Game guiding method and device, computer program product and electronic equipment

Abstract

The invention relates to the technical field of games, and provides a game guiding method and device, a computer program product and electronic equipment. The method comprises the following steps: acquiring a current characteristic value of a game situation characteristic of a current game at a current moment; inputting the current characteristic value into a pre-trained guide opportunity prediction model, and predicting a probability that a first change value of a fight resource of a fight participant of the current game is greater than a preset threshold after a first preset duration based on the guide opportunity prediction model; and according to the probability, determining a game guide opportunity so as to perform game guide on the battle participants of the current game at the game guide opportunity. According to the scheme, the accuracy and timeliness of game guidance can be improved.

Description

Technical Field

[0001] This disclosure relates to the field of game technology, and more specifically, to a game guidance method, a game guidance device, a computer program product, and an electronic device. Background Technology

[0002] Providing in-game guidance can help players make more accurate game operations and enhance their gaming experience.

[0003] In related technologies, a large language model can be used to guide players in the game. The current game state description is input into the large language model, which then determines whether guidance is needed and provides specific guidance content when it is determined that guidance is required.

[0004] However, large language models have high computational overhead, making it difficult to meet real-time requirements. Furthermore, large language models may conjure up reasons for guidance when no guidance is needed, resulting in low accuracy of game guidance.

[0005] It should be noted that the information disclosed in the background section above is only used to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] The purpose of this disclosure is to provide a game guidance method and apparatus, computer program product and electronic device, thereby improving the accuracy and efficiency of game guidance to at least a certain extent.

[0007] Other features and advantages of this disclosure will become apparent from the following detailed description, or may be learned in part from practice of this disclosure.

[0008] According to a first aspect of this disclosure, a game guidance method is provided, comprising: acquiring current feature values ​​of the current game situation characteristics at the current moment; inputting the current feature values ​​into a pre-trained guidance timing prediction model, and predicting, based on the guidance timing prediction model, the probability that a first change value of the combat resources of the combat participants in the current game is greater than a preset threshold after a first preset time period; determining a game guidance timing based on the probability, so as to guide the combat participants in the current game at the game guidance timing.

[0009] According to a second aspect of this disclosure, a game guidance device is provided, comprising: a current feature value acquisition module configured to acquire current feature values ​​of the current game situation characteristics at the current moment; a prediction module configured to input the current feature values ​​into a pre-trained guidance timing prediction model, and predict, based on the guidance timing prediction model, the probability that a first change value of the combat resources of the combat participants in the current game is greater than a preset threshold after a first preset time period; and a guidance timing determination module configured to determine, based on the probability, whether the current moment is a game guidance timing, so as to provide game guidance to the combat participants in the current game at the game guidance timing.

[0010] According to a third aspect of this disclosure, a computer program product comprising instructions is provided that, when run on a computer, causes the computer to perform the steps of the game-guiding method as described in the first aspect.

[0011] According to a fourth aspect of this disclosure, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the game guidance method as described in the first aspect of the above embodiments.

[0012] According to a fifth aspect of this disclosure, an electronic device is provided, comprising: a processor; and a storage device for storing one or more programs, which, when executed by the one or more processors, cause the one or more processors to implement the game guidance method as described in the first aspect of the above embodiments.

[0013] As can be seen from the above technical solutions, the game guidance method, game guidance device, and computer program product and electronic device implementing the game guidance method in the exemplary embodiments of this disclosure have at least the following advantages and positive effects: In some embodiments of the present disclosure, the technical solutions provided include, on the one hand, transforming the complex judgment of game guidance timing into a binary classification probability prediction problem by predicting the change value of battle resources, thereby improving the efficiency of guidance timing judgment through a lightweight guidance timing prediction model, and thus improving the efficiency and timeliness of game guidance; on the other hand, by predicting the change value of battle resources, the timing when a significant change in the game is about to occur can be accurately captured, thereby guiding the game at that timing and improving the accuracy and effectiveness of game guidance.

[0014] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0015] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure. It is obvious that the drawings described below are merely some embodiments of this disclosure, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0016] Figure 1 A schematic diagram of an exemplary system architecture to which embodiments of the present disclosure may be applied is shown; Figure 2 A flowchart illustrating a game guidance method according to an exemplary embodiment of this disclosure is shown. Figure 3 This diagram illustrates a flowchart of a method for determining a pre-trained guidance timing prediction model in an exemplary embodiment of the present disclosure. Figure 4 A flowchart illustrating a method for determining game guidance timing according to an exemplary embodiment of this disclosure is shown. Figure 5 This diagram illustrates a flowchart of a method for initiating game guidance at a game guidance time according to an exemplary embodiment of this disclosure; Figure 6 This diagram illustrates the structure of a game guidance device according to an exemplary embodiment of the present disclosure. Figure 7 A schematic diagram of the structure of an electronic device in an exemplary embodiment of this disclosure is shown. Detailed Implementation

[0017] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided to make this disclosure more comprehensive and complete, and to fully convey the concept of the example embodiments to those skilled in the art. The described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a full understanding of embodiments of this disclosure. However, those skilled in the art will recognize that the technical solutions of this disclosure can be practiced with one or more of the specific details omitted, or other methods, components, apparatus, steps, etc., can be employed. In other instances, well-known technical solutions are not shown or described in detail to avoid obscuring various aspects of this disclosure.

[0018] The terms “a,” “an,” “the,” and “the” are used in this specification to indicate the presence of one or more elements / components / etc.; the terms “including” and “having” are used to indicate an open-ended inclusion and to mean that there may be other elements / components / etc. in addition to the listed elements / components / etc.; the terms “first” and “second” are used only as markings and are not a limitation on the number of objects.

[0019] Furthermore, the accompanying drawings are merely illustrative of this disclosure and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0020] In MOBA (Multiplayer Online Battle Arena) games, the battlefield is complex and ever-changing, often making it difficult for players to grasp the situation. Therefore, providing game guidance during gameplay is crucial for lowering the learning curve and enhancing the user experience.

