a computer program or instructions, which, when executed by a processor, implement the steps in any of the information processing methods provided in embodiments of this application.

[0010] The solution adopted in this application involves obtaining a state data item set of at least one scene unit in a virtual scene; determining candidate recommendation items corresponding to each scene unit from a preset target recommendation rule library based on the state data item set of each scene unit; determining a target recommendation item matching the character's state information based on the matching relationship between the character's state information and the recommendation items, and the candidate recommendation items corresponding to each scene unit, and outputting the recommendation result corresponding to the target recommendation item; based on the above technical solution, a direct mapping relationship between the state data item set and the recommendation item is established through the preset target recommendation rule library, transforming the recommendation decision from traditional multi-branch step-by-step condition judgment to a matching search based on the mapping relationship. When it is necessary to adjust the recommendation rules, only the mapping data in the rule library needs to be modified without changing the code logic, thereby effectively reducing maintenance costs. At the same time, since the matching is performed using the mapping relationship, the exponential performance overhead caused by nested branches is avoided, thereby ensuring the performance controllability in the information processing process. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1This is a schematic flowchart of one embodiment of the information processing method provided in this application. Figure 2 This is a schematic diagram of a process for constructing a target recommendation rule base provided in an embodiment of this application; Figure 3 This is a schematic diagram of the structure of an information processing device provided in the embodiments of this application; Figure 4 This is a schematic diagram of the structure of an information processing system provided in the embodiments of this application; Figure 5 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0013] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application. At the same time, in the description of the embodiments of this application, the terms "first," "set," etc., are only used to distinguish descriptions and should not be construed as indicating or implying relative importance. Thus, features defined with "first" or "set" may explicitly or implicitly include one or more features. In the description of the embodiments of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0014] Specifically, this embodiment will be described from the perspective of an information processing device, which can be integrated into an electronic device. That is, the information processing method of this embodiment can be executed by an electronic device. Optionally, the electronic device may include a terminal device. The terminal device may be a mobile phone, tablet computer, smart Bluetooth device, laptop computer, game console, or personal computer (PC), etc.

[0015] The information processing method provided in this application can be applied to information processing systems. These systems may include terminal devices and servers. The terminal devices may be devices including receiving and transmitting hardware, i.e., devices with receiving and transmitting hardware capable of performing bidirectional communication over a bidirectional communication link. The terminal devices and servers can communicate bidirectionally via a network.

[0016] Optionally, the server can be a standalone server, or a server network or server cluster, including but not limited to computers, network hosts, single network servers, multiple network server sets, or cloud servers composed of multiple servers. Cloud servers consist of a large number of computers or network servers based on cloud computing.

[0017] The following detailed description is provided in conjunction with the accompanying drawings. In this embodiment, the execution subject is a terminal device as an example. It should be noted that the order of description in the following embodiments is not intended to limit the preferred order of the embodiments. Although a logical order is shown in the flowcharts, in some cases, the steps shown or described may be performed in a different order than that shown in the accompanying drawings.

[0018] To facilitate understanding of the technical solution of this application, the relevant concepts involved in this application will be introduced first.

[0019] Please refer to Figure 1 The specific process of this information processing method can be summarized in steps 101 to 103 as follows: Step 101: Obtain the state data item set of at least one scene unit in the virtual scene, wherein the state data item set is a combination of data representing the state of the scene unit.

[0020] In some embodiments, the virtual scene can be a scene in a game, such as a farm simulation game, or a casual game, simulation management game, or sandbox game that includes farm gameplay modules such as farm management, garden planting, and crop cultivation. Furthermore, a scene unit refers to the smallest geographical logical unit formed after logically dividing the virtual scene. Each scene unit corresponds to a fixed area on a map, each scene unit has a unique identifier and scene unit type, and each scene unit can also be associated with at least one state attribute (such as crop type, growth stage, watering status, etc. within the scene unit).

[0021] In some embodiments, a state data itemset is used to represent an ordered data tuple representing the state of a scene unit. The state data itemset includes at least a scene unit type field. It may also include extended state fields corresponding to the scene unit type. These extended state fields include at least one of the following: the identifier of an object within the scene unit, its growth stage, and its interaction state. Specifically, the identifier of an object within the scene unit represents the category and identity information of the crop, vegetation, or scene object contained within that scene unit; the growth stage represents the entire growth process of the crop within the scene unit, from sowing and watering to growth and eventual harvest; and the interaction state represents the types of operational behaviors that the current scene unit can support. The interaction state may include multiple independent fields, each representing an executable interactive operation corresponding to an object within the scene unit. The interaction state may specifically include watering state fields, fertilizing state fields, harvesting state fields, and reaping state fields, corresponding to whether the scene unit supports watering, fertilizing, fruit harvesting, and crop harvesting, respectively.

[0022] In some embodiments, scene units may include scene grids such as farmland type, flowerbed type, and woodland type that have crop growth and multiple state changes.

[0023] Specifically, a state data itemset is a tuple with fields arranged in an ordered manner. A state data itemset can consist of a scene unit type field and one or more extended state fields arranged in a fixed order. For example, taking a farmland scene unit as an example, the fields of the farmland type state data itemset could be ordered as scene unit type field + crop type field + growth stage field + watering status field. Alternatively, the fields could be ordered as scene unit type field + crop type field + growth stage field + watering status field + harvesting field, etc. By controlling the order of the state data itemset fields, the plot attributes, crop type, growth progress, and interactive status of the scene unit can be fully represented.

[0024] Of course, the fields in the state data item set can also include only the scene unit type field, without configuring additional extended state fields. In this case, only the scene unit type corresponding to the scene unit type field is included. This scene unit type can be aquatic scene units such as pond, lake, and stream, which have no crops growing, simple state changes, and fixed interaction methods. For example, taking the pond type scene unit as an example, the state data item set for the pond type can simply contain the scene unit type field. The pond scene unit attribute can be uniquely determined by the single scene unit type field, and the corresponding recommendation rules and interaction behaviors can be matched without additional state fields.

[0025] In some embodiments, the state data itemset of a water-type scene unit may also include extended state fields. In this case, the content of the extended state fields contained in the state data itemset of the water-type scene unit can be filled with default values, such as the value 0, null, or a preset placeholder identifier, to maintain the uniformity and regularity of the data structure of various state data itemsets. For example, taking a stream-type scene unit as an example, the state data itemset corresponding to the stream scene unit can be represented as (6, null), where the first digit 6 is the scene unit type code corresponding to the stream type, and the subsequent extended state fields are uniformly filled with null values. During the rule matching process, the matching can be completed directly based on the scene unit type field, ignoring the default placeholder extended state fields. This ensures that all state data itemsets have a uniform format, facilitates unified parsing and rule retrieval, and eliminates the need to adapt differentiated data structures separately for water-type scene units, thereby improving data standardization and adaptability.

[0026] In some embodiments, it is possible to monitor in real time whether a recommendation trigger event has occurred. The recommendation trigger event may be, for example, the operation character moving, the operation character actively switching the handheld prop, or the state of any scene unit changing (such as changing to "wet" after watering, the crop growth stage advancing, or changing to "empty land" after harvesting). After a recommendation trigger event is detected, in response to the recommendation trigger event, the scene unit state interface is called to obtain the state data item set of at least one scene unit.

[0027] In some embodiments, when a recommendation trigger event is triggered, a preset scene unit state interface is invoked to determine the state data item set of at least one scene unit. The at least one scene unit can be a set of candidate scene units determined based on the current location of the operating character and a preset range, or it can be all currently displayed scene units; this specification does not impose specific limitations.

[0028] In some embodiments, a unified scene unit state interface is predefined for each scene unit, such as GetRuleTuple(). This interface allows the retrieval of the state data item set for each scene unit. Specifically, the scene unit state interface can be used to obtain the scene unit type of each scene unit, and, as needed, match information such as the object's identifier, growth stage, and various interaction states within the scene unit. This information is then encapsulated according to a fixed field order to generate a standardized state data item set, thereby obtaining the state data item set for each scene unit within at least one scene unit.

[0029] In some embodiments, taking a farmland-type scene unit as an example, the state data itemset of this type of scene unit includes the following fields in order: scene unit type, crop number, watering status, growth stage, and whether it is harvested with a sickle. Accordingly, the tuple returned by the state interface of a farmland-type scene unit can be represented as: (scene unit type, crop number, watering status, growth stage, whether it is harvested with a sickle). For example, a specific farmland state data itemset could be (1, 4097, 0, 2, 0), where the first "1" indicates that the scene unit type is farmland, the second "4097" indicates that the crop is wheat, the third "0" indicates that it is not watered, the fourth "2" indicates that the growth stage is withered, and the fifth "0" indicates that it is not harvested with a sickle.

