Video Game Inventory Coach

Abstract

The video game inventory coach model AI model (e.g., 600, 708) can suggest discarding, acquiring, and / or using items in the video game inventory based on the user's gameplay style and other players' clusters (210). The inventory coach model can be turned on and off at the target player's discretion (1202).

Description

Technical Field

[0001] This application relates to a technically inventive non - stereotypical solution that necessarily results from computer technology and brings about specific technical improvements.

Background Art

[0002] As understood herein, video game players are often limited to a specific number of virtual objects that can be controlled at any point during the execution of a video game. At that time, other objects that could potentially be used or acquired may be missing. Similarly, as understood herein, a player may not particularly use various objects they have obtained, but based on the prospect of potential future use, they may hold onto those objects even if they ultimately do not use them. This eventually leads to inefficient game play, the retention of unused objects at the expense of other objects that could be used, repetitive or unnecessary actions, and an overall sub - optimal experience, especially for novice players. As prior art documents, there is U.S. Patent Application Publication No. 2020 / 0269136.

Summary of the Invention

[0003] Accordingly, a video game inventory coaching artificial intelligence (AI) model can be trained through machine learning to propose discarding, acquiring, and / or using items in a video game inventory based on a user's game play style, clusters of other players, and the points in the game the user has reached.

[0004] Thus, in one aspect, the device includes at least one computer storage that is not a temporary signal and that includes instructions executable by at least one processor to analyze a video game player's play style in relation to a particular video game being played by the video game player. The instructions are also executable to propose discarding, obtaining, and / or using at least one video game asset based on the analysis.

[0005] In some examples, the instructions can be made executable to propose discarding at least one video game asset based on the analysis, where the asset is held in the inventory of a video game player for a particular video game. The instructions can also be executable to propose obtaining at least one video game asset based on the analysis, where the asset is not currently held in the inventory of a video game player for a particular video game and / or where obtaining one or more assets is proposed. Additionally, or alternatively, the instructions can be executable to propose using at least one video game asset based on the analysis, where the asset is held in the inventory of a video game player for a particular video game.

[0006] If desired, the instructions can also be executable to provide the proposal using a machine learning configuration model and data of at least one reference player that is different from the video game player. In some examples, the data of at least one reference player can be assigned to a cluster defined by at least one common gameplay characteristic via the machine learning configuration model, and the at least one common gameplay characteristic is also identified as being exhibited by the video game player during play of a particular video game.

[0007] In various examples, the asset itself can include one or more of virtual weapons, virtual ammunition, virtual armor, virtual crafts, virtual potions, virtual plants, and / or virtual skills.

[0008] Also, in an exemplary embodiment, the analysis and recommendations can be performed by an inventory coach model executed through a video game console external to the video game environment itself. The inventory coach model can be configured to recommend video game assets for a plurality of different video games, if desired.

[0009] In addition, in some exemplary embodiments, instructions can be made executable to analyze the play style of a video game player associated with a particular video game, at least in part, by analyzing the combat style of the video game player and / or at least one other gameplay habit of the video game player. Based on the analysis of at least one gameplay habit and also based on at least a portion of the video games the user will play in the future, the instructions can then be made executable to recommend discarding, acquiring, and / or using at least one video game asset.

[0010] In another aspect, the method includes analyzing the gameplay of a video game player on a device with respect to a particular video game being played by the video game player. The method also includes recommending discarding, acquiring, and / or using at least one video game asset based on the analysis.

[0011] Thus, in some examples, the analysis of a video game player's gameplay can include an analysis of the combat style of the video game player for participating in combat in the video game. The combat style of the video game player can be determined to be ranged combat and / or melee combat.

[0012] Also, in various exemplary embodiments, the analysis can be performed using a video game console that facilitates the gameplay of a particular video game and a remotely located server that performs the analysis. In this way, the analysis and recommendations can be performed by an inventory coach model executed on the server, which communicates with the video game console via a graphical user interface displayed external to the video game environment to present the recommendations. In some examples, the video game player may be a first video game player, and the inventory coach model can be trained through machine learning to adjust the recommendations of video game assets based on the learned habits of a threshold number of other video game players other than the first video game player. The other video game players can include, for example, at least a second video game player determined to have a gameplay style that matches the gameplay style of the first video game player and at least a third video game player determined to have a gameplay style that does not necessarily match the gameplay style of the first video game player.

[0013] In yet another aspect, the assembly includes at least one computer including at least one processor programmed with instructions that identify a collection of assets of a first video game player for a particular video game being played by the first video game player. The processor is also programmed to associate the first video game player with a cluster of other video game players and, based on the association, propose to discard and / or use at least one of the assets.

[0014] In some examples, the processor can be programmed to associate a first video game player with a cluster based on the first video game player and a member of a cluster having a similar gameplay style for playing a particular video game as determined by an inventory coach model.

[0015] The details of the present application can be best understood with reference to the accompanying drawings in terms of both its structure and operation, in which like reference numerals refer to like parts.