[0021] In one related technology, a game tutorial dialogue is forcibly triggered at fixed time intervals, such as every 2 minutes. However, this mechanical guidance has low accuracy and is prone to frequently interrupting players during intense battles, affecting their gaming experience.

[0022] Another related technology involves implementing game guidance by setting fixed rule logic. For example, a set of fixed rule logic is preset, such as guiding the player when "the player dies" or "the tower falls". However, rule triggering is difficult to cover the infinitely possible battle situations in MOBA games, and the rules are costly to maintain and lack flexibility, resulting in insufficient timeliness and accuracy of game guidance.

[0023] In another related technology, the current game state description is input as text to an LLM (Large Language Model), which then determines whether intervention for game instruction is necessary. However, LLMs are typically trained on general corpora and lack deep reference to massive amounts of real-world online game data. This makes it difficult for them to make accurate inferences based on complex situational contexts, resulting in a high misjudgment rate. Furthermore, LLMs are prone to conjecturing reasons for game instruction when there are no obvious events, leading to insufficient accuracy in judging the timing of instruction. In addition, MOBA games require high-frequency real-time detection (e.g., detection per second), and LLM inference is costly and has high latency, making it difficult to meet real-time requirements.

[0024] To address the aforementioned problems, this disclosure provides a game guidance method and apparatus, which can be applied to... Figure 1 In the system architecture of the exemplary application environment shown.

[0025] like Figure 1 As shown, the system architecture 100 may include a terminal device 110 and a server 120. The terminal device 110 may be a smartphone, tablet, desktop computer, laptop, smart wearable device, or other similar device. The server 120 generally refers to the backend system providing services related to the game guidance method in this exemplary embodiment, and may be a single server or a cluster of multiple servers. The terminal device 110 and the server 120 can be connected via wired or wireless communication links for data interaction.

[0026] In one exemplary embodiment, the game guidance method described above can be executed by server 120. Correspondingly, a game guidance device can be installed in server 120 to implement the corresponding module functions. For example, when a user plays a game using the game client in terminal device 110, terminal device 110 sends the user's game data to server 120. Server 120 can acquire the current feature value of the current game situation at a fixed time interval, such as every second, and input this current feature value into the guidance timing prediction model. Based on the output of the guidance timing prediction model, server 120 determines whether the current moment is a game guidance opportunity. If it is, game guidance can be provided to the player based on the current game situation; otherwise, the determination can be made again in the next second.

[0027] In one exemplary embodiment, the game guidance method described above can also be jointly executed by the terminal device 110 and the server 120. Accordingly, some modules of the game guidance device can be located in the terminal device 110, and some modules can be located in the server 120 to implement corresponding module functions. For example, when a user plays a game using the game client in the terminal device 110, the game client in the terminal device 110 can be pre-configured with a guidance timing prediction model. During the game, the current feature value of the current game situation at the current moment can be obtained at a fixed time interval, such as every second. This current feature value is input into the guidance timing prediction model. Based on the output of the guidance timing prediction model, it is determined whether the current moment is a game guidance opportunity. If so, the current game situation information can be sent to the server 120, which then determines the specific game guidance content and sends it to the player's client.

[0028] It should be understood that Figure 1 The number of terminal devices and servers shown is merely illustrative. Depending on implementation needs, there can be any number of terminal devices and servers. For example, server 120 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms.

[0029] In one embodiment of this disclosure, the game guidance method can run on a local terminal device or a server. When the game interaction method runs on a server, the method can be implemented and executed based on a cloud interaction system, wherein the cloud interaction system includes a server and a client device.

[0030] In an optional implementation, various cloud applications, such as cloud gaming, can run under the cloud interaction system. Taking cloud gaming as an example, cloud gaming refers to a gaming method based on cloud computing. In the cloud gaming operating mode, the game program and the game screen presentation are separated. The storage and execution of the game interaction methods are completed on the cloud gaming server. The client device is used for receiving and sending data and presenting the game screen. For example, the client device can be a display device with data transmission capabilities located close to the user, such as a mobile terminal, television, computer, or PDA; however, the information processing is performed by the cloud gaming server in the cloud. When playing the game, the player operates the client device to send operation commands to the cloud gaming server. The cloud gaming server runs the game according to the operation commands, encodes and compresses the game screen and other data, returns it to the client device via the network, and finally, the client device decodes and outputs the game screen.

[0031] In an alternative implementation, taking a game as an example, the local terminal device stores the game program and is used to display the game screen. The local terminal device is used to interact with the player through a graphical user interface (GUI), i.e., conventionally by downloading, installing, and running the game program via an electronic device. The local terminal device can provide the GUI to the player in various ways, such as rendering it on the terminal's display screen or providing it to the player via holographic projection. For example, the local terminal device may include a display screen for displaying the GUI, which includes game screens, and a processor for running the game, generating the GUI, and controlling the display of the GUI on the display screen.

[0032] However, those skilled in the art will readily understand that the above application scenarios are merely illustrative and are not intended to limit the scope of this exemplary embodiment.

[0033] In one possible implementation, embodiments of the present invention provide a game guidance method that provides a graphical user interface (GUI) via a terminal device, the GUI including at least a portion of the game scene. The terminal device can be either the aforementioned local terminal device or a client device within the aforementioned cloud interaction system.

[0034] Figure 2 This diagram illustrates a flowchart of a game guidance method according to an exemplary embodiment of this disclosure, with reference to... Figure 2 The method includes: Step S210: Obtain the current feature value of the game situation characteristics at the current moment; Step S220: Input the current feature value into the pre-trained guidance timing prediction model, and predict the probability that the first change value of the battle resources of the battle participants in the current game is greater than a preset threshold after a first preset time period based on the guidance timing prediction model. Step S230: Determine the game guidance timing based on the probability, and guide the players in the current game at the game guidance timing.