[0030] In some embodiments, taking a pond-type scene unit as an example, since this type of scene unit has no complex states, the state data item set includes the following fields: scene unit type, and the extended state field is empty; correspondingly, the tuple returned by the state interface of the pond-type scene unit can be (scene unit type, ). For example, the state data item set of a scene unit is (2, ), where the first "2" indicates that the scene unit type is pond, and the extended state field after the first "2" is empty, indicating that the pond-type scene unit has no additional states.

[0031] In some embodiments, in gesture recommendation scenarios, the scene unit state interface, when called, additionally passes the current item slot parameter of the operating character, thereby further including the current item information of the operating character in the state data item set. This supports recommending different gestures for the same scene unit based on different current item information of the operating character. In this case, the return value of the scene unit state interface in the gesture recommendation scenario can be represented as (scene unit type, crop number, watering status, growth stage, whether to harvest with a sickle, item slot). In some embodiments, the state data item set is set as an immutable value type structure, which can be directly used as the search key of a hash dictionary. This enables constant-order fast hash lookups in the recommendation rule table, significantly simplifying multi-condition branch decision logic and improving rule matching efficiency and operational stability.

[0032] In practical applications, the state data set of a scene unit can be obtained through the scene unit state interface, for example, (1, 4097, 0, 2, 0). In this state data set, scene unit type = 1 (representing farmland), crop = 4097 (representing wheat), watering = 0 (representing no watering), growth stage = 2 (representing withered), and whether it is harvested by sickle = 0 (representing not harvested by sickle). If another scene unit's state data set is obtained, for example, (1, 4097, 1, 2, 0), where scene unit type = 1 (representing farmland), crop = 4097 (representing wheat), watering = 1 (representing watered), growth stage = 2 (representing withered), and whether it is harvested by sickle = 0 (representing not harvested by sickle), or if another scene unit's state data set is obtained, for example, (2,), where scene unit type = 2 (representing pond), there are no additional states. Thus, through the scene unit status interface, the inherent attributes and real-time status information of the scene unit can be adaptively filled into the scene unit type field and the corresponding extended status field according to the differences in the types of different scene units. For scene units such as farmland with multi-dimensional status changes, extended fields such as crop identification, growth stage, and various interaction states are automatically filled in. For scene units such as ponds with a single status and no additional attribute changes, only the scene unit type field needs to be retained, without redundant filling of invalid fields, so as to achieve differentiated and adaptive encapsulation of different types of scene units.

[0033] Based on the above scheme, the scene unit state interface can automatically adapt to the feature differences of different types of scene units, filling the scene unit type and extended states into the state data item set in a unified format. This eliminates the need to write separate data reading logic for each scene unit, demonstrating strong versatility and scalability. Furthermore, all scene unit attributes and states are carried by standardized state data item sets, with a unified and standardized data format. This provides a regular and reliable input for subsequent rule matching and hash table lookups, making the recommendation logic more concise and clear.

[0034] Step 102: Based on the state data item set of each scene unit, determine the candidate recommendation items corresponding to each scene unit from the preset target recommendation rule library. The target recommendation rule library includes the mapping relationship between the state data item set and the recommendation items.

[0035] In some embodiments, a target recommendation rule base can be pre-set, wherein, see Figure 2 The steps for building a target recommendation rule base include: Step 201: Obtain the preset recommended configuration data. The recommended configuration data includes multiple recommended items. Each recommended item is mapped to at least one set of state data items. Each set of state data items is associated with a recommended score.

[0036] In some embodiments, the recommended configuration data is a structured configuration file pre-configured by the game designer. This data can be in the form of an Excel spreadsheet, JSON configuration file, Lua configuration table, or data table resource. The recommended configuration data is configured externally, independent of the game program code, without requiring developers to modify the underlying logic. Game designers can independently add, remove, or adjust the mapping relationships between recommended items, status data sets, and recommended scores according to actual gameplay needs. Once configured, the recommended configuration data can be directly imported into the game project. After loading and parsing, it can participate in the construction of the target rule base, achieving configurable management of the recommendation logic. This decouples gameplay rules from program code, facilitating later iterations, optimizations, version updates, and content maintenance.

[0037] In some embodiments, the recommendation configuration data includes recommended items corresponding to all or part of the interactive scenarios in the game. These recommended items include recommended items, recommended scene units, and recommended gestures. Each recommended item is mapped to at least one set of state data items, and each mapping is associated with a recommendation score, used to prioritize multiple candidate recommended items. In this way, a unified set of recommendation configuration data can cover different types of recommendation needs in the game (such as prompting characters to switch items, highlighting operable scene units, and displaying interactive gestures), ensuring centralized management of recommendation rules and allowing designers to flexibly adjust recommendation strategies for different scenarios.

[0038] Step 202: Reverse the recommended configuration data to generate a hash mapping table with the status data item set as the key and the recommended item as the value.

[0039] Specifically, the recommendation configuration data indexed by recommendation items can be inverted. The status data item set within the recommendation configuration data becomes the primary key for retrieval, and the corresponding recommendation items, recommendation scores, and other information are integrated into corresponding key-value pairs, thus constructing a hash mapping table. This inversion transforms the traditional retrieval method of matching scenario units by recommendation item into a fast query structure that directly retrieves recommendation items by status data item set. This eliminates the need for branch traversal and allows for rapid matching solely through the key-value index.

[0040] Step 203: Use the hash mapping table as the target recommendation rule base.

[0041] Specifically, after determining the hash mapping table, the generated hash mapping table is loaded into the game client's memory, serving as the target recommendation rule base directly used by the runtime recommendation engine. The target recommendation rule base uses the state data item set as the sole retrieval criterion, enabling the game to quickly retrieve corresponding recommendation items by directly using the real-time acquired state data item set as the search key, thus providing rule support for subsequent recommendation filtering and output.

[0042] In some embodiments, taking a recommendation configuration table as an example, planners can configure each recommendation item, its triggering rules, and recommendation scores in the recommendation configuration table. The recommendation configuration table is an external structured configuration file that can be edited and maintained in tabular form. Planners configure the corresponding scene unit state triggering rules and matching recommendation scores for each recommendation item.

[0043] In some embodiments, the recommended configuration data can be set differently depending on the type of the recommended item. Since the types of recommended items can include recommended scene units, recommended props, and recommended gestures, the recommended configuration data can be correspondingly divided into scene unit recommended configuration data, prop recommended configuration data, and gesture recommended configuration data. The following example uses recommended scene units as the recommended item type and the recommended configuration data in tabular form. The scene unit recommended configuration table is shown in Table 1 below. This table includes the identifier, status data item set, and recommended score for each recommended item.

[0044] Table 1

[0045] In Table 1, the first field of each state data item set is the scene unit type field, where 1 represents farmland, 2 represents pond, 3 represents dry land, 5 represents lake, 6 represents stream, and 7 represents mine. The second field of the state data item set is: 1539 represents debris, 4097 represents wheat, 2819 represents ore, 5127 represents seeds, and 5128 represents trees. The third and subsequent fields of the state data item set represent different status information such as prop type, growth stage, watering status, and whether it is harvested with a sickle, depending on the scene unit type. For example, in the farmland state data item set (1, 4097, 0, 2, 0), the third field 0 indicates no watering, the fourth field 2 indicates the growth stage (withered), and the fifth field 0 indicates it is not harvested with a sickle; in the tree state data item set (3, 5128, 5, 0), the third field 5 indicates the growth stage, and the fourth field 0 indicates it is not harvested with an axe. For example, in (1, 4097, 0, 0, 0), the scene unit type is 1 (farmland); crop number is 4097 (wheat); watering status is 0 (not watered); growth stage is 0 (seed / newly sown); whether to harvest with a sickle is 0 (no); in (1, 5128, 5, 0), the scene unit type is 1 (farmland); crop number is 5128 (tree); growth stage is 5 (mature and sap can be collected); whether to place a sap collector is 0 (no); in (1, 1539, 1), the scene unit type is 1 (farmland); crop number is 1539 (debris; can be cleared with an axe = 1, indicating that the debris needs to be cleared with an axe, such as branches); in (1, 1539, 2), the scene unit type is 1 (farmland); crop number is 1539 (debris); interaction status flag is 2 (indicating that the debris needs to be cleared with a pickaxe, such as stones or a mixture of stones).

[0046] In some embodiments, miscellaneous items may include various items such as branches, stones, and weeds. These items can be distinguished by the required tools. For example, the required tool for branches is 1, which represents an axe; the required tool for stones is 2, which represents a pickaxe; and the required tool for weeds is 3, which represents a sickle.

