Computer systems and programs

Abstract

To provide a new technology that allows for randomization of content elements when generating content elements to be given to a user. [Solution] The server system 1100 sets part or all of the generation instruction information 10 that causes the generation AI to generate content elements 8 to be used in the content, using a random determination method. The set generation instruction information 10 is given to the generation AI to generate and acquire new content elements 8, and the system controls the provision of content using the acquired content elements 8.

Description

Technical Field

[0001] The present invention relates to a computer system and the like.

Background Art

[0002] In recent years, the use of a generation unit represented by generative AI (Artificial Intelligence) in content such as video games has been explored. For example, Patent Document 1 describes a technology that enables an avatar profile to be generated by generative AI and regenerated based on instructions from a user.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In a video game, which is an example of content, there are many cases where an opportunity is set to grant content elements such as items that can be used in the game to a user who is a player. For example, in an action RPG (Role Playing Game), item drops that occur when defeating enemy NPCs (Non Player Characters), clear bonuses that occur when clearing a game stage, virtual lotteries, etc. fall under this. Which content elements are granted are determined by random selection from within a predetermined candidate. And randomness can be said to be one of the interests in such granting opportunities. The same applies when granting content elements in a virtual space such as a metaverse other than video games.

[0005] When generating content elements to be given to a user by generative AI, it is desired not to impair randomness.

[0006] The problem that this invention aims to solve is to provide a new technology that can introduce randomness into the content elements provided to a user when the content generation unit generates the content elements to be given to the user. [Means for solving the problem]

[0007] The first invention for solving the above problems is a computer system that controls the provision of content to the user, A generation instruction information setting means (for example, the generation instruction information setting unit 210 in Figure 15, step S34 in Figure 18) sets the generation instruction information by determining some or all of the information of the generation instruction information to be given to a generation unit (for example, the server-side generation AI3 in Figure 1, the generation unit 208 in Figure 15) that generates a given content element to be used in the aforementioned content based on given generation instruction information, by determining the information of the generation instruction information using a random determination method. An acquisition control means (for example, the acquisition control unit 220 in Figure 15, step S86 in Figure 19) that controls the acquisition of the content element by providing the generation instruction information set by the generation instruction information setting means to the generation unit, The computer system comprises a provision control means (for example, the provision control unit 230 in Figure 15, step S160 in Figure 19) that performs control to provide the content using the content elements acquired by the acquisition control means.

[0008] According to the first invention, the computer system can generate content elements by setting the generation instruction information given to the generation unit to have randomness and providing it to the generation unit. In other words, when the generation unit generates content elements to be given to the user, it becomes possible to introduce randomness into the content elements to be given.

[0009] The second invention is a computer system in which the generation instruction information setting means sets the generation instruction information N times (N≧2; N is a natural number) so that when the acquisition of the content element by the acquisition control means is performed N times, the probability that the content element acquired during the N times satisfies a predetermined condition is changed (for example, steps S24 to S34 in Figure 18).

[0010] Furthermore, the third invention is a computer system in which, in the above-described computer system, the generation instruction information setting means sets the generation instruction information in such a way that the probability of satisfying the predetermined conditions is changed by changing the parameter values ​​related to the determination method (for example, the lottery probability for each rarity in the rarity lottery probability setting data 714 in Figure 17, and the probability 744 of the component definition data 740).

[0011] "N times" could refer to the number of times a content element is added on different occasions, or it could refer to the number of times it is added consecutively on a single occasion. "Predetermined conditions" could be, for example, conditions under which the content element is considered to be relatively frequently and easily obtainable when there is an opportunity to add it, and to have a common appearance and performance. If the predetermined conditions are met, the bonus will have a low value and be considered ordinary by the average user who does not have any particular preferences. This also applies to items with relatively low rarity.