Brief Description of the Drawings

[0016]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Mode for Carrying Out the Invention

[0017] The present disclosure generally relates to a computer ecosystem including, but not limited to, aspects of a home appliance (CE) device network such as a computer game network. The systems herein may include server components and client components that can be connected through a network, whereby data can be exchanged between the client components and the server components. The client components may include one or more computing devices, such as game consoles like Sony PlayStation (registered trademark) or game consoles made by Microsoft or Nintendo or other manufacturers, virtual reality (VR) headsets, augmented reality (AR) headsets, portable TVs (e.g., smart TVs, Internet-enabled TVs), portable computers such as laptop and tablet computers, and other mobile devices including smartphones and additional examples described below. These client devices may operate in various operating environments. For example, some of the client computers may use, by way of example, the Linux (registered trademark) operating system, Microsoft's operating system, or the Unix (registered trademark) operating system, or an operating system manufactured by Apple or Google. Using these operating environments, one or more browsing programs such as browsers created by Microsoft or Google or Mozilla, or other browser programs that can access websites hosted by the Internet servers described below may be executed. Also, one or more computer game programs may be executed using an operating environment in accordance with this principle.

[0018] The server and / or gateway may include one or more processors that execute instructions to configure a server that receives and transmits data through a network such as the Internet. Alternatively, the client and server may be connected through a local intranet or a virtual private network. The server or controller may be instantiated by a gaming machine such as Sony PlayStation (registered trademark), a personal computer, or the like.

[0019] Information may be exchanged between the client and the server through the network. For this purpose and for security, the server and / or client may include a firewall, a load balancer, a temporary storage, and a proxy, as well as other network infrastructure for reliability and security. One or more servers may form an apparatus that implements a method of providing a secure community such as an online social website to network members.

[0020] The processor may be a single-chip processor or a multi-chip processor that can execute logic by various lines such as address lines, data lines, and control lines, as well as registers and shift registers.

[0021] The components included in one embodiment can be used in any suitable combination in other embodiments. For example, any of the various components described herein and / or shown in the figures can be combined, exchanged, or excluded from other embodiments.

[0022] A "system having at least one of A, B, and C" (similarly, "a system having at least one of A, B, or C" and "a system having at least one of A, B, C") includes a system having A alone, B alone, C alone, A and B together, A and C together, B and C together, and / or A, B, and C together.

[0023] Referring specifically to FIG. 1 here, an exemplary system 10 is shown, and the system 10 may include one or more of the exemplary devices described above and detailed below according to the present principle. The first device among the exemplary devices included in the system 10 is a consumer electronics (CE) device such as an audio-video device (AVD) 12, which is not limited to this, such as an Internet-enabled TV equipped with a TV tuner (or equivalently, a set-top box that controls the TV). Alternatively, the AVD 12 may also be a computer-controlled Internet-enabled ( "smart") phone, a tablet computer, a notebook computer, an HMD, a wearable computer-controlled device, a computer-controlled Internet-enabled music player, a computer-controlled Internet-enabled headset, an implantable skin device, or other computer-controlled Internet-enabled implantable devices. In any case, it should be understood that the AVD 12 is configured to implement the present principle (for example, communicate with other CE devices to implement the present principle, execute the logic described herein, and perform any other functions and / or operations described herein).

[0024] Therefore, in order to implement such a principle, the AVD12 can be established by some or all of the components shown in FIG. 1. For example, the AVD12 may include one or more displays 14, and the one or more displays 14 may be implemented by a high-resolution or ultra-high-resolution "4K" or higher-resolution flat screen, and may be touch-responsive to receive user input signals by touching the display. The AVD12 may include one or more speakers 16 for outputting audio according to this principle, and at least one additional input device 18 such as an audio receiver / microphone for inputting audible commands to the AVD12 to control the AVD12. An exemplary AVD12 may also include one or more network interfaces 20 for communicating through at least one network 22 such as the Internet, WAN, LAN, etc. under the control of one or more processors 24. It may also include a graphics processor. Therefore, the interface 20 may be, but is not limited to, a Wi-Fi transceiver, and the Wi-Fi transceiver is an example of a wireless computer network interface such as, but not limited to, a mesh network transceiver. It should be understood that the processor 24 controls the AVD12 to implement the principle described herein, including other elements of the AVD12 such as controlling the display 14 to present an image thereon and receiving input therefrom. Further, it should be noted that the network interface 20 may be a wired or wireless modem or router, or other suitable interface such as a wireless telephone transceiver or the Wi-Fi transceiver described above.

[0025] In addition to the above, the AVD12 may also include one or more input ports 26, such as, for example, a high-definition multimedia interface (HDMI (registered trademark)) port or a USB port for physically connecting to another CE device, and / or a headphone port for connecting headphones to the AVD12 to provide audio to the user through the headphones. For example, the input port 26 may be connected wired or wirelessly to a cable of audio-video content or a satellite source 26a. Thus, the source 26a may be a separate or integrated set-top box, or a satellite receiver. Alternatively, the source 26a may be a game console or a disc player containing content. When implemented as a gaming machine, the source 26a may include some or all of the components described below in relation to the CE device 44.

[0026] The AVD12 may further include one or more computer memories 28, such as a disk-based or solid-state storage device, which are not temporary signals, and which in some cases are embodied in the AVD chassis as a stand-alone device or as a personal video recording device (PVR), or as a video disc player either inside or outside the AVD chassis for playing AV programs, or as a removable memory medium. Also, in some embodiments, the AVD12 may include a position receiver or location receiver, such as, but not limited to, a cellular phone receiver, a GPS receiver, and / or an altimeter 30, configured to receive geographical location information from a satellite base station or a cellular phone base station and provide the information to the processor 24, and / or to determine the altitude at which the AVD12 is disposed together with the processor 24. The component 30 may also be realized by an inertial measurement unit (IMU) typically including a combination of an accelerometer, a gyroscope, and a magnetometer to determine the position and orientation of the AVD12 in three dimensions.