[0047] For example, referring to Table 1, taking the recommended item as a sickle as an example, the sickle is identified as 101. The trigger rule list corresponding to the sickle includes [(1, 1539, 3), (1, 4097, 0, 2, 0), (1, 4097, 1, 2, 0)] and the corresponding score list: [70, 90, 90]. The numerical meanings of each status data item set are as follows: (Scene unit type = 1 farmland, crop = 1539 miscellaneous items, prop type = 3 sickle); (Scene unit type = 1 farmland, crop = 4097 wheat, watering = 0 unwatered, growth stage = 2 withered, whether to harvest with sickle = 0 no); (Scene unit type = 1 farmland, crop = 4097 wheat, watering = 1 watered, growth stage = 2 withered, whether to harvest with sickle = 0 no). And, taking the recommended item as a fishing rod as an example, the fishing rod is identified as 201, the trigger rule list corresponding to the fishing rod is: [(2,), (5,), (6,)] and the corresponding score list is: [100, 100, 100]. Among them, the numerical meanings of each state data item set are: (scene unit type = 2 pond), (scene unit type = 5 lake), (scene unit type = 6 stream).

[0048] Furthermore, after determining the scene unit recommendation configuration table, resource import is performed. At this point, the aforementioned forward mapping relationship with recommendation items as the main body can be automatically reversed and converted into a reverse mapping structure with state data item sets as keys and recommendation items and score information as values, generating a scene unit recommendation rule library that can be directly loaded and run. During game runtime, the reversed scene unit recommendation rule library is loaded into memory, forming the following mapping relationship: Recommendation rule mapping structure = {(1, 1539, 3): [(101, 70)], (1, 4097, 0, 2, 0): [(101, 90)], (1, 4097, 1, 2, 0): [(101, 90)], (2,): [(201, 100)], (5,): [(201, 100)], (6,): [(201, 100)]}. Of course, to simplify the recommendation rule mapping structure, (1, 1539, 3): [(101, 70)] can be adjusted to (1, 1539, 3): [(101, 0)], where 0 represents the first value in the score list [70, 90, 90]; for example, (5,): [(201, 100)] can also be adjusted to (5,): [(201, 1)], where 1 represents the second value in the score list [100, 100, 100].

[0049] Thus, by reversing the scene unit recommendation configuration table to form a hash mapping table with state data item sets as keys, the traditional multi-condition branch traversal judgment is optimized into a constant-level hash query, which can effectively improve retrieval speed and enhance game smoothness. Furthermore, the target recommendation rule base exists independently of the recommendation engine logic. When adding scene unit types, crop states, interaction methods, or recommendation items, only the recommendation configuration data needs to be updated, and then the target recommendation rule base needs to be updated, without modifying the core recommendation logic. This conforms to the open / closed principle and significantly reduces system coupling and later maintenance costs.

[0050] In some embodiments, for the prop recommendation configuration data and gesture configuration data, the prop rule recommendation library and gesture rule recommendation library can be obtained in the same way as the scene unit recommendation data described above, such that the target rule recommendation library includes one or more of the scene unit rule recommendation library, prop rule recommendation library, and gesture rule recommendation library. Furthermore, when the target rule recommendation library includes multiple rule recommendation libraries, for each scene unit, the state data itemset of that scene unit can be searched in each rule recommendation library to obtain the recommended items that each scene unit matches in each rule recommendation library. The recommended items that each scene unit matches in all recommendation rule libraries are then used as candidate recommended items for the scene unit.

[0051] In some embodiments, in response to the operation of adding a new scene unit type, a new state data item set corresponding to the new scene unit type can be added to the recommendation configuration data; and the target recommendation rule base can be updated based on the new state data item set.

[0052] Specifically, upon receiving an operation to add a new scene unit type, in response to the operation triggered by developers or planners, the system retrieves the parameters of the new scene unit type, including but not limited to the type identifier of the new scene unit (e.g., adding an "orchard scene unit" with an assigned type identifier of 8), the basic status fields corresponding to the scene unit (e.g., tree type, fruit status, maturity, etc.), the matching recommended items (e.g., picking tools, fertilizing tools, harvesting gestures, etc.), and the corresponding recommended score. Secondly, in the preset recommended configuration data (i.e., the planner configuration table), a new status data item set corresponding to the new scene unit type is added: following the system's unified status data item set encoding rules, the type identifier, basic status fields, and matching tool association identifiers of the new scene unit are encoded into a fixed-format, immutable, and hashable new status data item set. This ensures that the format of the new status data item set is consistent with the tuple format of existing farmland scene units, pond scene units, etc., in the system, guaranteeing compatibility for subsequent hash lookups. At the same time, the association between the newly added state data item set and the corresponding recommended items is supplemented in the recommended configuration data. That is, the candidate recommended items, recommended scores, and triggering conditions of the recommended items (such as whether specific handheld props or backpack props are required) are clarified for the newly added scene unit type under different states (such as the "immature" and "picking" states of the orchard scene unit), thus completing the update of the recommended configuration data.

[0053] Finally, based on the newly added state data itemsets and supplementary relationships, the target recommendation rule base is automatically updated. For example, a preset import tool can be used to reverse the mapping of the updated recommendation configuration data, using the newly added state data itemsets as keys and the corresponding list of recommendation items (including recommendation scores, trigger conditions, etc.) as values, and adding them to the original target recommendation rule base (hash mapping table). If the newly added state data itemsets have associated existing recommendation items, the recommendation items and scores corresponding to that tuple are updated synchronously to ensure that the target recommendation rule base and recommendation configuration data are synchronized in real time. Furthermore, after the update is complete, the state information of the newly added scene units can be obtained through the state interface, encoded into a new state data itemset, and then a hash lookup can be performed in the target recommendation rule base to achieve candidate recommendation item matching for the new scene unit type, without needing to restart the system or modify the core retrieval and matching logic.

[0054] Thus, through the aforementioned extension mechanism, without modifying core code logic such as the recommendation engine, state interface, and hash lookup, adaptation for new scene unit types can be completed simply by updating the recommendation configuration data and the target recommendation rule base, significantly reducing the development cost and cycle of game version iterations. Furthermore, the state data itemset of the new scene unit follows the system's unified encoding rules, maintaining the same format as the original state data itemset. The update of the target recommendation rule base uses the original import table reversal logic, ensuring that the recommendation process for the new scene unit type is completely consistent with the original scene units, avoiding recommendation anomalies caused by format differences and reducing later maintenance costs.

[0055] In some embodiments, after the target recommendation rule base is pre-created and loaded, and after obtaining the state data item set of each scene unit through step 101, each state data item set is used as an independent retrieval identifier to traverse the target recommendation rule base that has been loaded in memory. For each scene unit, key-value precise matching is performed in the target recommendation rule base using the state data item set of that scene unit to quickly hit all recommendation items associated with the state data item set of that scene unit, and all the hit recommendation items are used as candidate recommendation items corresponding to that scene unit. Among them, state data item sets with different semantics and different field combinations can match different types of candidate recommendation items such as recommended props, recommended scene units, and recommended gestures; at the same time, the matching results carry corresponding recommendation scores, providing a weight basis for subsequent selection of target recommendation items. For example, if the state data item set of a certain scene unit is (1, 1539, 3), then using (1, 1539, 3) as the key, a search is performed in the target recommendation rule base. If (101, 0) and (1, 1539, 3) are found in the target recommendation rule base, then the candidate recommendation item for this scene unit is determined to be the sickle (101), and the corresponding recommendation score is the first value in [70, 90, 90], which is 70. In this way, constant-level fast retrieval can be achieved based on the characteristics of the hash mapping table, without the need for multi-condition branch traversal judgment, resulting in high matching efficiency, stable running time, and improved information processing response speed. Moreover, the standardized state data item set can uniformly adapt to various scene unit types such as farmland, forest land, mines, and water areas, and is compatible with multiple recommendation items such as recommended props, recommended scene units, and recommended gestures, demonstrating wide adaptability and strong versatility.

[0056] In some embodiments, a single index query can be performed in the target recommendation rule base based on the hash value of the state data item set of each scene unit to determine the candidate recommendation item corresponding to each scene unit.

[0057] In some embodiments, to further improve the query efficiency of candidate recommendations and ensure real-time response of game interactions, a single index query can be performed in the target recommendation rule base based on the hash value of the state data item set of each scene unit to quickly determine the candidate recommendations corresponding to each scene unit.