[0012] According to the second or third invention, the computer system can change the probability that a content element acquired and given to the user over N turns satisfies predetermined conditions. Therefore, by appropriately setting the predetermined conditions and the way in which they are changed, for example, while ensuring randomness, if the user has several opportunities to be given a content element, the probability of an extraordinary content element being given to the user increases, thereby enhancing the interest associated with the opportunity to give a content element.

[0013] The fourth invention is a computer system in which the generation instruction information setting means sets the generation instruction information in such a way that it changes the possibility of satisfying the predetermined conditions by making the parameter value related to the determination method when setting the generation instruction information for the nth time different from the parameter value when setting the generation instruction information for the (n+1)th time (for example, steps S26 and S30 in Figure 18 for each execution of Loop A).

[0014] According to the fourth invention, the computer system can change the probability that both the content element generated, acquired, and assigned on the nth time and the content element generated, acquired, and assigned on the (n+1)th time satisfy predetermined conditions.

[0015] The fifth invention is a computer system in which the generation instruction information setting means sets the generation instruction information N times (N≧2) so that when the acquisition of the content element by the acquisition control means is performed N times, the probability that the content element acquired during the N times satisfies a predetermined condition is changed (for example, steps S88 to S96 in Figure 19 for each execution of loop A).

[0016] Furthermore, the sixth invention is a computer system in which the generation instruction information setting means sets the generation instruction information such that the probability of the generation instruction information satisfying a predetermined condition over N times changes.

[0017] According to the fifth or sixth invention, the computer system can change the likelihood that content elements acquired over N times will satisfy predetermined conditions.

[0018] The seventh invention is a computer system that further includes generation instruction information presentation control means (for example, step S202 in FIG. 25) for performing control to present to the user some or all of the generation instruction information set by the generation instruction information setting means.

[0019] According to the seventh invention, the computer system can present to the user some or all of the set generation instruction information 10.

[0020] The eighth invention is a computer system in which the generation instruction information setting means performs control to update some or all of the information presented by the generation instruction information presentation control means based on the operation of the user (for example, step S206 in FIG. 25).

[0021] According to the eighth invention, the computer system can update some or all of the presented generation instruction information based on the operation of the user.

[0022] The ninth invention is a computer system in which the generation instruction information includes subject identification information for identifying the subject of the content element and setting value designation information for designating a setting parameter value related to the subject, and the generation instruction information setting means determines the subject identified by the subject identification information and / or the setting parameter value designated by the setting value designation information by the determination method and sets the generation instruction information (for example, the component setting process in FIG. 20).

[0023] Further, the tenth invention is a computer system in which the subject identification information is composed of options in a hierarchical structure of information for specifying details of the subject for each type of the subject.

[0024] According to the ninth or tenth invention, the computer system can determine a subject specified by subject specification information and / or a set parameter value specified by set value specification information by a determination method having randomness, and set generation instruction information.

[0025] The eleventh invention is a computer system in the above computer system, wherein the content is a game, and the subject is at least one of a character, an item, a game stage, and music appearing in the game.

[0026] According to the eleventh invention, the computer system can provide at least one of a character, an item, a game stage, and music that can be used in a game to a user as a content element.

[0027] The twelfth invention is a computer system in the above computer system, wherein the generation instruction information setting means makes the determination by the determination method based on the acquisition result of the content element acquired in the past by the acquisition control means, and sets the generation instruction information (for example, a probability change amount ΔP calculated by a function f having the cumulative grant count M in FIG. 13 as a variable, steps S26 to S34 in FIG. 18).

[0028] Further, the thirteenth invention is a computer system in the above computer system, wherein the generation instruction information setting means makes the determination by the determination method in which a parameter value specifying the randomness related to the determination method is changed based on the acquisition result.

[0029] According to the twelfth or thirteenth invention, the computer system can set generation instruction information by a determination method based on the acquisition result of the content element acquired in the past, that is, the past grant result of the content element to the grant target user.