[0027] Continuing the description of the AVD12, in some embodiments, the AVD12 may include one or more cameras 32, and the one or more cameras 32 may be digital cameras such as thermal imaging cameras, web cameras, etc., and / or cameras integrated into the AVD12 to collect photos / images and / or videos according to this principle and controllable by the processor 24. Also, included in the AVD12 may be a Bluetooth® transceiver 34 and other NFC elements 36 for communicating with other devices using Bluetooth® and / or Near Field Communication (NFC) technologies respectively. An exemplary NFC element may be a Radio Frequency Identification (RFID) element.

[0028] Furthermore, the AVD12 may include one or more auxiliary sensors 38 that provide inputs to the processor 24 (such as motion sensors like accelerometers, gyroscopes, cyclometers, etc., or magnetic sensors, infrared (IR) sensors, optical sensors, speed sensors and / or cadence sensors, gesture sensors (e.g., sensors for detecting gesture commands)). The AVD12 may include a wireless television broadcast port 40 for receiving wireless (OTA) TV broadcasts that provide inputs to the processor 24. It should be noted that in addition to the above, the AVD12 may also include an IR transmitter and / or IR receiver and / or IR transceiver 42 such as an Infrared Data Association (IRDA) device. A battery (not shown) may be provided to power the AVD12, and a kinetic energy harvester that can convert kinetic energy into electricity to charge the battery and / or power the AVD12 may be possible. A Graphics Processing Unit (GPU) 44 and a Field Programmable Gate Array 46 may also be included. One or more tactile generators 47 may be provided to generate tactile signals that can be detected by a person holding or touching the device.

[0029] Referring further to FIG. 1, in addition to the AVD 12, the system 10 may include one or more other CE device types. In one embodiment, the first CE device 48 may be a computer game console that can be used to send the audio and video of a computer game to the AVD 12 via commands sent directly to the AVD 12 and / or through the server described below, while the second CE device 50 may include components similar to those of the first CE device 48. In the example shown, the second CE device 50 may be configured as a computer game controller operated by a player or a head-mounted display (HMD) worn by the player. In the example shown, only two CE devices are shown, and it should be understood that fewer or more devices may be used. The devices herein may implement some or all of the components shown for the AVD 12. Any of the components shown in the following figures may incorporate some or all of the components shown for the AVD 12.

[0030] Referring now to at least one of the servers 52 described above, the server 52 includes at least one server processor 54, at least one tangible computer-readable storage medium 56 such as disk-based storage or solid-state storage, and at least one network interface 58 that enables communication with other devices in FIG. 1 through the network 22 under the control of the server processor 54 and that can actually facilitate communication between the server and client devices in accordance with this principle. Note that the network interface 58 can be, for example, a wired or wireless modem or router, a Wi-Fi transceiver, or other suitable interface such as, for example, a wireless telephone transceiver.

[0031] Accordingly, in some embodiments, server 52 can be an Internet server or an entire server “farm,” can include “cloud” functionality, can perform “cloud” functionality, whereby the devices of system 10 can access a “cloud” environment, e.g., via server 52 in an exemplary embodiment regarding a network gaming application. Or, server 52 can be implemented by one or more gaming machines, or other computers in the same room as or near the other devices shown in FIG. 1.

[0032] The components shown in the following figures can include some or all of the components shown in FIG. 1. The technology described herein can be downloaded from Internet server storage and embodied in a computer application (an “app”) stored on any storage implementation computer described herein. Additionally or alternatively, the technology can be implemented in whole or in part by a server located remotely from a video game console that can also be used to implement some or all of the technology.

[0033] With this in mind, the detailed description of the present application recognizes that some video games offer seemingly endless possibilities for weapons, ammunition, and armor, as well as other inventory items that can be collected and used. For example, the game may allow the player to collect various plants, skins, potions, resources, and other assets / items used to craft supplies such as ammunition and upgraded weapons. However, the game may also limit the amount of assets that a player's virtual character can hold at any one time. Thus, in accordance with this principle, the inventory coach can operate across all such games on various gaming platforms and guide the player as to when to obtain, sell, and / or drop various items / resources that the player may or may not need or consider useful in the future, depending on which part or stage of the game the player has reached.

[0034] The inventory coach can also help the player determine whether the player is using the best virtual weapons, ammunition, and / or armor for a particular game, taking into account the player's own play style. Thus, if the player has a tendency to attack or kill virtual enemies from a distance, the inventory coach can suggest weapons that perform ranged attacks, such as long bows or sniper rifles, along with corresponding armor if available in the game. Similarly, if the player has a tendency to engage in close combat, the inventory coach can suggest shotguns or swords along with appropriate armor.

[0035] Accordingly, the inventory coach can notify the player when, for example, the player is carrying around a lot of irrelevant virtual items in the vicinity for crafting and equipment upgrades, or when the player can potentially sell or discard some of these items without negatively impacting the player's chances of game success. The inventory coach can also notify the player whether the player is using the most appropriate weapons and armor from the player's inventory or collection of other assets, taking into account the player's specific playstyle, and further whether the player might want to obtain new or different items or skills for an enhanced, more efficient gaming experience.