[0035] exist Figure 2 In the technical solution provided by the embodiment shown, on the one hand, the complex judgment of game guidance timing is transformed into a binary classification probability prediction problem by predicting the change value of battle resources. Thus, the guidance timing is judged by a lightweight guidance timing prediction model, improving the efficiency of guidance timing judgment and thereby improving the efficiency and timeliness of game guidance. On the other hand, by predicting the change value of battle resources, the timing when the game is about to undergo significant changes can be accurately captured, thereby guiding the game at that time and improving the accuracy and effectiveness of game guidance.

[0036] The following is a detailed explanation of the specific implementation method of "Step S210, obtaining the current feature value of the game situation characteristics at the current moment".

[0037] In one exemplary implementation, the game situation characteristics can be understood as features that can objectively reflect the state and trend of the game during the game, such as feature data that can reflect the current strength of the players, the real-time battlefield environment, resource distribution, skill status and the occurrence of key events, which change dynamically in real time with the progress of the game.

[0038] In one exemplary embodiment, the game situation features include game combat data of the game participants, which includes one or more of the following: the total value of virtual economic resources of the participants, the number of defensive buildings, the cumulative number of enemy targets defeated by the participants, the position coordinates of the virtual characters controlled by the participants, the remaining cooldown time of preset skills, the survival status of preset non-player controlled characters, the virtual health of the participants, the difference in virtual economic resources between different participants, and the difference in virtual economic resources between different participants whose controlled virtual characters have the same function.

[0039] For example, game situation characteristics may include the combat resources of the combat participants (such as gold and experience points), character attribute status (such as level and virtual health), skill status (such as the remaining cooldown time of key skills), battlefield environment status (such as the number of defensive buildings and the survival status of neutral monsters), combat behavior data (such as the number of kills and assists of both sides), and character position information, etc., which represent the game state. Of course, game situation characteristics may include other content, such as the location coordinates of the core heroes of both sides, the survival status and health of epic monsters, etc., and this exemplary implementation does not impose any special limitations on this.

[0040] It should be noted that the game situation characteristics can include raw game combat data that can be directly read from the game engine interface without any processing, such as current virtual health, current amount of gold coins, and current level of the game account. It can also include game combat data of these elements that have been preprocessed, such as introducing time dimension statistical trends or spatial dimension calculation relative relationships to obtain derived features, such as the difference in gold coins between the two sides, the distance between the two sides, and the change in the difference in gold coins between the two sides over a period of time.

[0041] In one exemplary implementation, game situation features can be selected from game battle data based on the correlation between game battle data and changes in battle resources. For example, the top N candidate game battle data points ranked by their correlation with changes in battle resources can be selected as game situation features from all candidate game battle data points.

[0042] In another exemplary implementation, the characteristics of the game situation can also be determined according to needs or experience, and this exemplary implementation does not impose any special limitations on this.

[0043] For example, during the game, the current feature value of the game situation at the current moment can be obtained at a preset time interval, such as obtaining the current feature value of the game situation at the current moment every second.

[0044] The preset time interval can be customized according to actual needs. The smaller the preset time interval, the more frequently the game guidance timing is judged, which will make it easier to capture the accurate game guidance timing. The larger the preset time interval, the less frequently the game guidance timing is judged, which may miss the best game guidance timing.

[0045] The following is a detailed description of the specific implementation of "step S220, inputting the current feature value into the pre-trained guidance timing prediction model, and predicting the probability that the first change value of the battle resources of the current game participants is greater than a preset threshold after a first preset time period based on the guidance timing prediction model".

[0046] In one exemplary implementation, the change value includes one or more of the change value of the sum of the battle resources of each battle participant and the change value of the battle resources of any battle participant; the battle resources include one or more of virtual economic resources and virtual experience points.

[0047] Taking virtual economic resources as an example, the guidance timing prediction model can predict whether the change in the sum of the virtual economic resources of both sides exceeds a preset threshold, or whether the change in the virtual economic resources of any single side exceeds a preset threshold. In other words, whether a major game event will occur in the future can be measured by whether the sum of the virtual economic resources of both sides changes significantly in a future period, thereby determining whether game guidance is currently needed. Similarly, whether a major game event will occur in the future can be measured by whether the virtual economic resources of any single side change significantly, thereby determining whether game guidance is currently needed.

[0048] For example, Figure 3 This diagram illustrates a flowchart of a method for determining a pre-trained guidance timing prediction model according to an exemplary embodiment of this disclosure. (Reference) Figure 3 The method may include steps S310 to S350. Wherein: In step S310, historical feature values ​​of the game situation characteristics at each sampling time of the historical game are obtained to obtain sample feature data.

[0049] For example, a large amount of historical game replay data can be extracted from the game server logs. Then, according to the preset time interval mentioned above, such as sampling each game every second, the historical feature values ​​of the game situation characteristics at that sampling time of the historical game can be extracted for the sampling time T, thereby obtaining a large amount of sample feature data.

[0050] For example, if there are 100 historical game matches, and 100 data points are collected from each of these 100 historical game matches, then a total of 10,000 sample feature data points can be collected.

[0051] In step S320, for each sample feature data, a second change value of the battle resources of the battle participants in the historical game game is determined from the historical game game after a first preset time period at the sampling time, and a sample label corresponding to the sample feature data is generated according to whether the second change value is greater than a preset threshold.

[0052] For example, the first preset duration can be customized based on needs or experience. However, the first preset duration should not be set too long or too short. If it is set too long, it will be too far removed from the current moment, with too many variables, thus negating the purpose of game guidance. If it is set too short, it will be too close to the current moment, and players may not have enough time to change the situation based on the guidance information. Therefore, the first preset duration can be a duration that is not too far removed from the current moment and allows players enough time to change the situation based on the game guidance information.