[0030] The fourteenth invention is a computer system in which, in the above-described computer system, the generation instruction information setting means changes the parameter values ​​based on the user information of the user (for example, a probability change amount ΔP calculated by a function f that takes the value of the reference individual information in Figure 13 as a variable, the change of probabilities in steps S26 and S30 using the probability change amount ΔP set in step S24 in Figure 18, and steps S32 and S34 using the changed probabilities).

[0031] According to the 14th invention, a computer system can set generation instruction information using a random decision method based on user information.

[0032] The fifteenth invention is a computer system in which, in the above-described computer system, the content is a game, the user information includes at least information indicating the progress of the game, and the generation instruction information setting means changes the parameter values ​​based on the information indicating the progress.

[0033] According to the 15th invention, a computer system can set generation instruction information using a random decision method that is based at least on information indicating the progress of the game.

[0034] The sixteenth invention is a computer system in which, in the above-described computer system, the user information includes a plurality of individual pieces of information, and the generation instruction information setting means selects from the plurality of individual pieces of information to be used for changing the parameter value (for example, step S24 in Figure 18).

[0035] According to the 16th invention, a computer system can set generation instruction information using a random decision method based on individual information selected from a plurality of individual pieces of information.

[0036] The 17th invention is a computer system in which, in the above-described computer system, the generation unit generates the content elements using generation AI (Artificial Intelligence), and the evaluation receiving means (for example, the evaluation receiving unit 234 in Figure 24, step S220 in Figure 26) receives the user's evaluation of the content elements acquired by the acquisition control means, and the generation AI performs learning processing based on the received evaluation (for example, the additional learning control unit 236 in Figure 24, step S222 in Figure 26).

[0037] According to the 17th invention, the computer system can receive user evaluations of the generated content elements and allow the generating AI to undergo additional learning based on the received evaluations.

[0038] The 18th invention is a computer system further comprising the above-described computer system, evaluation receiving means (for example, evaluation receiving unit 234 in Figure 24, step S220 in Figure 26) for receiving the user's evaluation of the content element acquired by the acquisition control means, and reacquisition execution control means (for example, reacquisition execution control unit 238 in Figure 24, steps S224 to S234 in Figure 26) for causing the setting of the generation instruction information by the generation instruction information setting means and the acquisition of the content element by the acquisition control means to be executed again when the received evaluation satisfies predetermined reacquisition conditions.

[0039] According to the 18th invention, the computer system can receive user evaluations of generated content elements and re-obtain a replacement for the content element based on the received evaluations.

[0040] The 19th invention is a program for causing a computer system to perform control over providing content to a user, comprising: a generation instruction information setting means for setting generation instruction information by determining, using a random determination method, part or all of the information of the generation instruction information to be given to a generation unit that generates given content elements to be used in the content based on given generation instruction information; an acquisition control means for performing control over providing the generation instruction information set by the generation instruction information setting means to the generation unit to acquire the content elements; and a provision control means for performing control over providing the content using the content elements acquired by the acquisition control means (for example, the server program 501 in Figure 14 and the content provision program 505 in Figure 22).

[0041] According to the 19th invention, it is possible to realize a program that can enable a computer system to perform the same functions as the first invention. [Brief explanation of the drawing]

[0042] [Figure 1] A system configuration diagram showing an example of a content delivery system. [Figure 2] A diagram illustrating the opportunities for adding content elements. [Figure 3] A diagram showing an example of a template for generation instruction information. [Figure 4] A diagram illustrating the overall structure of the component definition data. [Figure 5] A diagram showing an example of the data structure of component definition data. [Figure 6] A diagram showing an example of the data structure of component definition data. [Figure 7] A diagram showing an example of the data structure of component definition data. [Figure 8] A diagram showing an example of the data structure of component definition data. [Figure 9] A diagram showing an example of the data structure of component definition data. [Figure 10]A diagram showing an example of the data structure of component definition data. [Figure 11] A diagram illustrating the quality of content elements. [Figure 12] A diagram illustrating the first adjustment function related to the assignment of content elements. [Figure 13] A diagram illustrating the second adjustment function related to the assignment of content elements. [Figure 14] A diagram showing examples of programs and data stored by a server system. [Figure 15] A diagram showing an example of the functional configuration of the server processing unit. [Figure 16] A diagram showing an example of the data structure of user information. [Figure 17] A diagram showing an example of the data structure for granting opportunity control data. [Figure 18] A flowchart illustrating the processing flow related to granting opportunities. [Figure 19] Flowchart continuing from Figure 18. [Figure 20] A flowchart illustrating the flow of the component configuration process. [Figure 21] A flowchart to explain the process for determining rare items. [Figure 22] A diagram illustrating a modified example. [Figure 23] A diagram illustrating a modified example. [Figure 24] A diagram illustrating a modified example. [Figure 25] A diagram illustrating a modified example. [Figure 26] A diagram illustrating a modified example. [Modes for carrying out the invention]