[0036] To achieve the foregoing, the inventory coach processes telemetry data collected by a video game console or other computer to determine specific coaching suggestions for providing, which can be remotely stored from the console to an Internet-based server. The inventory coach can use machine learning (ML) to learn how other players have handled the same part of the same game and with cluster gamers following a similar playstyle. The coach can then analyze the inventory coach that can help the player progress through the game more quickly. The coach can also consider inventory items and items that might be needed for each item for future upgrades and crafting, based on the current playstyle and how far the player has already progressed through the game. For example, if the player is carrying around an item that seems unnecessary because all upgrades have been achieved, or if the item will not be used for the rest of the game, the coach can recommend selling or dropping the item.

[0037] In some exemplary embodiments, the inventory coach can be activated during a particular game by going out to the shell interface for the console and starting / initiating the coach. However, in other situations, the inventory coach can be activated directly from the game itself. When the inventory coach is started from the shell interface, if the game is not currently active or not started from the player's console, the coach can present the player with the threshold number of the most recently played games (e.g., the last three games played on that console) and ask the player to select the game for which to activate the inventory coaching for that game.

[0038] Figure 2 shows the overall logic in accordance with what is described above, where this logic can be executed by one or more devices such as a personal computer, a video game console, and / or a remotely located server in any suitable combination. Starting from block 200, the device can execute the video game itself to facilitate gameplay. The logic can then proceed to block 202, where the device can analyze the gameplay of a particular player, for example, by performing feature engineering to identify the particular habits / preferences of the player. The particular player will be described below as the "first player" with reference to Figure 2. However, it is understood that this first player may or may not be a beginner or a casual gamer.

[0039] Next, from block 202, the logic can proceed to block 204, where the device can identify elite players who play the same video game exceptionally well and can also be determined based on quantifiable metrics established by the game developer or console manufacturer. The metrics can include, in any suitable combination, success (or failure), highest score, length of the minimum time to complete a level or the game itself, most virtual characters killed, most virtual currency acquired, and so on. In some examples, elite players can be identified and grouped according to overall performance regardless of gameplay style. Additionally or alternatively, elite players can be identified and grouped according to performance using a given gameplay style. Elite players in general and elite players for each gameplay style can function as reference groups or clusters for purposes further described below.

[0040] However, first, it should be noted that the logic of FIG. 2 can then proceed from block 204 to block 206. At block 206, the device can train an inventory coach artificial intelligence (AI) model via various machine learning techniques to infer, as output, one or more game performance metrics for a given set of inputs for an elite player, where the inputs can include the use of specific game assets (e.g., virtual items and / or game skills) alone and / or in combination with each other, and optionally the use at a particular point in the game. The output can also include an overall game play style. Thus, the output and insights derived from the correlation between the input and output of the model can be used during deployment to make proposals for specific assets to a given player of a game that includes a first player. The output that can be used for training can include various game play styles that generally indicate success or failure, and / or various metrics (or one or more of the specific quantifiable metrics described above), at a particular aspect or point in time of the game, and / or overall, in relation to the use of a given set of assets that establish the corresponding training input.

[0041] Thus, in this principle, various machine learning models including deep learning models can be used. Machine learning models in accordance with this principle can use various algorithms trained in ways including supervised learning, unsupervised learning, semi-supervised learning, reinforcement learning, functional learning, self-learning, and other forms of learning. Examples of such algorithms that can be implemented by a computer circuit include one or more neural networks such as convolutional neural networks (CNNs), recurrent neural networks (RNNs), and the type of RNN known as long short-term memory (LSTM) networks. Support vector machines (SVMs) and Bayesian networks can also be considered examples of machine learning models.

[0042] Thus, as understood herein, performing machine learning can include accessing training data and then training the model to enable the model to process additional data for making inferences. An artificial neural network / artificial intelligence trained by machine learning can thus include an input layer, an output layer, and a plurality of hidden layers therebetween, and those layers are configured and weighted to make inferences about appropriate outputs.

[0043] Once the inventory coach model is trained (or further trained as more data becomes available for elite players), the logic can then proceed to block 208, where the device can use the trained inventory coach model to associate the first player with one or more clusters of other players for each block 204, the clusters of elite players sharing the same or a similar identified game play style as the first player. The logic can then proceed to block 210.

[0044] In block 210, the inventory coach model can propose one or more video game assets that can already exist in, be discarded from, be acquired, and / or be used in the inventory of the first player, based on the general success of elite players regardless of game play style and / or based on the first player's own game play style that matches a particular cluster. In some examples, the proposal can also be adjusted based on the remaining portion of the game that the user has not yet succeeded in, such that items that would otherwise be useful according to the first player's game play style but are only useful for past aspects of games that have already been played are omitted.

[0045] Figure 3 shows a schematic diagram of a process that can be used in conjunction with the logic of Figure 2 to use clusters of various players to provide suggestions to a first player. The first player of Figure 2 is generally designated as target player 300 in Figure 3, in which case the console or computer of the target player accesses inventory coach model 302 as described herein. Model 302 itself can group general elite players into a first cluster 304, regardless of the game play style, and one or more recommendations 306 for target player 300 can be created based on the overall game play and achievements of that cluster (cluster 304). Model 302 can also group elite players with a game play style similar to that of target player 300 into a second cluster 308, and one or more recommendations 310 for target player 300 can be created based on the game play and achievements of that cluster (cluster 308). Note that in some embodiments, a given elite player can be associated with both clusters 304, 308 if the reference player is both a general elite player and an elite player with the same game play style as target player 300. In any case, note that ultimately, the combined result 312 can be presented to target player 300 in accordance with this specification.