[0058] Specifically, the current state information of each scene unit can be obtained through the state interface. Following preset encoding rules, the scene unit's type identifier, state fields, associated parameters, and other information are encoded into a fixed-format, immutable, and hashable state data item set (including existing and newly added state data item sets). This ensures the uniqueness and hashability of each state data item set, providing a foundation for subsequent hash calculations. Then, a hash calculation is performed on each scene unit's state data item set to generate a unique hash value. For example, built-in hash functions (such as Python's `hash()` or `__hash__`) can be used to perform a one-way hash operation on the state data item set, obtaining a fixed-length hash value. This hash value corresponds one-to-one with the state data item set and can serve as a unique index identifier for the state data item set. Subsequently, the generated state data item set hash value is used as the retrieval index to perform a single index query operation in the target recommendation rule base (hash map table). Since the target recommendation rule base is pre-generated by a table import tool, it has already established a mapping relationship between the state data item set and the candidate recommendation items. Moreover, the hash map table supports O(1) level constant time retrieval based on hash values. Therefore, the corresponding candidate recommendation item can be directly located through the state data item set hash value without traversing the entire rule base or performing multiple matching queries. After the query is completed, the candidate recommendation items corresponding to the state data item set are directly returned. The list contains all the recommendation items adapted to the current state of the scene unit, the corresponding recommendation scores, and the triggering conditions, providing data support for subsequent selection of target recommendation items based on the role status information of the operation role. If the state data item set hash value does not match the corresponding record in the target recommendation rule base, it is determined that there are no candidate recommendation items for the scene unit, and the query process is terminated.

[0059] Thus, by performing a single index query using the hash value of the state data itemset, and relying on the O(1) level retrieval characteristic of the hash mapping table, candidate recommendations can be quickly located without multiple traversals or multi-condition matching, significantly reducing query time, ensuring real-time response during game interaction, and improving the smoothness of character operation. Moreover, a single index query reduces the computational load of loop judgments and multiple rounds of table lookups. Especially when there are many scene units and complex game scenes, it can effectively reduce the consumption of device computing power, avoid game lag caused by excessive query time, and optimize the overall system performance.

[0060] Step 103: Based on the matching relationship between the character's status information and the recommended items in the game, as well as the candidate recommended items corresponding to each scene unit, determine the target recommended item that matches the character's status information, and output the recommendation result corresponding to the target recommended item.

[0061] In some embodiments, the character's status information includes at least one of the following: current item information, inventory item information, and item slot information. After obtaining the candidate recommendations for each scene unit, the character's status information is correlated and matched with each candidate recommendation to establish an adaptation relationship between the character's game state and the candidate recommendations. Alternatively, the character's status information can also be correlated and matched with each candidate recommendation based on the candidate recommendations and their scores for each scene unit.

[0062] Specifically, for each candidate recommendation item in at least one scene unit, the system filters and compares them based on the current item information, owned items, and operable gestures of the player character. Candidate recommendations that are not compatible with the player character's game state are eliminated, while those that are compatible with the player character's actual game scene, operating habits, and owned items are retained. Then, the candidate recommendations are prioritized according to their recommendation scores, and the one with the highest priority and best matching degree is selected from the suitable candidate recommendations as the final target recommendation item.

[0063] Specifically, after determining the target recommendation item, the corresponding recommendation result is output according to the preset interface display rules: if the target recommendation item is a recommended scene unit, the target recommended scene unit is rendered and displayed using the set display methods such as highlighting and special marking; if the target recommendation item is a recommended prop, a quick switching control is generated in the game interface to facilitate the operation character to switch the held prop with one click; if the target recommendation item is a recommended gesture, the corresponding gesture operation control pops up in the interactive interface to guide the operation character to complete the corresponding interactive behavior.

[0064] By adopting the above technical solution, recommendations no longer rely solely on the single state of scene units. Instead, they integrate character status information such as the character's held items and inventory items for dual matching and filtering. This avoids situations where the recommended item is not held by the character or where the recommended action conflicts with the currently held item, significantly improving the rationality and relevance of the recommendations. Furthermore, when determining the target recommendation, the system can also perform weighted sorting based on the recommendation scores of the candidate items. This allows for automatic selection of the optimal solution from multiple candidate items, achieving intelligent and optimal recommendation that aligns with the game's gameplay logic and the character's operating habits.

[0065] In some embodiments, if the character status information includes the current item information of the character and the candidate recommendation includes recommended scene units, then at least one scene unit can be traversed in order from nearest to farthest, and it can be determined in turn whether there is a recommendation that matches the current item information in the candidate recommendation for each scene unit; the first successfully matched scene unit is taken as the target recommended scene unit, wherein the target recommendation includes the target recommended scene unit.

[0066] Specifically, the system first obtains the current item information of the player character. Items corresponding to this information could include, for example, a sickle, a kettle, or an axe. After identifying candidate recommendations for each scene unit, the system traverses at least one scene unit from nearest to farthest, using the player character's current location as a baseline. During this traversal, for each traversed scene unit, it checks whether the candidate recommendation matches the player character's current item information, and selects the first successfully matched scene unit as the target recommended scene unit. This method, traversing and filtering from near to far based on the player character's location, prioritizes recommending nearby operable scene units, aligning with the player character's natural interaction habits and improving game smoothness and immersion. Furthermore, using the first successfully matched scene unit as the target recommended scene unit reduces the number of scene unit matches required, thereby reducing game processing power consumption and ensuring efficient and smooth interface interaction. Furthermore, by matching the candidate recommendations corresponding to the scene unit with the current item information, invalid recommendations that cannot be operated by the recommended scene unit can be avoided, thereby improving the accuracy and usability of the recommendations and making the user experience better.

[0067] At this point, after determining the target recommended scene unit, the target recommended scene unit can be displayed in a set display mode, wherein the set display mode is different from the display mode of other scene units in at least one scene unit.

[0068] In some embodiments, the display method can be set to highlight, special mark, flash, etc. to render and display the target recommended scene unit. This can make the target recommended scene unit visually distinct from other ordinary scene units in the map, making it easier for the operator to quickly locate the interactive target recommended scene unit without having to manually search and identify them one by one, thus reducing the operator's cognitive burden.

[0069] In some embodiments, the scene units are traversed layer by layer from near to far according to the Manhattan distance, with the scene unit where the operator is located as the center. First, the current interactive scene unit at a distance of 0 (i.e., the scene unit where the operator is currently located) is checked, then the surrounding scene units at a distance of 1 (the four neighboring areas above, below, left, and right) are checked, and then the outer scene units at a distance of 2 (all scene units with a Manhattan distance of 2) are checked.

[0070] In practical applications, when a recommended scene unit is used as the recommendation target, the first scene unit within the game scene units surrounding the player character that perfectly matches the player character's currently held item can be retrieved as the target recommended scene unit. Specifically, a distance-first retrieval strategy is adopted. Centered on the player character's location, surrounding scene units are traversed in a hierarchical order from closest to furthest. The traversal priority is as follows: the currently interacting scene unit at distance 0, the surrounding scene units at distance 1, the outermost scene units at distance 2, and so on, expanding the retrieval range outwards level by level. Once a matching scene unit is found during the retrieval process, the subsequent traversal search is immediately terminated, and no further outward retrieval is performed, ensuring that the closest interactive scene unit to the player character is prioritized for recommendation. Furthermore, for each scene unit to be traversed, the matching and judgment steps are as follows: Step 1: Call the scene unit status interface to obtain the status data item set corresponding to the current scene unit; Step 2: Use the status data item set of the scene unit as the search key to perform a precise hash search in the target recommendation rule library loaded in memory. If no corresponding recommendation item is found, skip the scene unit and continue searching for the next scene unit; Step 3: If the hash search hits the corresponding candidate recommendation item, traverse the hit candidate recommendation items one by one and determine whether the item type associated with the candidate recommendation item is consistent with the current item type of the operating character; Step 4: If the item types of the two match successfully, determine the current scene unit as the target recommendation scene unit and immediately terminate the overall traversal and retrieval process.

[0071] For example, the retrieval process described above is explained below: First, obtain the current handheld item type of the character, the dictionary of surrounding scene units (grouped by distance), and the current interaction grid (which may be empty). Then, construct a candidate scene unit list in order of distance from nearest to farthest: if the current interaction grid is not empty, add it as the first element to the candidate list; subsequently, add all scene units from the distance 1 scene unit list to the candidate list in sequence; finally, add all scene units from the distance 2 scene unit list to the candidate list in sequence. After obtaining the candidate list, traverse each scene unit in the candidate list in order of distance from nearest to farthest. For each scene unit currently traversed, perform the following steps: First, call the status interface of the scene unit to obtain the rule key corresponding to the scene unit; then determine whether the rule key exists in the preset recommendation rule table: if it does not exist, skip the current scene unit and continue processing the next one. If it exists, obtain the recommendation item list corresponding to the rule key and traverse each recommendation item in the list. For each recommended item, check if its associated item type matches the current item type of the player character. If they match, return that scene unit as the target recommended scene unit and terminate the entire search process. If none of the recommended items for the current scene unit match, proceed to the next scene unit. If no matching item is found after traversing all candidate scene units, return empty, indicating no recommended results.