[0043] Examples of embodiments of the present invention will be described below, but it goes without saying that the embodiments to which the present invention can be applied are not limited to the following embodiments.

[0044] Figure 1 is a system configuration diagram showing an example of the configuration of a content provision system according to this embodiment. The content provision system 1000 is a computer system that provides content offering virtual experiences in a virtual space, such as video games, virtual activities, and shopping, to multiple registered users. In other words, the content provision system 1000 provides virtual experiences to users. Hereafter, the content provision system 1000 will be described as providing video games.

[0045] The content provision system 1000 is a computer system that includes a server system 1100 and user terminals 1500 for each user, all connected via a network 99 for data communication.

[0046] Network 99 refers to a communication path capable of data communication. In other words, Network 99 includes not only LANs (Local Area Networks) using dedicated lines (dedicated cables) or Ethernet (registered trademark) for direct connections, but also telephone communication networks, cable networks, and the Internet.

[0047] The server system 1100 has a control board 1150 mounted on the main unit 1101. The control board 1150 is equipped with various microprocessors such as a CPU (Central Processing Unit) 1151, a GPU (Graphics Processing Unit), and a DSP (Digital Signal Processor), various IC memories 1152 such as VRAM, RAM, and ROM, and a communication device 1153. Some or all of the functions mounted on the control board 1150 may be implemented using an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or a SoC (System on a Chip).

[0048] Although Figure 1 depicts the server system 1100 as a single server device, it may also be implemented using multiple devices. For example, the server system 1100 may be configured with multiple servers, each responsible for a specific function, connected to each other via an internal bus or network 99 to enable data communication.

[0049] The server system 1100 has a server-side generated AI 3. Server-side generated AI3 is implemented using a machine learning-based AI model on hardware employing a multi-core architecture (for example, a group of GPUs and memory, a group of AI chips, etc.). Note that the hardware for server-side generated AI3 is not limited to that of server system 1100.

[0050] The server-side generation AI3 may be implemented as a single multimodal generation AI model, or it may be configured as a group of AIs having multiple type-specific generation AIs 4 (4a, 4b, ...) prepared for different types of data to be generated. In either configuration, the server-side generation AI3 is a generation unit that generates content elements.

[0051] User terminal 1500 serves as a Man-Machine Interface (MMIF) for user 2 to play content as a player. For example, if the content to be played is a game, user terminal 1500 functions as a gameplay terminal for user 2, who is the player. In actual operation, it is common for multiple user terminals 1500 to communicate with the server system 1100 simultaneously.

[0052] The user terminal 1500 is a computer system that can connect to the network 99, such as a personal computer, smartphone, wearable computer, portable game console, home game console, or tablet computer.

[0053] The user terminal 1500 is a computer equipped with an operation input device, an image display device, a communication device, and a control board 1550 for performing calculations. Examples of operation input devices include a touch panel 1506, a keyboard, a game controller, and a mouse. Examples of image display devices include a touch panel 1506, a head-mounted display, and a glasses-type display.