[0046] Therefore, the inventory coaching model 302 can calculate player clusters based on in-game data collected by each console of elite- criterion players, which is reported to a server associated with the console manufacturer. Therefore, feature engineering for each elite-criterion player is performed using the respective in-game data for the individual criterion players at the server to extract useful features / learn player habits. Methods / models available for feature engineering include transformer networks such as BERT, RoBERTa, XLM-RoBERTa, and feature engineering techniques such as transformation, binning, scaling, projection, etc. A clustering algorithm can then be applied to the extracted features, and methods / models for such clustering can include clustering algorithms such as K-means clustering and hierarchical clustering. Note that similar methods and models can also ultimately be used for feature engineering and clustering in relation to the gameplay of the target player 300.

[0047] For elite players identified to function within the reference groups / clusters 304, 308 themselves, design metrics can be used to quantify the game performance of reference players in order to identify elite players from a broader set of players from whom telemetry data has been reported. This metric can aggregate data such as success rate of activities, time consumption, score / rating. Thus, players can be sorted by game performance metrics, and a percentage or number of the upper threshold of players can be used as a given reference group / cluster qualification (such as the top 10% or top 100 players). Next, a server or other computer that processes in-game data for reference players can create one general reference group from the entire player population (e.g., different players with different game play styles grouped together), and can also create cluster-specific / game play style-specific reference groups based on learned game play habits.

[0048] These two clusters can then be used to generate recommendations regarding inventory resource management for the target player 300. This can be done by a server or other computer first comparing the resource distributions and detecting significant discrepancies. For example, the target player may currently have a number of "Small Potion", "Purified Water" and "Bamboo" in their personal player inventory, but it is recognized that the general reference group and / or game play style-specific reference group to which the target player 300 is assigned tend to not have any of these items.

[0049] Next, the server or other computer can examine the inventory management actions employed by the reference players of the reference group / cluster after comparing the resource distributions to detect significant discrepancies. For example, players in one or more reference clusters can craft "Large Potion" from "Small Potion" and "Purified Water" while dropping all "Bamboo". The server can then construct meaningful recommendations from the information extracted above, such as suggesting to discard those items from the inventory of target player 300 or providing proposals to convert certain items into "Large Options" similar to how the reference player(s) did it.

[0050] Recommendations can be generated based on the items and skills that can be equipped. This can be done by console developers, system administrators, or other technicians who construct one or more interpretable regression models between the use of items / skills and game performance metrics to identify items and combinations of items that may improve the performance of one's game (e.g., the performance of target player 300). The methods / models that can be used include generalized linear models and tree-based models such as random forests and XGBoost. For training such models and also for deployment, the inputs to the model structure can be the use of various items / skills and their use in various combinations (expressed as interaction terms). The output may then be a game performance metric. Here, during the training phase(s), the output can be labeled as "good", "bad", and "elite / outstanding" game play, or other labeling can be used as desired (e.g., if supervised machine learning is employed).

[0051] Thus, when the target player accesses the inventory coach model according to the description of FIG. 7 or by other means, the current inventory management state of the player can be compared with the actions taken by that reference player who has been found to be excellent in the game at the same time point (e.g., in terms of game progress). The recommendations can be created based on two groups of players, namely, the general elite players 304 who have the best performance throughout the game, even if they do not necessarily use the same game play style, and the high-performance players 308 who are algorithmically determined to have a game play style similar to that of the target player 300.

[0052] By using two reference player groups that are not specific to the game play style, the recommendations / suggestions made to the target player 300 by the inventory coach model can be diversified and personalized. Thus, in addition to providing general recommendations that help the player optimize their existing resources and obtain resources beneficial for progress (e.g., using the generalized cluster 304), personalized recommendations based on the game play style can also be provided based on the previous game experience of the target player.

[0053] From a game design perspective, it should be noted that the resources / assets available in the game can have unique interactive dynamics, so the optimization of resources / assets for a given target player can depend on the player's game play style in that game. For example, in an action adventure game, players who prefer to participate in brawling battles can receive suggestions for acquiring items and skills that maximize damage output and defense at close range. On the other hand, players who enjoy long-range combat can benefit from long-range weapons and skills. Table 400 in FIG. 4 shows examples of such assets according to the game play style.

[0054] As shown in FIG. 4, an exemplary data table 400 is shown, and these data tables can also be used in combination with the logic of FIG. 2. Table 400 can be constructed based on empirical data of various characteristics of elite standard players by game developers or console developers, and / or can be input by the inventory coaching model described herein based on identified elite standard player actions and / or "good" action inferences made using one or more of the models discussed herein. Thus, Table 400 can be appropriately used to provide suggestions to target players such as player 300.

[0055] As shown in FIG. 4, the first column 402 can indicate the game play style shown as "play style abstraction" in FIG. 4. Here too, it should be noted that the game play style can be determined using feature engineering and clustering, for example, to assign a given player to a particular cluster specific to a particular game play style, as described herein.

[0056] As also shown in FIG. 4, when a specific gameplay style for a target player is inferred, data within column 404 for each gameplay style can be accessed to identify weapons that can be proposed to be acquired or used by the target player (e.g., if already in the target player's inventory). Data within column 406 can also be accessed for each gameplay style to identify equipment and / or armor that can be proposed to the target player to acquire or wear if already present in the target player's inventory. Similarly, data in column 408 can be accessed for each gameplay style to identify skills that can be proposed to the target player to acquire or use if already in the target player's inventory. Note that the specific data shown in FIG. 4 is presented as an example and is not intended to be limiting.