[0072] In some embodiments, the character status information includes the current item information of the operating character, and the candidate recommendation items include the target recommendation item. In this case, at least one candidate recommendation item that is inconsistent with the current item information can be selected from the candidate recommendation items corresponding to each scene unit; from the at least one candidate recommendation item, one candidate recommendation item that meets the item recommendation conditions is determined as the target recommendation item, wherein the target recommendation item includes the target recommendation item.

[0073] In some embodiments, when it is detected that the candidate recommendation includes the target recommended item and the character status information includes the current item information, at least one scene unit can be traversed in order from near to far, based on the current location of the operating character. During the traversal, for each traversed scene unit, it is determined whether the candidate recommended item corresponding to the scene unit is consistent with the current item information. If they are inconsistent, it is determined that the candidate recommended item is inconsistent with the current item information. The above determination is performed for each scene unit to obtain all candidate recommended items that are inconsistent with the current item information as at least one candidate recommended item. Then, one candidate recommended item that meets the item recommendation conditions is selected from the at least one candidate recommended item as the target recommended item.

[0074] In some embodiments, when determining candidate recommended items that are inconsistent with the current item information, the judgment may be made only for a portion of the scene units in at least one scene unit.

[0075] In some embodiments, the item recommendation criteria can be set according to actual needs. For example, it can be at least one of the following: recommendation score priority criteria, item-to-current scene unit scene adaptation criteria, item-to-character inventory holding status criteria, and item-to-function complementarity criteria. The recommendation score priority criteria prioritize candidate recommended items with higher recommendation scores; a higher score indicates a higher degree of matching between the item and the current scene unit state and gameplay scenario. The item-to-current scene unit scene adaptation criteria require that the function type of the candidate recommended item matches the operational needs of the current scene unit; for example, a sickle is suitable for a farmland debris scene, a fishing rod for a water scene, and a pickaxe for a mine stone scene. The item-to-character inventory holding status criteria prioritize candidate recommended items that the character already owns and can use directly, avoiding recommending invalid items that the character does not possess. The item-to-function complementarity criteria select items with functions different from the character's currently held item that can complete the specific operations of the current scene unit, achieving complementary operation functions and improving interaction convenience.

[0076] In some embodiments, when determining the target recommended item, the recommendation score corresponding to each candidate recommended item can also be obtained, and at least one candidate recommended item can be sorted according to the recommendation score; the first candidate recommended item in the sorting result is determined as the target recommended item.

[0077] Specifically, when obtaining the recommendation score for each candidate recommended item, since the target recommendation rule base contains the recommendation score for each recommended item, all candidate recommended items can be sorted from the recommendation score for each candidate recommended item in the target recommendation rule base in descending order of score. After sorting, the first candidate recommended item with the highest score in the sorted sequence is selected as the target recommended item, ensuring the rationality of the priority of the recommendation results and the adaptability to the scenario.

[0078] In some embodiments, if the character's item inventory contains the target recommended item, a quick switch control is displayed on the game's screen. The quick switch control is used to switch the character's handheld item to the target recommended item.

[0079] Specifically, after determining the target recommended item, the system calls the character's inventory status interface to check if the target recommended item exists in the character's inventory. If the target recommended item is found, a quick switch control is rendered and displayed in a preset location on the game display interface (such as the side of the interface or near the bottom quick bar). This control is associated with the identification information of the target recommended item. The character can quickly switch the currently held item to the target recommended item by clicking the control, without having to manually enter the inventory to search for and select the item. This simplifies the operation process, improves the efficiency of switching items, and enhances the user experience.

[0080] In practical applications, when recommended items are used as the recommendation target, recommended items that require the character to switch to a different handheld item to complete the corresponding operation can be selected within the game scene unit around the character. This intelligently guides the character to change items and achieves intelligent item recommendation.

[0081] Specifically, the item recommendation strategy requires comparing the recommendation scores of multiple candidate items. Therefore, a strategy of collecting data first and then sorting is adopted. For example, scene units at a distance of 1 can be scanned in one round with the highest priority. If candidate results are already available, the highest-scoring item is directly output, without expanding to scene units at a distance of 2, because the demand for closer scene units is more urgent. The specific steps of the item recommendation strategy are as follows: Step 1: Obtain the state data item set and look up the table; For each scene unit within the current distance level (processing distance 1 first, then distance 2), the scene unit state interface is called to obtain the state data item set of each scene unit, and a hash lookup is performed in the target recommendation rule base using the state data item set of that scene unit as the key. If the lookup does not find a match (i.e., the scene unit has no recommended items), the scene unit is skipped, and the next scene unit is processed. Step Two: Filter Shortcut Button Markers. For each recommended item found in the lookup table, check if it has a "Shortcut Button" marker. A "Shortcut Button" marker indicates that the recommended item is a candidate item and can be displayed as a UI button. If a recommended item does not have a "Shortcut Button" marker, proceed to the next scene unit. Step Three: Filter Recommended Items Different from the Currently Held Item. Further check the item types associated with the retained recommended items, retaining only those whose item types do not match the character's current item type. That is, only recommend items that need to be switched, avoiding recommending items already held by the character. Step Four: Maintain the Highest Score within the Same Distance Layer. Specifically, within the same distance layer, use a recommended score table (with recommended items as keys and the current highest score as the value) and a scene unit table (with recommended items as keys and the corresponding scene unit where the score appears). For each recommended item that passes the above filtering, obtain its recommended score. If the recommended score is greater than the existing value in the score table (or the recommended item has not yet been recorded), update the score table and scene unit table. Thus, when the same recommended item appears in multiple scene units, only its highest recommended score and corresponding scene unit are retained. Step 5: Output the highest score if there are candidates in the same layer. Specifically, after processing all scene units in the current distance layer, if the score table is not empty, the recommended item with the highest score is selected as the optimal recommended item, and the recommended item and its corresponding scene unit are returned. Due to the distance-first strategy, once a candidate result is obtained in the current distance layer, the algorithm terminates immediately and does not scan further distance layers. If there is no candidate result in the current distance layer, the next distance layer is processed (e.g., distance 2). If no candidate is found after traversing all distance layers, an empty value is returned, indicating that there is no recommended item this time. Furthermore, after determining the optimal recommended item, a quick switch control (e.g., a button) can be generated on the game display interface. The quick switch control can display the icon and name of the recommended item. After the character clicks the quick switch control, the character's handheld item is automatically switched to the target recommended item, and the character can be further guided to operate the corresponding scene unit.

[0082] Specifically, the item recommendation strategy used for the above item recommendation conditions is explained below: First, obtain the player's currently held item type and a dictionary of surrounding scene units grouped by distance (including scene unit lists for distance 1 and distance 2). Then, process each distance level in order of increasing distance: process distance 1 level first; if no recommendation result is generated for this level, then process distance 2 level. For the currently processing distance level (e.g., distance 1), perform the following steps: First, initialize an empty score table (to record the highest score of each recommended item in the current level) and an empty scene unit table (to record the scene unit corresponding to the highest score). Second, traverse all scene units in this distance level. For each traversed scene unit, first call the current scene unit's status interface to obtain the rule key of the scene unit. Then, determine if the rule key exists in the preset recommendation rule table: if it does not exist, skip the current scene unit and continue processing the next one. If it exists, obtain the list of recommended items corresponding to the rule key and traverse each recommended item in it. For each recommendation, the following filtering is performed sequentially: If a recommendation does not have a shortcut button marker, it is ignored. If the item type associated with the recommendation is equal to the currently held item type (i.e., no switching is required), it is ignored. Otherwise, the recommendation score for that item is obtained. If the score is greater than the score currently recorded in the score table (or the recommendation has not yet been recorded), the score table is updated (mapping the recommendation to the current score), and the scene unit table is updated (mapping the recommendation to the current scene unit). After traversing all scene units at the current distance level, the score table is checked for emptiness: If not empty, the recommendation with the highest score is selected as the optimal recommendation, and the scene unit corresponding to the recommendation is obtained from the scene unit table. Then, (optimal recommendation, scene unit) is returned, and the entire retrieval process ends (no further processing of more distant distance levels). If empty, the next distance level (distance 2) is processed. If no candidate recommendations are found after processing distances 1 and 2, an empty value is returned, indicating no recommended item.

[0083] The aforementioned item recommendation strategy achieves efficient and accurate item recommendations through methods such as distance-based layering, shortcut button marking, filtering for inconsistent held items, retaining the highest score within the same layer, and stopping upon the appearance of a candidate. It avoids invalid prompts (not recommending items already available), only displays recommendations suitable for UI shortcut buttons, and prioritizes pushing the highest-scoring item corresponding to the nearest scene unit, aligning with the player's operating habits. Furthermore, due to the use of hash table lookups and a limited number of traversals, computational overhead is extremely low, ensuring real-time response.