[0054] The control board 1550 is equipped with a CPU 1551, various microprocessors such as a GPU and DSP, various IC memories 1552 such as VRAM, RAM, and ROM, and a communication module 1553 that connects to the network 99. These elements mounted on the control board 1550 are electrically connected via bus circuits and the like, enabling data reading and writing, and signal transmission and reception. Part or all of the control board 1550 may be an ASIC, FPGA, or SoC.

[0055] The control board 1550 stores programs and various data necessary to realize the functions of the user terminal 1500 in the IC memory 1552. The user terminal 1500 realizes the functions of a play terminal for playing content by executing a predetermined application program.

[0056] The user terminal 1500 has an AI 5 generated on the terminal side. The terminal-generated AI5 is implemented using a machine learning-based AI model on hardware employing a multi-core architecture (e.g., GPU clusters, memory, AI chip clusters, etc.). The hardware for the terminal-generated AI5 is not limited to the user terminal 1500.

[0057] The terminal-side generation AI 5 may be implemented as a single multimodal generation AI model, or it may be configured as a group having multiple type-specific generation AIs 6 (6a, 6b, ...) prepared for each type of data to be generated. In either configuration, the terminal-side generation AI 5 is a generation unit that generates content elements.

[0058] Figure 2 illustrates the opportunity to grant users content elements that can be used within the game. The games provided by the content provision system 1000 offer various reward opportunities. These opportunities include, for example, item drops that occur when defeating enemy NPCs (Non-Player Characters), clear bonuses that occur when clearing game stages, and virtual lotteries. Figure 2 depicts a virtual lottery machine 7 as a representative example of a reward opportunity.

[0059] When an opportunity to grant content arises, the server system 1100 creates generation instruction information 10 and provides it to the server-side generation AI 3 to generate data for the content element 8 to be granted to the user.

[0060] The generation instruction information 10 in Figure 2 is an example where the content element is "sword," and it includes natural language and various data attached as reference information (such as a sample specifying the data format of the generated result, and reference image data). The generation instruction information 10 may also be written using something other than natural language (for example, a dedicated prompt).

[0061] The generation instruction information 10 is created by filling in the blanks in templates provided for each category of content element. The categories of content elements are set as appropriate according to the game content and game rules. For example, attack equipment, defensive equipment, support tools, characters (companions, followers, etc.), attribute assignment to characters, skill assignment to characters, etc.

[0062] Figure 3 shows an example of template 12 for content elements with the classification "Content Element Classification = Attack Equipment, Attack Equipment Classification = Sword". The generation conditions of the generation instruction information 10 include various items that specify what the content element is and what its content and specifications are. These various items will be called the "constituent elements" of the generation instruction information 10. The dashed white rectangles represent the parts where the specific content of the constituent elements is set and correspond to blank spaces.

[0063] Specifically, for example, classification, type, and category are constituent elements that identify "what that content element is," and are "subject identification information" for identifying the subject of content element 8. If content element 8 is a tool or object such as a weapon, specifications for each part, color scheme, etc., may also be included in subject identification information.

[0064] More specifically, for example, the types and values ​​of ability parameters such as granted attributes, granted skills, complementary skills, skill levels, attack power, and durability are components that identify "what kind of content and specifications that content element has." This is "setting value specification information" for specifying the setting parameter values ​​related to the subject of content element 8.

[0065] Depending on the game's world setting and rules, it may also be possible to include attributes, skills, and complementary skills as part of the character identification information.

[0066] Figure 4 is a diagram illustrating the overall structure of the component definition data 740, which defines the options for the specific content of the components of the generation instruction information 10. The white rectangles in Figure 4 are symbols representing one component definition data 740. One component definition data 740 is prepared for each component of the generation instruction information 10 and defines various information related to the setting options.

[0067] The component definition data 740 constitutes multiple groups. These groups are formed according to the type of rarity initially set for the content element, and the component definition data 740 within the same group have a hierarchical structure. The left side of Figure 4 indicates a higher layer, and the right side indicates a lower layer, with the straight lines connecting the symbols indicating the correspondence between them.