[0053] Taking a first preset duration of 10 seconds as an example, for the sample feature data corresponding to each sampling moment, the change value of the virtual economic resources of both sides in the corresponding historical game match 10 seconds after the sampling moment compared to the sampling moment can be collected. If the change value is greater than the preset threshold, the sample label is set to 1; otherwise, the sample label is set to 0.

[0054] Among them, a sample label of 1 indicates that the economy of both sides will change significantly in the future, indicating that a major event such as an explosive team battle or resource competition is about to occur. A sample label of 0 indicates that the sum of the economies will fluctuate relatively steadily in the short term, and no significant events will occur.

[0055] For example, the preset threshold can be customized according to needs or experience. Taking gold coins as the battle resource, if it is determined from experience or statistical analysis of historical data that an important game event usually occurs when the change in the sum of gold coins is greater than 2000, then the preset threshold can be set to 2000.

[0056] Through the above steps S310 to S320, the automatic generation of sample feature data and sample labels can be achieved, thereby helping to improve the training efficiency of the timing prediction model and reduce the training cost of the model.

[0057] In step S330, the sample feature data is input into an initial machine learning model, and the predicted value of the change in the battle resources of the participants in the historical game is obtained according to the output of the machine learning model after a first preset time period has elapsed at the sampling time indicated by the sample feature data.

[0058] For example, after obtaining sample feature data and corresponding sample labels, an initial machine learning model can be trained based on the sample feature data and sample labels, and a pre-trained guidance timing prediction model can be obtained based on the training results.

[0059] In one exemplary implementation, the initial machine learning model may include any machine learning model capable of binary classification, such as XGBoost (Extreme Gradient Boosting), LightGBM (Light Gradient Boosting Machine), or a lightweight MLP (Multi-Layer Perceptron) neural network.

[0060] In this disclosure, determining the timing of game guidance is transformed into a binary classification problem. The decision to guide the game is made by predicting whether the situation will change drastically in the short term. In other words, the guidance timing prediction model in this disclosure only needs to determine whether the current moment is a suitable time for game guidance, i.e., it only requires a binary classification judgment. Therefore, the guidance timing prediction model in this disclosure is a lightweight binary classification learning model. This reduces the computational cost of the model, thereby helping to improve the efficiency of determining the game guidance timing and the efficiency of game guidance.

[0061] For example, sample feature data can be batch-input into an initial machine learning model. The output of the machine learning model can be used to obtain the predicted value of the change in the combat resources of the participants in the historical game after a first preset time period following the sampling time indicated by the sample feature data. For example, the predicted value of the change in the sum of the virtual economic resources of the participants in the game at the time corresponding to 10 seconds after the sampling time indicated by the sample feature data can be obtained.

[0062] In step S340, the training loss is determined based on the degree of difference between the predicted value of the change and the sample label.

[0063] For example, the training loss can be determined by the cross-entropy loss or hinge loss between the predicted value of the change value and the sample label.

[0064] In step S350, the machine learning model is iteratively trained according to the training loss to obtain the pre-trained guidance timing prediction model.

[0065] For example, the machine learning model can be iteratively trained with the goal of minimizing the training loss until the training loss is less than a preset loss value or the number of iterations reaches a preset number, at which point training stops, thus obtaining a candidate guidance timing prediction model. Then, the candidate guidance timing prediction model can be tested using performance metrics such as prediction accuracy and recall. If the test is passed, the candidate guidance timing prediction model that passes the test is designated as the pre-trained guidance timing prediction model. If the test fails, the training and test datasets are re-split, and the machine learning model is trained again until the test is passed.

[0066] Through steps S210 to S250 described above, a pre-trained guidance timing prediction model can be obtained. Then, the acquired current feature values ​​can be input into the pre-trained guidance timing prediction model. The pre-trained guidance timing prediction model can predict, based on the input current feature values, the probability that the first change value of the combat resources of the combat participants in the current game match will be greater than a preset threshold after a first preset time period at the current moment.

[0067] The following is a detailed description of the specific implementation of "step S230, determining the game guidance timing based on the probability, and guiding the players in the current game match at the game guidance timing".

[0068] For example, Figure 4 This diagram illustrates a flowchart of a method for determining game initiation timing according to an exemplary embodiment of this disclosure. (Reference) Figure 4 The method may include steps S410 to S460. Wherein: In step S410, it is determined whether the probability is greater than the first preset probability threshold. If it is, proceed to step S420; otherwise, proceed to step S460.

[0069] In one exemplary implementation, the first preset probability threshold can be customized based on experience or needs. For example, the first preset probability threshold can be any value greater than or equal to 0.5 and less than 1. The larger the first preset probability threshold is set, the higher the reliability of the final determined game guidance timing; the smaller the first preset probability threshold is set, the lower the reliability of the final determined game guidance timing.

[0070] For example, if the probability is greater than a first preset probability threshold, the current moment can be directly determined as a game-guiding opportunity. Alternatively, subsequent steps S420 to S460 can be used to further determine whether the current moment is a game-guiding opportunity. If the probability is less than or equal to the first preset probability threshold, it indicates that the possibility of a major event occurring in the future is low, so the current moment can be determined as not a game-guiding opportunity.

[0071] In step S420, it is determined whether the duration of the current time relative to the most recent game guidance time is greater than the second preset duration. If not, proceed to step S430; if yes, proceed to step S440.

[0072] For example, the time elapsed since the last game tutorial can be calculated, and it can be determined whether this time elapsed is greater than a second preset time elapsed.

[0073] The second preset duration can be customized according to needs or experience. For example, the second preset duration can be 30 seconds. The second preset duration can be understood as the pre-set guide cooldown time, that is, the minimum time interval between two adjacent game guide moments.