[0084] In some embodiments, the candidate recommendation items include recommended gestures, and the target recommended gesture that matches the character status information can be filtered from the candidate recommended gestures corresponding to each scene unit based on the character status information. The target recommendation items include target recommended gestures, and the character status information includes at least one of the following: the character's backpack item information, the character's social status information, and the character's current prop information.

[0085] Specifically, the process begins by obtaining the state data set for each scene unit through the scene unit state interface. This data is then used to look up candidate recommended gestures in the target recommendation rule base hash table. Next, real-time character state information is collected, including the types and quantities of items already owned in the character's inventory, the character's current social interaction status, and the type of handheld item the character is currently using. Each candidate recommended gesture is then matched and validated against the character's various state information. Based on inventory item information, recommended gestures that match the character's existing items and trigger the gesture interaction are selected. Based on social status information, gestures suitable for multi-person interaction and friend collaboration scenarios are selected. Based on current item information, gestures matching the function of the currently held item and enabling corresponding interactive actions are selected. If multiple valid candidate gestures remain after this filtering, a unique target recommended gesture is determined according to preset rules. This multi-dimensional validation ensures a high degree of compatibility between the target recommended gesture and the character's held items, social scenarios, and current operating items, avoiding recommendation bias caused by single-dimensional judgments and improving the rationality and accuracy of gesture recommendations.

[0086] In some embodiments, when calling the scene unit state interface, the current item slot parameter of the operating character can be additionally passed in, so that the generated state data set includes not only the state information of the scene unit itself, but also the state of the operating character's held item. For example, the gesture recommendation rule tuple format of the farmland scene unit is: (scene unit type, crop item type, watering status, growth stage, whether to harvest with a sickle, item slot), which allows the same scene unit to be mapped to different recommended gestures depending on the operating character holding different items (such as holding a watering can vs. empty-handed), thus achieving differentiated guidance.

[0087] In some embodiments, gesture recommendations use separate gesture recommendation configuration data, which has a similar structure to item recommendation configuration data, but the recommended items represent gesture operation types (e.g., "NPC assistance," "follow the character," "cooperative interaction," "place item," "dialogue," etc.). Through reversal processing, a hash map is generated with the (extended) state data item set as the key and the gesture recommendation item list as the value, serving as the gesture recommendation rule base. Thus, by performing a hash lookup in the gesture recommendation rule base, candidate gesture recommendations for each scene unit can be obtained.

[0088] In some embodiments, certain social interaction gestures have the highest recommendation priority in the gesture recommendation process. For example, when recommending gestures, the system can first detect whether the player character is currently in a preset special social state. Exemplary special social states include, but are not limited to: whether an NPC is following the player character, whether there are cooperating players (such as teammates in a multiplayer online scenario), and whether the player character is in a specific event scenario (such as a festival event scenario). If the player character falls into any of these special social states, the gesture type corresponding to that special social state is directly returned (e.g., the "call for NPC assistance" gesture in an NPC-following scenario, and the "cooperative action" gesture in a multiplayer cooperative scenario), without further filtering based on inventory information, general social state information, etc. This priority determination strategy ensures the timeliness of social interaction and game immersion, aligning with the core needs of the player character's social interaction.

[0089] In some embodiments, for candidate gesture recommendations not covered by special social statuses, the triggering conditions for each gesture recommendation can be further validated. If a candidate gesture is associated with an item triggering condition (for example, the corresponding raw materials must be in the backpack when near a processing machine, or ore must be in the backpack when near a furnace), the character's backpack status interface is called to obtain the character's current backpack item information and query whether the specified item ID and corresponding quantity required to trigger the gesture exist in the backpack. At this time, only candidate gesture recommendations that meet the above item triggering conditions are retained, and invalid recommendations that cannot be triggered because the corresponding item is not held in the backpack are removed.

[0090] In some embodiments, if a candidate gesture is associated with a social trigger condition (for example, the NPC must be in a talkable state when approaching it, or the animal must be tamed when approaching it), then the current social status information of the operating character is obtained. This social status information includes, but is not limited to, the target NPC's interaction permission status, favorability threshold, and current dialogue state. In this case, only candidate gesture recommendations that meet the aforementioned social trigger conditions are retained, ensuring that the recommended gestures can be triggered normally in the current social context and improving the effectiveness of gesture recommendations.

[0091] Therefore, after the above steps of screening, if multiple valid candidate gesture recommendations still exist, a unique target recommended gesture can be determined according to one of the following preset rules. For example, it can be sorted by recommendation score. In this case, all valid candidate gestures can be sorted from high to low according to the preset recommendation scores in the gesture recommendation rule library, and the candidate gesture with the highest score can be selected as the target recommended gesture. Of course, it can also be sorted according to preset priority rules. For example, it can be selected as the target recommended gesture according to the preset gesture priority (for example, social gestures > item-related gestures > prop-related gestures).

[0092] In some embodiments, after determining the target recommended gesture, gesture controls corresponding to the target recommended gesture can also be displayed on the game's display interface.

[0093] Specifically, gesture controls can take the form of icons, buttons, or dynamic prompts, and are usually located near the scene unit corresponding to the target recommended gesture (e.g., floating above the scene unit or close to the edge of the scene unit). They can also be fixed in a specific area of ​​the screen (e.g., the shortcut bar). Gesture controls can also display the name of the gesture, the icon, and a brief operation prompt (e.g., "Dialogue", "Put in materials", "Request help").

[0094] In some embodiments, based on the type of the target recommended gesture (such as "NPC assistance," "placing items," "dialogue," etc.), corresponding control styles, icons, text descriptions, and other materials are loaded from the resource manager. These resources can be obtained by associating them with control identifiers in the gesture recommendation configuration data. Furthermore, when displaying gesture controls, the gesture controls can be displayed near the scene unit that triggered the gesture, allowing the operator to intuitively associate the control with the interactive object. For example, for a machine scene unit, the control is displayed above or to the side of the machine; for an NPC scene unit, the control floats above or beside the NPC's head. If the scene unit is off-screen, it can be initially indicated by an edge arrow, and its position dynamically adjusted after the scene unit enters the field of view.

[0095] In some embodiments, click or touch events can be bound to gesture controls. When the character clicks the control, the corresponding gesture interaction logic is immediately executed, such as opening a processing interface, initiating an NPC dialogue, requesting to follow, or engaging in cooperative interaction. Simultaneously, the control's state can be automatically hidden or updated after the interaction to avoid repeated triggering. Furthermore, the gesture control is only displayed during the recommended period. When the character moves out of the scene unit's range, switches handheld items, changes inventory items, or changes social status, causing the gesture to no longer match, the gesture control can be promptly removed or updated to ensure the real-time nature and accuracy of the recommendation information. For example, when the character approaches a processing machine and has the required raw materials in their inventory, the target recommended gesture can be determined as "Place Item," and a control with a "+" icon can be displayed above the machine. When the character clicks this control, the game interface directly pops up a window for placing materials, eliminating the need for the character to manually search for the machine's interaction button. In this way, by automatically identifying and displaying gesture controls, recommended interactive actions can be presented to the user in the form of visual controls. The user does not need to remember or search for the interaction method; they can simply click the control to complete the operation. This is especially effective for novice users or in complex scenarios. Moreover, the gesture controls are refreshed or disappear in real time according to changes in the user's status (backpack, location, social status), ensuring that the recommended content is always effective and avoiding confusion caused by outdated or ineffective guidance.

[0096] Based on the above technical solution, a direct mapping relationship between state data itemsets and recommended items is established by pre-setting a target recommendation rule base. This transforms the recommendation decision-making process from traditional multi-branch, step-by-step conditional judgment to a matching search based on the mapping relationship. When it is necessary to adjust the recommendation rules, only the mapping data in the rule base needs to be modified without changing the code logic, thus effectively reducing maintenance costs. Furthermore, since matching is performed using mapping relationships, the exponential performance overhead caused by nested branches is avoided, thereby ensuring the controllability of performance during information processing.

[0097] This application also provides an information processing device, such as... Figure 3 As shown, it includes: The state data acquisition module 301 is used to acquire a state data item set of at least one scene unit in the virtual scene, wherein the state data item set is a combination of data representing the state of the scene unit; The candidate recommendation acquisition module 302 is used to determine the candidate recommendation items corresponding to each scene unit from the preset target recommendation rule library based on the state data item set of each scene unit. The target recommendation rule library includes the mapping relationship between the state data item set and the recommendation items. The recommendation module 303 is used to determine the target recommendation item that matches the character's state information based on the matching relationship between the character's state information and the recommendation items, as well as the candidate recommendation items corresponding to each scene unit, and output the recommendation result corresponding to the target recommendation item.