[0068] In creating the generation instruction information 10, the component definition data 740 for the top layer (first layer) component is referenced, and one of the setting options specified in the definition data is selected to set the content of that component. Then, for the next layer (second layer) component specified in the selected setting option, the component definition data 740 is referenced, and one of the setting options specified in the definition data is selected to set the content of that component. This process is repeated until all next layer components have been specified.

[0069] In other words, the creation of the generation instruction information 10 is achieved by gradually descending to the lower levels according to the hierarchical structure and specifically setting the content of the constituent elements.

[0070] Figure 5 shows an example of component definition data 740 (740a) that defines the component "Classification of content elements," which is one of the component definition data 740 of the first layer (top layer).

[0071] Figure 6 shows an example of the data structure of the second-layer component definition data 740 (740b) specified as the next-layer component when "Attack Equipment" is selected and set in "Classification of Content Elements" in Figure 5. It defines the setting options for what type of attack equipment to generate.

[0072] Figure 7 shows an example of the data structure of the third-level component definition data 740 (740c) specified as the next-level component when "Sword" is selected and set in "Attack Equipment" in Figure 6. It defines the setting options for what type of sword to generate.

[0073] Figure 8 shows an example of component definition data 740 (740d) that defines the component "assigned skill," which is one of the component definition data 740 of the first layer (top layer).

[0074] Figure 9 shows an example of the data structure of the second layer component definition data 740 (740e) specified as the next layer component when "Radar" is selected and set in "Assigned Skills" in Figure 8. It defines the setting options for what type of attack equipment to generate.

[0075] Figure 10 shows an example of component definition data 740 (740f) that defines the component "design theme," which is one of the component definition data 740 of the first layer (top layer).

[0076] As shown in Figures 5 to 10, the component definition data 740 commonly includes a component ID 741, an application condition 742, a setting option 743, a probability 744, and a next-level component type 745.

[0077] The application condition 742 stores one of the types of rarity and indicates which of the groups of component definition data 740 shown in Figure 4 it belongs to.

[0078] The setting options 743 are choices for the settings of the component in question. The setting options 743 may include options equivalent to "Not set" or "Unspecified," indicating that the generating AI may set them randomly. For example, the component definition data 740 (740a) in Figure 5 shows the setting options 743 for the subject of the generated content element 8.

[0079] In this embodiment, the content provided by the content provision system 1000 is exemplified as a game, so the types and number of setting options 743 should be set appropriately according to the game genre, game world settings, game rules, etc. For example, in the case of an action RPG, setting options 743 may include items such as attack equipment, defensive equipment, and support tools, as well as characters that can be recruited as allies. In addition, the right to grant attributes and skills to characters, music such as sound effects and background music, game stages, etc., may also be set.

[0080] The probability 744 is provided for each type of setting option 743 and represents the probability applied in the lottery process to determine which setting option 743 to select. In other words, probability 744 is a parameter value related to a random decision method.

[0081] The next-level component type 745 is provided for each type of setting option 743, and specifies which component to configure next when the setting option 743 for that component is selected. There may be one or more next-level component types 745 configured, or there may be no configuration at all. In cases where no next-level component type 745 is configured, the setting option 743 for that component is the lowest level and the most detailed setting in the content element's specification.

[0082] The component definition data 740 may include the specification of additional items to be set in relation to the selection of setting options 743, or the specification of reference information for the generation instruction information 10. For example, the component definition data 740(740a) in Figure 5 and the component definition data 740(740d) in Figure 8 have an additional item specification, the initial setting of the capability parameter 746, for each type of setting option 743. This setting specifies the type and initial value (not shown) of the setting parameter that defines the capability and effect of the content element of the corresponding setting option 743.

[0083] For example, the component definition data 740 (740b) in Figure 6 has a data format reference 747 for each type of setting option 743, which specifies the reference information for the generation instruction information 10. It stores reference data for specifying what data format the server-side generation AI 3 will output the generation results in.