[0074] For example, during the cooldown period, users may not be guided, avoiding frequent guidance that could disturb them. In other words, if the current moment does not fall within the cooldown period, the method described in this disclosure can be used to determine whether the current moment is a suitable time for game guidance; if the current moment does fall within the cooldown period, it can be directly determined that the current moment is not a suitable time for game guidance. Alternatively, the method can periodically determine whether the current moment is a candidate time for game guidance based on a fixed time interval. That is, if the probability is greater than a first preset probability threshold, the current moment is determined to be a candidate time for game guidance, and then it is determined whether the current moment is within the cooldown period. If it is, the current moment is determined not to be a suitable time for game guidance; otherwise, the current moment is determined to be a suitable time for game guidance.

[0075] For example, during the cooldown period, it can also be determined whether the current moment is the time for game guidance based on whether a first preset game event occurs or whether the probability is greater than a second preset probability threshold.

[0076] In step S430, it is determined whether the probability is greater than the second probability threshold or whether the current game event is the first preset game event. If so, proceed to step S440; otherwise, proceed to step S450.

[0077] In one exemplary implementation, the first preset probability threshold is less than the second preset probability threshold.

[0078] For example, if the probability is greater than the second preset probability threshold, it indicates a high likelihood of a major event occurring in the future. In this case, the current moment, regardless of whether it is within the guidance cooldown period, can be identified as a game guidance opportunity, and thus game guidance can proceed. Alternatively, if the probability is greater than the first preset probability threshold and the first preset game event is currently occurring, the current moment, regardless of whether it is within the guidance cooldown period, can also be identified as a game guidance opportunity.

[0079] The second preset probability threshold and the first preset game event can be customized according to needs or experience. For example, the first preset game event can be a major game event such as a BOSS team battle. This exemplary implementation does not impose any special limitations on this.

[0080] In step S440, the current moment is determined as the game guidance time.

[0081] For example, if the probability is greater than the first probability threshold and the time elapsed since the last game tutorial is greater than the second preset time, it means that the current moment is not within the tutorial cooldown period and the probability is greater than the first probability threshold. Therefore, the current moment can be directly determined as the game tutorial time.

[0082] In step S450, it is determined that the current moment is not a game-guiding opportunity.

[0083] For example, if the time elapsed since the last game tutorial is less than or equal to the second preset time, it means that the current time is within the tutorial cooldown period. It can be directly determined that the current time is not a game tutorial opportunity. Alternatively, if the first preset game event has not occurred at the current time or the probability is less than or equal to the second probability threshold, it can be determined that the current time is not a game tutorial opportunity.

[0084] Through steps S410 to S460 above, the timing of game guidance can be determined based on the probability of significant economic fluctuations in the future predicted by the model, thereby improving the efficiency and accuracy of determining the timing of game guidance. Furthermore, with a guidance cooldown time set, the interference caused to players by continuous and frequent guidance can be avoided, thus enhancing the player's game guidance experience.

[0085] For example, Figure 5 This diagram illustrates a flowchart of a method for initiating game guidance at a game guidance time according to an exemplary embodiment of this disclosure, with reference to... Figure 5 The method may include steps S510 to S550. Wherein: In step S510, the game event of the current game match at the current guiding moment indicated by the game guiding timing is determined.

[0086] For example, the game event corresponding to the current tutorial moment can be determined using battlefield environment data at that moment. This could be achieved by using information such as turret health and status, jungle monster survival status, minion wave position and health at the current tutorial moment. Of course, other information can also be used to determine the game event at the current tutorial moment; this exemplary implementation does not impose any special limitations on this.

[0087] In step S520, the game event is matched with the second preset game event in the preset guide content library.

[0088] For example, a preset tutorial content library can be set in advance, which can be configured with a second preset game event and its corresponding game tutorial content. The second preset game event and its corresponding game tutorial content in the preset tutorial content library can be customized based on experience or needs, and this exemplary embodiment does not impose any special limitations on this.

[0089] In other words, the guidance content can be understood as a pre-configured knowledge base mapping "game events - corresponding tactics". It can monitor key discrete events in the game process in real time (e.g., the enemy disappears from sight, our team gathers to attack the dragon, a player uses a key skill, etc.). Based on the monitored events, it queries this preset knowledge base to find corresponding tactical suggestions or operation prompts, generates text and pushes it to the player, thereby guiding the player in the game.

[0090] Each second preset game event can correspond to one or more preset game tutorial content. That is, the same second preset game event can correspond to multiple preset game tutorial content.

[0091] For example, game events and second preset game events can be encoded using a text model to obtain corresponding text encoding features. Then, the similarity between the corresponding text encoding features is calculated. The similarity is used to determine whether a match is successful. If the similarity is greater than the similarity threshold, the match is successful; otherwise, the match fails. The similarity threshold can also be customized according to requirements or practical experience. This exemplary implementation does not impose any special limitations on this.

[0092] Of course, character matching can also be used, such as performing regular expression matching on the text corresponding to the game event and the text corresponding to the second preset game event, and determining whether the match is successful based on the regular expression matching result.

[0093] In step S530, it is determined whether the match is successful. If successful, proceed to step S540; otherwise, proceed to step S550.

[0094] In step S540, the preset game guidance content corresponding to the successfully matched second preset game event is determined from the preset guidance content library, and game guidance is provided to the participants in the current game based on the preset game guidance content.

[0095] For example, if a match is successfully matched, the preset game tutorial content corresponding to the second preset game event can be directly determined as the current game tutorial content, and then pushed to the corresponding game client to guide the player. If there are multiple preset game tutorial contents corresponding to the second preset game event, one preset game tutorial content can be randomly selected as the current game tutorial content for the current game match.

[0096] In step S550, the current situation information of the current game match at the current guidance time is obtained, and the current situation information is input into the pre-trained game guidance content generation model. The current game guidance content is determined according to the output of the game guidance content generation model, so as to guide the players in the current game match based on the current game guidance content.