[0098] In some embodiments, the recommendation module 303 is configured to, when the character status information includes the current item information of the character and the candidate recommendation items include recommended scene units, traverse at least one scene unit in order from nearest to farthest, and sequentially determine whether there is a recommendation item that matches the current item information among the candidate recommendation items corresponding to each scene unit; and take the first successfully matched scene unit as the target recommended scene unit, wherein the target recommendation item includes the target recommended scene unit.

[0099] In some embodiments, the recommendation module 303 is used to display target recommended scene units in a set display mode, wherein the set display mode is different from the display mode of other scene units in at least one scene unit.

[0100] In some embodiments, the recommendation module 303 is configured to, when the character status information includes the current item information of the operating character and the candidate recommendation items include the target recommendation item, filter at least one candidate recommendation item that is inconsistent with the current item information from the candidate recommendation items corresponding to each scene unit; and determine one candidate recommendation item that meets the item recommendation conditions from the at least one candidate recommendation item as the target recommendation item, wherein the target recommendation item includes the target recommendation item.

[0101] In some embodiments, the recommendation module 303 is used to obtain the recommendation score corresponding to each candidate recommended item, and sort at least one candidate recommended item according to the recommendation score; and determine the first candidate recommended item in the sorting result as the target recommended item.

[0102] In some embodiments, the recommendation module 303 is used to display a quick switch control on the game's display interface if the target recommended item exists in the character's item inventory. The quick switch control is used to switch the character's handheld item to the target recommended item.

[0103] In some embodiments, the recommendation module 303 is configured to, when the candidate recommendation item includes a recommended gesture, filter out the target recommended gesture that matches the character state information from the candidate recommended gestures corresponding to each scene unit based on the character state information, wherein the target recommendation item includes the target recommended gesture, and the character state information includes at least one of the following: the backpack item information of the operating character, the social status information of the operating character, and the current prop information of the operating character.

[0104] In some embodiments, the recommendation module 303 is used to display gesture controls corresponding to the target recommended gesture on the game's display interface.

[0105] In some embodiments, the state data acquisition module 301 is used to call the scene unit state interface to obtain at least one scene unit state data item set in response to a recommendation trigger event, wherein the state data item set is a tuple with fields arranged in an ordered manner, and the state data item set includes at least a scene unit type field.

[0106] In some embodiments, the state data item set further includes an extended state field corresponding to the scene unit type, wherein the extended state field includes at least one of the following: the identifier, growth stage, and interaction state of the object within the scene unit.

[0107] In some embodiments, the device further includes: The recommendation rule base construction module is used to obtain preset recommendation configuration data, which includes multiple recommendation items. Each recommendation item is mapped to at least one state data item set, and each state data item set is associated with a recommendation score. The recommendation configuration data is reversed to generate a hash mapping table with state data item sets as keys and recommendation items as values. The hash mapping table is used as the target recommendation rule base.

[0108] In some embodiments, the device further includes: The recommendation rule base update module is used to respond to the operation of adding a new scene unit type by adding a new state data item set corresponding to the new scene unit type to the recommendation configuration data; and updating the target recommendation rule base based on the new state data item set.

[0109] In some embodiments, the recommendation rule base update module is used to perform a single index query in the target recommendation rule base based on the hash value of the state data item set of each scene unit to determine the candidate recommendation item corresponding to each scene unit.

[0110] In some embodiments, this application also provides an information processing system, specifically as follows: Figure 4As shown. Taking scene units as grids as an example, this recommendation system includes a grid component 40, which includes farmland grids 401, pond grids 402, dryland grids 403, and special grids 404. The recommendation system also includes an interface layer 41, which includes a state interface 411. The state interface 411 is used to encode the current state of each grid (such as grid type, crop type, growth stage, watering status, etc.) into a grid tuple with a fixed format. The grid tuple serves as the unique index key for subsequent recommendations and is input to the data layer 42. The data layer 42 includes a planning configuration table 421, which stores the relationship between recommended items and the list of triggering grid tuples and their corresponding recommendation scores in a forward mapping form. The table is then reversed using a table import tool to generate a hash mapping table with grid tuples as keys and recommended items as values, which is the target recommendation rule base. The recommendation engine 43 performs an O(1) hash lookup in the target recommendation rule base of the data layer 42 based on the grid tuple of the current grid obtained from the state interface 411 to obtain the candidate recommendation items corresponding to each grid, forming a candidate recommendation item list. Subsequently, the recommendation engine 43 obtains the real-time character status information of the operating character (including the held props, backpack items, social status, etc.), filters and sorts the candidate recommendation item list, and determines the target recommendation result. Among them, the output results corresponding to the target recommendation result are divided into three categories: recommended grids are used to highlight the target scene unit that matches the current prop information in the virtual scene, guiding the operating character to interact nearby; recommended props are used to display quick change buttons on the interface, guiding the operating character to switch to props that are more suitable for the current scene; recommended gestures are used to display gesture controls such as NPC assistance, cooperation, or item placement near specific interactive objects, simplifying player interaction operations. This information processing system achieves a unified abstraction of grid status through a status interface, decouples recommendation rules from code logic through a table import tool, ensures real-time interactive performance through hash table lookup, and each module is independently scalable. When adding grid types, recommendation items, or interactive gestures, only the planning configuration table needs to be updated to take effect, without modifying the core code logic.

[0111] Accordingly, this application also provides an electronic device, which can be a terminal, such as a smartphone, tablet computer, laptop computer, touch screen, game console, personal computer (PC), personal digital assistant (PDA), or other terminal device. Alternatively, the electronic device can be a server.

[0112] like Figure 5 As shown, Figure 5This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device 1100 includes a processor 1101 with one or more processing cores, a memory 1102 with one or more computer-readable storage media, and a computer program stored on the memory 1102 and executable on the processor. The processor 1101 and the memory 1102 are electrically connected. Those skilled in the art will understand that the electronic device structure shown in the figure does not constitute a limitation on the electronic device, and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0113] The processor 1101 is the control center of the electronic device 1100. It connects various parts of the electronic device 1100 via various interfaces and lines. By running or loading software programs and / or units stored in the memory 1102, and by calling data stored in the memory 1102, it executes various functions of the electronic device 1100 and processes data, thereby providing overall monitoring of the electronic device 1100. The processor 1101 can be a central processing unit (CPU), a graphics processing unit (GPU), a network processor (NP), etc., and can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application.

[0114] In this embodiment, the processor 1101 in the electronic device 1100 loads the instructions corresponding to the processes of one or more applications into the memory 1102 according to the following steps, and the processor 1101 runs the applications stored in the memory 1102 to realize various functions, such as: obtaining a state data item set of at least one scene unit in a virtual scene, wherein the state data item set is a combination of data representing the state of the scene unit; determining candidate recommendation items corresponding to each scene unit from a preset target recommendation rule library based on the state data item set of each scene unit, wherein the target recommendation rule library includes a mapping relationship between state data item sets and recommendation items; determining a target recommendation item matching the character state information based on the matching relationship between the character state information of the game character and the recommendation item, and the candidate recommendation item corresponding to each scene unit, and outputting the recommendation result corresponding to the target recommendation item.

[0115] The solution adopted in this application establishes a direct mapping relationship between state data item sets and recommended items by pre-setting a target recommendation rule base. This transforms the recommendation decision-making process from traditional multi-branch, step-by-step conditional judgment to a matching search based on the mapping relationship. When it is necessary to adjust the recommendation rules, only the mapping data in the rule base needs to be modified without changing the code logic, thereby effectively reducing maintenance costs. Furthermore, since matching is performed using a mapping relationship, the exponential performance overhead caused by nested branches is avoided, thus ensuring the controllability of performance during information processing.

[0116] For details on the implementation of each of the above operations, please refer to the previous examples, which will not be repeated here.