[0057] FIG. 5 shows another schematic view of a process executable by an inventory coach model in accordance with the present principles. As illustrated, raw in-game data 500 for a particular player can be processed using feature engineering 502 to identify useful features 504. The raw in-game data 500 can be generated by video game telemetry that can be collected or generated by a target player's local video game console / computer that executes a video game based on, for example, controller input, output from the video game application itself as the target player progresses through a given video game. Controller input can include, for example, game movement, input to switch weapons within an inventory cache, weapons and ammunition used against an enemy and the resulting damage, worn armor, and damage received by the enemy each time it is hit. The raw in-game data can then be reported to a server or other computer using an inventory coach model in accordance with the present principles.

[0058] Once useful feature 504 is identified, clustering algorithm 506 can be executed to assign the target player to one or more player clusters 508 that include clusters that match the game play style of the target player. Here, note that in some examples, if two or more game play styles are determined for the target player, the target player may be assigned to clusters specific to the two or more game play styles.

[0059] Referring now to FIG. 6, an exemplary artificial intelligence (AI) architecture 600 for an inventory coaching model is shown. As illustrated, architecture 600 can include a feature extractor 602 that can perform feature engineering using raw in-game data 604 as input as described herein. Next, the output 606 of the useful features can be provided to a cluster classifier 608 as input for the cluster classifier 608 to use a clustering algorithm on the extracted features 606 to provide one or more clusters determined from the useful features 606 as output 610. Next, a proposal engine 612 can employ the output 610 from the cluster classifier 608 as input to identify one or more proposals to make to the target player to whom the raw in-game data 604 belongs, using a table such as table 400 of FIG. 4 for example.

[0060] Accordingly, by using the various logics, methods, tables, and architectures described above, recommendations or proposals can be provided to a given target player using the performance of the most successful players in history regardless of gameplay style and / or using other benchmark players who have been most successful using a similar gameplay style. The proposals can relate to optimizing existing resources already available in the player's game inventory. The resources can include manageable items, abilities, skills, etc. Optimization can include item equipping, selling, crafting, changing, and / or dropping. The proposals can be in the form of achievable resource / asset proposals, in which case achievable further depends on the player's current game level, money, location, etc., such that only assets available at a given level of the game are proposed, only assets that the player can afford with the current amount of virtual currency available to the player are proposed, and / or only assets that the player can acquire (or acquire a lot) or discard at a given geolocation in the game virtual world are proposed.

[0061] Accordingly, recommendations and proposals may be made based on the player's current progress in the game (e.g., character level, game progress, timestamp, etc.). Accordingly, resource optimization can help significantly improve the player's game performance through the inventory coach model itself that provides the player with easy access to resource optimization assistance.

[0062] FIG. 7 shows an exemplary dialog in a graphical user interface (GUI) 700 that is displayed on a display under the control of a video game console or other computer, controls aspects of the console / computer's operating system (OS), and provides access to certain functions at the OS level external to the particular video game environment itself. The GUI 700 is reachable by pressing certain designated buttons on a keyboard, mouse, video game controller, joystick, etc. that may be used to control the game and / or console. As shown, various selectable "apps" can be presented, including an app 702 selectable to start a particular video game having the fiction title "Hunted Island". Also, an online store app 704 can be selected to launch an online marketplace from which other video games can be purchased and downloaded, and a settings app 706 can be selected to launch a settings application that can configure even more settings of the console. For example, app 706 can be selected to instruct the console to display the GUI 1200 of FIG. 12, which will be described later.

[0063] Further, as shown in FIG. 7, the GUI 700 can include an inventory coach app 708 that can embody the inventory coach model(s) described above. In a situation where the game is not currently active and is being run on the console, selection of app 708 can cause the GUI 800 of FIG. 8 to be displayed. However, in a situation where a particular video game is currently active and the player may have paused it to progress to the GUI 700, selection of app 708 can cause the console to directly present the GUI 900 of FIG. 9, to put it briefly.

[0064] However, first referring to FIG. 8, it should be noted that the displayed GUI 800 may include an indication 802 that the video game is not currently active. The GUI 800 can also list various selectable video game apps 804-808 that can be selected, which can then cause another GUI, such as GUI 900 in FIG. 9, to be displayed for each video game (although it may be outside the environment of the video game itself), and provide the player with suggestions from the inventory coach regarding the specific video game associated with any app selected from selectors 804-808. Also, in some situations, even if the game may not currently be active, the inventory coach can still provide suggestions for the selected game based on the player's previous game play and game state, saved manually or automatically before the most recent shutdown or termination of each video game app itself. Also, it should be noted that if the player wants to receive inventory coaching for games other than the three most recently executed and listed games, by selecting another selector 810, the player can browse to the desired game from the game library.

[0065] Therefore, after going out to the shell interface 700 and starting the inventory coach, the player may be able to access various suggestions for one or more video games. Generally, in some examples, the suggestions may be provided separately for different clusters. For example, the suggestions may be that elite players who are generally good at the game tend to take actions A, B, C, while elite players who play similarly to the target player tend to take actions D, E, F.

[0066] More specific examples in accordance with the above matters are shown in FIG. 9. Here, the GUI 900 for the inventory coach can be presented on a display controlled by a console or other computer used for the game. Here too, if the game is already being executed and the current game state data is maintained in the RAM, it is possible to automatically reach the GUI 900 in response to the selection of the coach application 708. On the other hand, it should be noted that if the game is not currently being executed and not maintained in the RAM, it is possible to reach the GUI 900 by selecting each application from the GUI 800.