[0097] For example, in the event of a match failure, the current situation information corresponding to the current guidance moment can be fed into a pre-trained game guidance content generation model, and the current game guidance content can be obtained through the output of this model.

[0098] In one exemplary implementation, the current situation information can be customized as needed, such as one or more of the following: current operational behavior data (e.g., skill release, target selection, movement trajectory), character status data (e.g., health, equipment, coordinates, field of vision), battlefield environment data (e.g., turret health, status), and battle result data (e.g., economic difference, number of kills, number of assists, etc.).

[0099] For example, a pre-trained game tutorial content generation model can include a pre-trained large language model, which can be used to generate game tutorial content.

[0100] Through steps S510 to S550 above, after determining the game guidance timing, the LLM and the preset tactics corresponding to the locally preset second preset game event can be combined. If game guidance can be performed using the locally set preset tactics, there is no need to call the large language model. Only when game guidance cannot be performed using the locally set preset tactics will the large language model be called. In this way, accurate generation of game guidance content can be achieved while reducing dependence on the large language model, thereby improving overall guidance efficiency.

[0101] Of course, detailed current situation information can also be sent directly to the cloud-based LLM to request the generation of specific tutorial text. Since this disclosure uses a large language model to generate specific tutorial text only when the current moment is determined to be the game-guiding opportunity, it can minimize the computational overhead of the large language model and improve game-guiding efficiency.

[0102] In one exemplary implementation, after determining the timing for game guidance, game guidance can be provided to both sides of the battle, or it can be provided only to one side, such as the side with less gold. Alternatively, game guidance can be provided to all members of the battle, or only to certain specific members of the battle, such as the game account corresponding to a core hero, or to the member with the least gold. This exemplary implementation does not impose any particular limitations on these methods.

[0103] For example, when game players play games in the game client through their game accounts, the game guidance provided to game players in this disclosure can be understood as guiding the game players corresponding to their game accounts.

[0104] In this disclosure, the game guidance timing prediction model is a lightweight model. Its parameter size is orders of magnitude smaller than that of large language models, thus requiring minimal computing power to achieve real-time guidance timing monitoring down to the second, ensuring real-time guidance while reducing server load. Furthermore, this game guidance timing prediction model is trained on a large amount of real-world game data, enabling it to keenly capture early signs of battlefield changes that are difficult for humans to perceive (such as positioning characteristics before a dramatic shift in economic advantage), thereby improving the accuracy of game guidance timing judgment. Moreover, through this game guidance timing prediction model, accurate intervention can be made at crucial moments when the battlefield is about to undergo significant changes, improving the practicality of the tutorial information and player acceptance, thereby enhancing the player's game guidance experience.

[0105] Furthermore, it should be noted that the above figures are merely illustrative representations of the processes included in the method according to exemplary embodiments of the present invention, and are not intended to be limiting. It is readily understood that the processes shown in the above figures do not indicate or limit the temporal order of these processes. Additionally, it is readily understood that these processes may be executed synchronously or asynchronously, for example, in multiple modules.

[0106] Furthermore, exemplary embodiments of this disclosure also provide a game guiding device. (See reference...) Figure 6 As shown, the game guidance device 600 includes the following program modules: a current feature value acquisition module 610, configured to acquire the current feature value of the game situation characteristics of the current game at the current moment; a prediction module 620, configured to input the current feature value into a pre-trained guidance timing prediction model, and predict, based on the guidance timing prediction model, the probability that the first change value of the combat resources of the combat participants in the current game is greater than a preset threshold after a first preset time; and a guidance timing determination module 630, configured to determine whether the current moment is a game guidance timing based on the probability, so as to provide game guidance to the combat participants in the current game at the game guidance timing.

[0107] In one exemplary embodiment, the determination of the pre-trained guidance timing prediction model includes: acquiring historical feature values ​​of the game situation characteristics at each sampling time in historical game matches to obtain sample feature data; for each sample feature data, determining a second change value of the combat resources of the combat participants in the historical game matches after a first preset time has elapsed at the sampling time, and generating a sample label corresponding to the sample feature data based on whether the second change value is greater than a preset threshold; inputting the sample feature data into an initial machine learning model, and obtaining a predicted value of the change value of the combat resources of the combat participants in the historical game matches after a first preset time has elapsed at the sampling time indicated by the sample feature data based on the output of the machine learning model; determining a training loss based on the degree of difference between the predicted value of the change value and the sample label; and iteratively training the machine learning model based on the training loss to obtain the pre-trained guidance timing prediction model.

[0108] In one exemplary embodiment, the game situation characteristics include game combat data of the game participants, which includes one or more of the following: the total value of virtual economic resources of the participants, the number of defensive buildings, the cumulative number of enemy targets defeated by the participants, the position coordinates of the virtual characters controlled by the participants, the remaining cooldown time of preset skills, the survival status of preset non-player controlled characters, the virtual health of the participants, the difference in virtual economic resources between different participants, and the difference in virtual economic resources between different participants whose controlled virtual characters have the same function.

[0109] In one exemplary implementation, determining the game guidance timing based on the probability includes: determining the duration of the guidance time corresponding to the most recent game guidance timing from the current moment; and determining the current moment as a game guidance timing if the probability is greater than a first preset probability threshold and the duration is greater than or equal to a second preset duration; wherein, if the duration is less than the second preset duration, determining the current moment as a game guidance timing if the probability is greater than the first preset probability threshold and the current game event of the current game match is detected as a first preset game event or the probability is greater than the second preset probability threshold, wherein the first preset probability threshold is less than the second preset probability threshold.

[0110] In one exemplary implementation, the step of providing game guidance to the participants in the current game match at the game guidance timing includes: determining the game event of the current game match at the current guidance time indicated by the game guidance timing; matching the game event with a second preset game event in a preset guidance content library; and providing game guidance to the participants in the current game match based on the matching result.