[0117] Optional, such as Figure 5 As shown, the electronic device 1100 also includes: a touch display screen 1103, a radio frequency circuit 1104, an audio circuit 1105, an input unit 1106, and a power supply 1107. The processor 1101 is electrically connected to the touch display screen 1103, the radio frequency circuit 1104, the audio circuit 1105, the input unit 1106, and the power supply 1107. Those skilled in the art will understand that... Figure 5 The electronic device structure shown does not constitute a limitation on the electronic device and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0118] The touch display screen 1103 can be used to display a graphical user interface (GUI) and receive operation commands generated by the user interacting with the GUI. The touch display screen 1103 may include a display panel and a touch panel. The display panel can be used to display information input by the user or information provided to the user, as well as various graphical user interfaces of the electronic device. These graphical user interfaces can be composed of graphics, text, icons, video, and any combination thereof. Optionally, the display panel can be configured using a liquid crystal display (LCD), organic light-emitting diode (OLED), or other similar technologies. The touch panel can be used to collect touch operations performed by the user on or near it (such as operations performed by the user using a finger, stylus, or any suitable object or accessory on or near the touch panel), generate corresponding operation commands, and execute the corresponding program according to the operation commands. Optionally, the touch panel may include a touch detection device and a touch controller. The touch detection device detects the user's touch location and the signal generated by the touch operation, transmitting the signal to the touch controller. The touch controller receives touch information from the touch detection device, converts it into touch point coordinates, and sends it to the processor 1101. It can also receive and execute commands from the processor 1101. The touch panel can cover the display panel. When the touch panel detects a touch operation on or near it, it transmits the information to the processor 1101 to determine the type of touch event. Subsequently, the processor 1101 provides corresponding visual output on the display panel based on the type of touch event. In this embodiment, the touch panel and the display panel can be integrated into the touch display screen 1103 to achieve input and output functions. However, in some embodiments, the touch panel and the touch display screen 1103 can be implemented as two independent components to achieve input and output functions. That is, the touch display screen 1103 can also be used as part of the input unit 1106 to achieve input functions.

[0119] The radio frequency circuit 1104 can be used to transmit and receive radio frequency signals to establish wireless communication with network devices or other electronic devices, and to transmit and receive signals with network devices or other electronic devices.

[0120] Audio circuit 1105 can be used to provide an audio interface between a user and an electronic device via a speaker and a microphone. Audio circuit 1105 can convert received audio data into electrical signals and transmit them to the speaker, where the speaker converts them into sound signals for output. Conversely, the microphone converts collected sound signals into electrical signals, which are then received by audio circuit 1105, converted back into audio data, and then processed by processor 1101 before being transmitted via radio frequency circuit 1104 to, for example, another electronic device, or output to memory 1102 for further processing. Audio circuit 1105 may also include an earphone jack to provide communication between peripheral headphones and electronic devices.

[0121] The input unit 1106 can be used to receive input numbers, characters, or user characteristic information (such as fingerprints, iris, facial information, etc.), and to generate keyboard, mouse, joystick, optical, or trackball signal inputs related to user settings and function control.

[0122] Power supply 1107 is used to supply power to various components of electronic device 1100. Optionally, power supply 1107 can be logically connected to processor 1101 through a power management system, thereby enabling functions such as charging, discharging, and power consumption management through the power management system. Power supply 1107 may also include one or more DC or AC power supplies, recharging devices, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components.

[0123] although Figure 5 As not shown in the diagram, the electronic device 1100 may also include a camera, sensor, wireless fidelity module, Bluetooth module, etc., which will not be described in detail here.

[0124] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0125] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be performed by instructions, or by instructions controlling related hardware. These instructions can be stored in a computer-readable storage medium and loaded and executed by a processor.

[0126] To this end, embodiments of this application provide a computer-readable storage medium storing multiple computer programs that can be loaded by a processor to execute any of the information processing methods provided in embodiments of this application. The method involves: acquiring a state data item set of at least one scene unit in a virtual scene, wherein the state data item set is a combination of data representing the state of the scene unit; determining candidate recommendation items corresponding to each scene unit from a preset target recommendation rule base based on the state data item set of each scene unit, wherein the target recommendation rule base includes a mapping relationship between state data item sets and recommendation items; determining a target recommendation item matching the character's state information based on the matching relationship between the character's state information and the recommendation items, and the candidate recommendation items corresponding to each scene unit, and outputting a recommendation result corresponding to the target recommendation item.

[0127] For details on the implementation of each of the above operations, please refer to the previous examples, which will not be repeated here.

[0128] The computer-readable storage medium may include: read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0129] Since the computer program stored in the computer-readable storage medium can execute any of the information processing methods provided in the embodiments of this application, it can achieve the beneficial effects that any of the information processing methods provided in the embodiments of this application can achieve, as detailed in the preceding embodiments, and will not be repeated here.

[0130] According to one aspect of this application, a computer program product or computer program is also provided, comprising computer instructions stored in a computer-readable storage medium. A processor of an electronic device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the electronic device to perform the methods provided in the various optional implementations of the above embodiments.

[0131] In the above embodiments of the information processing method, apparatus, computer-readable storage medium, electronic device, and computer program product, the descriptions of each embodiment have different focuses. Parts not described in detail in a particular embodiment can be referred to in the relevant descriptions of other embodiments. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes and beneficial effects of the information processing apparatus, computer-readable storage medium, computer program product, electronic device, and their corresponding units described above can be referred to the description of the information processing method in the above embodiments, and will not be repeated here.

[0132] The foregoing has provided a detailed description of an information processing method, apparatus, electronic device, computer-readable storage medium, and computer program product provided in the embodiments of this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the embodiments above are only for the purpose of helping to understand the methods and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. An information processing method, characterized in that, include: Obtain a state data item set of at least one scene unit in a virtual scene, wherein the state data item set is a combination of data characterizing the state of the scene unit; Based on the state data item set of each scene unit, candidate recommendation items corresponding to each scene unit are determined from a preset target recommendation rule library, wherein the target recommendation rule library includes a mapping relationship between state data item sets and recommendation items; Based on the matching relationship between the character's status information in the game and the recommended items, as well as the candidate recommended items corresponding to each scene unit, the target recommended item that matches the character's status information is determined, and the recommendation result corresponding to the target recommended item is output.

2. The method as described in claim 1, characterized in that, The character status information includes the current item information of the controlled character, and the candidate recommendation items include target recommendation scene units. The step of determining the target recommendation item matching the character status information based on the matching relationship between the character status information and the recommendation items in the game, and the candidate recommendation items corresponding to each scene unit, includes: Traverse the at least one scene unit in order from nearest to farthest, and determine in turn whether there is a recommended item that matches the current prop information among the candidate recommended items corresponding to each scene unit; The first successfully matched scene unit is taken as the target recommended scene unit, wherein the target recommended item includes the target recommended scene unit.

3. The method as described in claim 2, characterized in that, The output of the recommendation result corresponding to the target recommendation item includes: The target recommended scene unit is displayed in a set display mode, wherein the set display mode is different from the display mode of other scene units in the at least one scene unit.

4. The method as described in claim 1, characterized in that, The character status information includes the current item information of the controlled character, and the candidate recommendation items include target recommendation items. The process of determining the target recommendation item matching the character status information based on the matching relationship between the character status information and the recommendation items in the game, and the candidate recommendation items corresponding to each scene unit, includes: From the candidate recommended props corresponding to each scene unit, at least one candidate recommended prop that is inconsistent with the current prop information is selected; From the at least one candidate recommended item, one candidate recommended item that meets the item recommendation criteria is determined as the target recommended item, wherein the target recommended item includes the target recommended item.

5. The method as described in claim 4, characterized in that, The step of determining one candidate recommended item from the at least one candidate recommended item that meets the item recommendation criteria as the target recommended item includes: Obtain the recommendation score corresponding to each candidate recommended item, and sort the at least one candidate recommended item according to the recommendation score; The first candidate recommended item in the ranking results is determined as the target recommended item.

6. The method as described in claim 1, characterized in that, The candidate recommendation items include recommended gestures. The process of determining the target recommendation item matching the character's state information based on the matching relationship between the character's state information in the game and the recommendation items, as well as the candidate recommendation items corresponding to each scene unit, includes: Based on the character status information, target recommended gestures that match the character status information are selected from the candidate recommended gestures corresponding to each scene unit. The target recommended item includes the target recommended gesture, and the character status information includes at least one of the following: the character's backpack item information, the character's social status information, and the character's current item information.

7. The method as described in claim 1, characterized in that, The acquisition of the state data item set of at least one scene unit in the virtual scene includes: Call the scene unit status interface to obtain the status data item set of the at least one scene unit, wherein the status data item set is a tuple with fields arranged in an ordered manner, and the status data item set includes at least a scene unit type field.

8. The method as described in claim 7, characterized in that, The state data item set also includes extended state fields corresponding to the scene unit type, wherein the extended state fields include at least one of the following: the identifier, growth stage, and interaction state of the object within the scene unit.

9. The method according to any one of claims 1-8, characterized in that, The steps for constructing the target recommendation rule base include: Obtain preset recommended configuration data, which includes multiple recommended items. Each recommended item is mapped to at least one set of state data items, and each set of state data items is associated with a recommended score. The recommended configuration data is reversed to generate a hash mapping table with the status data item set as the key and the recommended item as the value; The hash mapping table is used as the target recommendation rule base.

10. The method according to any one of claims 1-8, characterized in that, The step of determining candidate recommendation items corresponding to each scene unit from a preset target recommendation rule base based on the state data item set of each scene unit includes: Based on the hash value of the state data item set of each scene unit, a single index query is performed in the target recommendation rule base to determine the candidate recommendation item corresponding to each scene unit.

Images