[0067] In either case, the GUI 900 can be presented. As shown in FIG. 9, at a specific stage of the game where the target player is currently located, regardless of the game play style, it can include a display 902 that the entire elite player tends to drop the pistol and acquire the shotgun (proposal 904), and / or tends to switch from armor X to armor Y (proposal 906). It should be noted that each of the proposals 904 and 906 in this example is established by a selectable button or other type of selector that can be selected to provide the game itself with a command to execute the proposed action. Therefore, the selection of the selector 906 can instruct the game to drop the pistol currently held by the virtual character of the target player (if actually already held), and / or select the shotgun in the inventory of the target player. Similarly, the selection of the selector 908 can instruct the game to remove armor X from the player's virtual character (if already worn), and / or place armor Y on the player's virtual character.

[0068] As also shown in FIG. 9, the GUI 900 can also include a display 908 that at a particular stage of the game in which the target player is currently located, elite players with the same gameplay style as the target player tend to drop pistols and acquire swords (proposal 910), and / or tend to use and continue to wear Armor X (proposal 912). Here too, note that for each of proposals 910, 912, it can be selected to provide a command to the game to perform the relevant action (e.g., acquire a sword according to proposal 910, or use and wear / contiue with Armor X according to proposal 912).

[0069] FIG. 10 shows another GUI 1000 that can be presented as part of an inventory coach for providing proposals to a target player. Based on the current progress of the target player to a certain level of a video game, a display 1002 is shown that includes a proposal that the player does not need "Magic Potion 53" for the rest of the game. This proposal can be made by the inventory coach, for example, based on the determination that the player has already passed a particular level of the game where "Magic Potion 53" is used and it will not be useful in subsequent levels of the game. Thus, the display 1002 can propose that the target player's virtual character discard it and create room in the player's limited inventory to acquire weapons that the player may find useful, based on the player's gameplay style. Note that the proposal is the same in any case, and the proposal is not divided between general elite players and elite players of the particular type of gameplay style in this example.

[0070] In any case, it should be noted that the selector 1004 may be selected by the target player to supply a command to the game for the player's virtual character to drop or discard "Magic Potion 53" from the player's available inventory. The selector 1006 gives a command to the player's virtual character for the player's virtual character to obtain a vine ladder, assuming that the corresponding item is nearby or in a state where it can be immediately obtained, or displays a virtual map of the game world indicating the virtual position where the vine ladder may be located, so that the player can know where to navigate their virtual character to obtain the vine ladder. Furthermore, if the player does not wish to supply any commands, the selector 1008 can be selected to simply return to the game from which the player disconnected when the inventory coach was called.

[0071] FIG. 11 shows yet another GUI 1100 that can be presented as part of an inventory coach to provide a proposal to the target player. The display 1102 including the proposal shows that it appears as if the player prefers combat in facilities closer than remote combat, based on the identified game play style of the target player. In this example, it should also be noted that the subsequent proposals are not divided between general elite players and elite players of a specific game play style here, since all the proposals are specific to the particular game play style of the target player.

[0072] As shown in FIG. 11, each particular proposal may be presented in the form of selectors 1104-1114, each of which may be selectable to give commands to the game according to the text proposals presented on the face of each selector itself. By way of example, a proposal can include a switch from a sniper rifle, which may be suitable for long-range combat, to a shotgun, which may be suitable for close combat (selector 1104). For similar reasons, selector 1106 can be selected to switch from the use of a sniper rifle to the use of a sword. Similarly, selector 1108 can be selected to give a command to drop a sniper rifle on the virtual character of the target player, whereas selector 1110 can be selected to command the game to present this map or other data from which the player can find a pistol suitable for close combat. If desired, the player can also select selector 1112, which commands the player's virtual character to wear armor suitable for combat in close quarters combat. If the user does not wish to select any of the other selectors 1104-1112, selector 1114 can be selected to simply return to the game itself.

[0073] Continuing with the detailed description with reference to FIG. 12, it shows an exemplary GUI 1200 that can be displayed on a display controlled by a game console or other computer to configure one or more settings related to an inventory coach in accordance with the present principle. For example, GUI 1200 may be displayed in response to the selection of selector 706 from GUI 700 of FIG. 7. In the following examples, it should be understood that each option or sub-option can be selected by selecting the respective checkbox adjacent to each option or sub-option.

[0074] As shown in FIG. 12, the GUI 1200 can include a first option 1202 that configures or enables the inventory coach described herein for use by the console in making proposals to the target player. Thus, the inventory coach can be turned on by selecting option 1202 and turned off by deselecting option 1202. The proposals can be reached from the console's shell interface and provided according to the GUI described above. However, in some embodiments, the proposals and / or the GUI itself can be presented in response to a pause command as a pop-up proposal overlaid on the game's graphics, from the game-specific options menu, and / or while the player is playing the game in real time while within the game environment. In practice, option 1208 can be selected to specifically configure or set up the inventory coach to present in-game proposals while the player is playing in real time, either by indicating that the proposals are available via overlay graphics and / or by presenting a non-full-screen version of the GUIs 900, 1000, and 1100 themselves.

[0075] FIG. 12 also shows that sub-options 1204 and 1206 can be presented under option 1202. Sub-option 1204 can be selected to configure or set up the inventory coach to specifically use the identified gameplay style of the target player when making proposals, while sub-option 1206 can be selected to configure or set up the inventory coach to specifically use the current game state of the target player (e.g., the stage or level currently reached in the game, achievements, and progress already made) when making proposals.