[0111] In one exemplary implementation, the step of providing game guidance to the participants in the current game match based on the matching result includes: if the match is successful, determining the preset game guidance content corresponding to the second preset game event that was successfully matched from the preset guidance content library, and providing game guidance to the participants in the current game match based on the preset game guidance content; if the match fails, obtaining the current situation information of the current game match at the current guidance time, inputting the current situation information into a pre-trained game guidance content generation model, determining the current game guidance content based on the output of the game guidance content generation model, and providing game guidance to the participants in the current game match based on the current game guidance content.

[0112] In one exemplary implementation, any change value includes one or more of the change value of the sum of the battle resources of each battle participant and the change value of the battle resources of any battle participant; the battle resources include one or more of virtual economic resources and virtual experience points.

[0113] The specific details of each part of the above-mentioned device have been described in detail in the method section of the implementation plan. For any undisclosed details, please refer to the implementation plan of the method section, and therefore will not be repeated here.

[0114] It should be noted that although several modules or units for the device used to perform actions have been mentioned in the detailed description above, this division is not mandatory. In fact, according to exemplary embodiments of this disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units.

[0115] Furthermore, although the steps of the method in this disclosure are described in a specific order in the accompanying drawings, this does not require or imply that the steps must be performed in that specific order, or that all the steps shown must be performed to achieve the desired result. Additional or alternative steps may be omitted, multiple steps may be combined into one step, and / or a step may be broken down into multiple steps.

[0116] Exemplary embodiments of this disclosure also provide a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the above-described game guidance method.

[0117] In one implementation, the computer program product can be a tangible product containing a computer program, such as a computer-readable storage medium storing the computer program. The readable storage medium can be a storage medium based on electrical, magnetic, optical, electromagnetic, infrared, or other signals, including but not limited to: random access memory (RAM), read-only memory (ROM), magnetic tape, floppy disk, flash memory, hard disk drive (HDD), solid-state drive (SSD), etc. For example, the computer program product can be implemented as a non-volatile storage medium storing a computer program, such as read-only memory, NAND flash memory, etc.

[0118] In one implementation, the computer program product can be an intangible product containing a computer program. For example, the computer program product can be implemented as a virtual digital product, such as an executable file, installation package, or other digital file storing the computer program.

[0119] Computer program code can be written in one or more programming languages. Examples of programming languages ​​include C, Java, C++, and Python. Program code can execute entirely on the user's computing device, partially on the user's computing device, or as a standalone software package. It can also execute partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, such as a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via an internet connection provided by a mobile network operator).

[0120] Computer programs can be carried or transmitted via signals such as electricity, magnetism, light, electromagnetic fields, and infrared radiation. Electronic devices can convert signals carrying computer programs into digital signals, thereby running the computer programs. When a computer program runs on an electronic device, its code is used to cause the electronic device to execute (more specifically, the processor of the electronic device to execute) the method steps of various exemplary embodiments of this disclosure. For example, the game guidance method described above can be executed, which includes the following steps: obtaining the current feature value of the game situation characteristics of the current game at the current moment; inputting the current feature value into a pre-trained guidance timing prediction model, and predicting, based on the guidance timing prediction model, the probability that the first change value of the combat resources of the combat participants in the current game is greater than a preset threshold after a first preset time period; determining the game guidance timing according to the probability, so as to guide the combat participants in the current game at the game guidance timing.

[0121] By executing the above method steps through a computer program, on the one hand, the complex judgment of game guidance timing is transformed into a binary classification probability prediction problem by predicting changes in combat resources. This allows for the use of a lightweight guidance timing prediction model, improving the efficiency of guidance timing judgment and thus enhancing the efficiency and timeliness of game guidance. On the other hand, predicting changes in combat resources can accurately capture moments when significant changes in the game are about to occur, allowing for game guidance at those moments, thus improving the accuracy and effectiveness of game guidance. (Method options can be added here based on client preferences.)

[0122] Exemplary embodiments of this disclosure also provide an electronic device, such as the terminal device 110 or server 120 described above. The electronic device may include a processor and a memory. The memory stores executable instructions for the processor, such as computer programs. The processor executes these executable instructions to perform the method steps of various exemplary embodiments of this disclosure. Furthermore, the electronic device may also include a display for displaying a graphical user interface.

[0123] The following is for reference. Figure 7 The electronic device is illustrated by way of a general-purpose computing device. It should be understood that... Figure 7 The electronic device 700 shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments disclosed herein.

[0124] like Figure 7 As shown, the electronic device 700 may include: a processor 710, a memory 720, a bus 730, an I / O (input / output) interface 740, a network adapter 750, and a display 760.

[0125] The memory 720 may include volatile memory, such as RAM 721 and cache unit 722, and may also include non-volatile memory, such as ROM 723. The memory 720 may also include one or more program modules 724, including but not limited to: an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include an implementation of a network environment. For example, program module 724 may include the modules described above.

[0126] The processor 710 may include one or more processing units, such as an AP (Application Processor), a modem processor, a GPU (Graphics Processing Unit), an ISP (Image Signal Processor), a controller, an encoder, a decoder, a DSP (Digital Signal Processor), a baseband processor, and / or an NPU (Neural-Network Processing Unit).

[0127] The processor 710 can be used to execute executable instructions stored in the memory 720, such as the game guidance method described above, which includes the following steps: obtaining the current feature value of the current game situation at the current moment; inputting the current feature value into a pre-trained guidance timing prediction model, and predicting, based on the guidance timing prediction model, the probability that the first change value of the combat resources of the combat participants in the current game is greater than a preset threshold after a first preset time; determining the game guidance timing according to the probability, so as to guide the combat participants in the current game at the game guidance timing.

[0128] Implementing the above method through computer programs achieves two main benefits. First, by predicting changes in combat resources, the complex judgment of game guidance timing is transformed into a binary classification probability prediction problem. This allows for the use of a lightweight guidance timing prediction model, improving the efficiency of guidance timing judgment and thus enhancing the efficiency and timeliness of game guidance. Second, by predicting changes in combat resources, the timing of significant changes in the game can be accurately captured, allowing for game guidance at that moment and improving the accuracy and effectiveness of game guidance.