[0076] Further, if desired, Option 1210 can be presented as part of GUI 1200, and the inventory coach can be instructed to present only the suggestions determined from the identified game play style of the target player to the target player. Thus, if Option 1210 is selected, it is not possible to present suggestions derived from overall elite standard players (e.g., regardless of the game play style described above), and only suggestions derived from standard players with a similar game play style can be presented.

[0077] Moving on from FIG. 12, it should be noted more generally that the assigned clusters of a given target player for a given game play style can change if the player's own style changes or improves over the course of playing a particular video game, and those changes can be recognized by the inventory coach. Thus, timestamps for various activities as shown in game telemetry data may be used, for example, to identify such transitions in a role-playing game. In the case of a sports game where each game or match within the video game itself (e.g., an 8-minute basketball or football game) has a defined length, each discrete and separate game / match can be analyzed separately to determine the specific game play style of that particular game / match. Then, if the target player is identified as transitioning from one game play style cluster to another, future suggestions can be based on the new game play style cluster to which the target player is transitioning.

[0078] Although the principles have been described with reference to some exemplary embodiments, it is recognized that these are not intended to be limiting and that the subject matter claimed herein can be practiced using various alternative arrangements.

Claims

1. A device, comprising: At least one computer storage device, which is not a transitory signal, and which is accessed by at least one processor; analyzing a playing style of a video game player in relation to a particular video game being played by the video game player; the at least one computer storage having instructions recorded thereon executable to provide, via a graphical user interface (GUI) presented on a display, at least two different types of suggestions for discarding, acquiring, and / or using at least one video game asset based on the analysis; Equipped with The device, wherein the GUI provides a first suggestion to discard, acquire and / or use a first video game asset based on past gameplay data of elite players who share the same play style as the video game player, and a second suggestion to discard, acquire and / or use a second video game asset based on past gameplay data of elite players in general regardless of the play style of the video game player.

2. The instruction: The device of claim 1 , further comprising: a processor configured to: determine whether or not a video game player has a video game asset that is associated with a video game player and a video game player configured to play the video game player on a video game device;

3. The instruction: The device of claim 1 , further comprising: a device operable to suggest acquiring at least one video game asset based on the analysis, the asset not currently held in the video game player's inventory for the particular video game, and / or one or more of the assets are suggested to be acquired.

4. The instruction: The device of claim 1 , further comprising: a device operable to suggest using at least one video game asset based on the analysis, the asset being maintained in the video game player's inventory for the particular video game.

5. The instruction:

11. The device of claim 1, wherein the device is executable to make the suggestions using a machine learning configuration model and data of at least one reference player, the reference player being different from the video game player.

6. 6. The device of claim 5, wherein the data of the at least one reference player is assigned via the machine learning configuration model to a cluster defined by at least one common gameplay characteristic, the at least one common gameplay characteristic also being identified as exhibited by the video game player while playing the particular video game.

7. The device of claim 1 , wherein the assets include one or more of a virtual weapon, a virtual ammunition, a virtual armor, a virtual craft, a virtual potion, and a virtual plant.

8. The device of claim 1 , wherein the analysis and recommendations are performed by an inventory coach model that runs through a video game console outside of the video game environment itself.

9. The device of claim 8 , wherein the inventory coach model is configured to provide video game asset suggestions for a plurality of different video games.

10. The instruction: The device of claim 1 , wherein the device is executable to analyze a playing style of a video game player in relation to the particular video game, at least in part, by analyzing a fighting style of the video game player.

11. The instruction: analyzing a play style of the video game player in connection with the particular video game, at least in part, by analyzing at least one game playing habit of the video game player; The device of claim 1 , further comprising: a device operable to suggest that the at least one video game asset be discarded, acquired, and / or used based on the analysis of the at least one game playing habit and based at least in part on the video games that the user will play in the future.

12. A method executed by at least one computer including at least one processor, comprising: said processor analyzing at said device a playing style of said video game player in association with a particular video game being played by said video game player; and providing, via a graphical user interface (GUI) presented on a display, at least two different types of suggestions for discarding, acquiring, and / or using at least one video game asset based on the analysis; Including, The method, wherein the GUI provides a first suggestion to discard, acquire and / or use a first video game asset based on past game play data of elite players who share the same play style as the video game player, and a second suggestion to discard, acquire and / or use a second video game asset based on past game play data of elite players in general regardless of the play style of the video game player.

13. The method of claim 12 , wherein analyzing the gameplay of the video game player includes analyzing a fighting style of the video game player for participating in video game combat.

14. The method of claim 13 , wherein the fighting style of the video game player is determined to be one or more of: ranged combat, close combat.

15. 13. The method of claim 12, wherein the analysis is performed using a video game console that facilitates gameplay of the particular video game and using a remotely located server that performs the analysis.

16. 16. The method of claim 15, wherein the analysis and recommendations are performed by an inventory coach model running on the server, the server communicating with the video game console to present the recommendations via a graphical user interface displayed outside of the video game environment.

17. 17. The method of claim 16, wherein the video game player is a first video game player, and the inventory coach model is trained through machine learning to tailor video game asset suggestions based on learned habits of a threshold number of other video game players other than the first video game player.

18. 18. The method of claim 17, wherein the other video game players include at least a second video game player determined to have a game play style that matches the game play style of the first video game player, and at least a third video game player determined to have a game play style that does not necessarily match the game play style of the first video game player.

Images