[0129] Bus 730 is used to connect different components of electronic device 700 and may include a data bus, an address bus and a control bus.

[0130] Electronic device 700 can communicate with one or more external devices 800 (such as keyboard, mouse, external controller, etc.) through I / O interface 740.

[0131] Electronic device 700 can communicate with one or more networks via network adapter 750. For example, network adapter 750 can provide mobile communication solutions such as 3G / 4G / 5G, or wireless communication solutions such as wireless LAN, Bluetooth, and near-field communication. Network adapter 750 can communicate with other modules of electronic device 700 via bus 730.

[0132] Electronic device 700 can display a graphical user interface, such as a game guide interface, through monitor 760.

[0133] although Figure 7 As not shown in the diagram, other hardware and / or software modules may also be configured in the electronic device 700, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0134] Those skilled in the art will understand that various aspects of this disclosure can be implemented in the following forms: a completely hardware implementation, a completely software implementation (including firmware, microcode, etc.), or a combination of hardware and software implementations, which may be referred to as "circuit", "module" or "system" respectively.

[0135] It should be understood that this disclosure is not limited to the specific methods, steps, or structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. Those skilled in the art will readily conceive of other embodiments based on the specific implementations provided in this disclosure. Therefore, the specific implementations provided in this disclosure are merely exemplary, and the scope and spirit of this disclosure are indicated by the claims, and should cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary technical means in the art not disclosed in this disclosure.

Claims

1. A game guidance method, characterized in that, include: Obtain the current feature value of the current game situation at the current moment; The current feature value is input into a pre-trained guidance timing prediction model. Based on the guidance timing prediction model, the probability that the first change value of the battle resources of the battle participants in the current game match is greater than a preset threshold is predicted after a first preset time. Based on the probability, a game guidance timing is determined, and game guidance is provided to the participants in the current game match at the game guidance timing.

2. The method according to claim 1, characterized in that, The methods for determining the pre-trained guidance timing prediction model include: Obtain historical feature values ​​of the game situation characteristics at each sampling time in historical game matches to obtain sample feature data; For each sample feature data, from the historical game matches, determine the second change value of the battle resources of the battle participants in the historical game matches after a first preset time period at the sampling time, and generate the sample label corresponding to the sample feature data according to whether the second change value is greater than a preset threshold. The sample feature data is input into an initial machine learning model, and the predicted value of the change in the battle resources of the participants in the historical game is obtained based on the output of the machine learning model after a first preset time period has elapsed at the sampling time indicated by the sample feature data. The training loss is determined based on the degree of difference between the predicted value of the change and the sample label; The machine learning model is iteratively trained based on the training loss to obtain the pre-trained guidance timing prediction model.

3. The method according to claim 1 or 2, characterized in that, The game situation characteristics include game combat data of the participating parties in the game, which includes one or more of the following: the total value of virtual economic resources of the participating parties, the number of defensive buildings, the cumulative number of enemy targets defeated by the participating parties, the position coordinates of the virtual characters controlled by the participating parties, the remaining cooldown time of preset skills, the survival status of preset non-player controlled characters, the virtual health of the participating parties, the difference in virtual economic resources between different participating parties, and the difference in virtual economic resources between different participating parties whose controlled virtual characters have the same function.

4. The method according to claim 1, characterized in that, Determining the game guidance timing based on the probability includes: Determine the duration of the current moment since the last game tutorial. If the probability is greater than a first preset probability threshold and the duration is greater than or equal to a second preset duration, the current moment is determined to be the game guidance moment; Wherein, if the duration is less than the second preset duration, and the probability is greater than the first preset probability threshold and the current game event of the current game is detected to be the first preset game event or the probability is greater than the second preset probability threshold, the current moment is determined to be the game guidance moment, wherein the first preset probability threshold is less than the second preset probability threshold.

5. The method according to claim 1, characterized in that, The process of providing game guidance to the participants in the current game match at the specified game guidance time includes: Determine the game events of the current game match at the current tutorial moment indicated by the game tutorial timing; The game event is matched with the second preset game event in the preset guidance content library, and game guidance is provided to the participants in the current game match based on the matching result.

6. The method according to claim 5, characterized in that, The step of providing game guidance to the participants in the current game match based on the matching results includes: If a match is successful, the preset game guidance content corresponding to the second preset game event that was successfully matched is determined from the preset guidance content library, and game guidance is provided to the participants in the current game based on the preset game guidance content. In the event of a match failure, the current situation information of the current game match at the current guidance time is obtained, and the current situation information is input into a pre-trained game guidance content generation model. The current game guidance content is determined based on the output of the game guidance content generation model, so as to guide the players in the current game match based on the current game guidance content.

7. The method according to claim 1 or 2, characterized in that, Any change value includes one or more of the change value of the sum of the battle resources of all battle participants and the change value of the battle resources of any battle participant; the battle resources include one or more of virtual economic resources and virtual experience points.

8. A game guidance device, characterized in that, include: The current feature value acquisition module is configured to acquire the current feature value of the game situation characteristics at the current moment; The prediction module is configured to input the current feature value into a pre-trained guidance timing prediction model, and based on the guidance timing prediction model, predict the probability that the first change value of the battle resources of the battle participants in the current game match is greater than a preset threshold after a first preset time. The guidance timing determination module is configured to determine, based on the probability, whether the current moment is a game guidance timing, so as to guide the players in the current game match at the game guidance timing.

9. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method described in any one of claims 1 to 7.

10. An electronic device, characterized in that, include: One or more processors; A storage device for storing one or more programs, which, when executed by one or more processors, cause the one or more processors to implement the method as described in any one of claims 1 to 7.

Images