[0084] For example, the component definition data 740 (740f) in Figure 10 has a reference image list 748 for each type of setting option 743, which is a reference information for the generation instruction information 10. The reference image list 748 specifies the type of image data to be used as reference data to help the generation AI understand the content indicated by the setting option 743.

[0085] Returning to Figure 2, let's reconsider having the server-side generation AI 3 generate content element 8. When a generation AI is made to generate something, the result is strongly influenced by its learned content. Therefore, even if the generation instructions are the same, generating multiple times will result in some variation. However, the variation is small, and from the user's perspective, it falls within the category of "more or less the same." In other words, from a game developer's point of view, simply providing the same generation instructions is not enough to obtain sufficiently random generation results from the generation AI.

[0086] Therefore, the server system 1100 determines some or all of the information of the components of the generation instruction information 10 by a lottery process that applies a lottery probability based on probability 744. In other words, the generation instruction information 10 is set using a random determination method. Thus, the content elements for each granting opportunity will be generated as various classifications and specifications that go beyond the category of "similar." In other words, sufficient randomness is ensured in the content elements granted to each granting opportunity.

[0087] Figure 11 is a diagram illustrating the difference in quality when the assigned content element 8 is a sword, which is an attack equipment. When we focus on the "quality" of content element 8 from the user's perspective—whether it's good or bad for the user receiving it—the quality varies. For example, even if the content element is a sword, which is an attack equipment, differences in rarity, sword classification, and granted skills will result in the generation and granting of content elements with various levels of quality.

[0088] "Quality" is generally rated higher for items with higher rarity. Furthermore, in the component definition data 740 for the second layer and beyond, the setting options 743 are registered earlier for options that result in a higher quality rating for the generated content element, while the probability 744 is set to be registered lower. In other words, content elements with more setting options 743 registered earlier will be rated higher in quality.

[0089] As mentioned above, while it is possible to ensure randomness while using a generation AI for the granting opportunities, if it were completely random, it is possible that even with many granting opportunities, no high-quality content elements would be granted even once. A "high-quality content element" is a content element that meets predetermined high-quality conditions, and may be, for example, a content element with a rarity of "A" or "S" among the components of the generation instruction information 10. In addition, the setting of a predetermined granting skill with high rarity may also be included as a high-quality condition.

[0090] Figure 12 is a diagram illustrating the first adjustment function related to the assignment of content elements.

[0091] If there are many opportunities to award items but no high-quality content elements are awarded even once, it diminishes the appeal of the awarding opportunities. Therefore, to avoid such a situation, the server system 1100 adjusts so that high-quality content elements are awarded at least once in a "consecutive awarding opportunity". Here, "consecutive awarding opportunities" may refer to, for example, a situation where, if one awarding opportunity is a consecutive lottery using the virtual lottery machine 7 (see Figure 2), the number of consecutive lotteries meets a predetermined standard number (for example, 10 times). Alternatively, it may refer to past awarding opportunities in a predetermined past period and the current awarding opportunity.

[0092] The adjustment function can be implemented, for example, as follows: As shown in the example in Figure 12, if one grant opportunity involves 10 consecutive draws using the virtual lottery machine 7, all content element candidates 9 (9a, 9b, ...) for all 10 draws are determined before the grant is executed, and it is determined whether any high-quality content is included among the content element candidates 9.

[0093] If none are included, this can be achieved by replacing one or more content element candidates 9 with high-quality content. In the example in Figure 12, the fourth content element candidate 9(9d) was initially "B rare," but it has been replaced with content element candidate 9(9s), which is "S rare."

[0094] Alternatively, for a randomly determined number of times out of 10, a high rarity level that is considered superior content could be forcibly set, or the probability 744 applied to setting option 743, which increases the quality level, could be set to "100%" for that time only.

[0095] If the adjustment function is to be implemented in such a way that the probability of receiving a high-quality content element at least once is higher than usual, then it could be implemented by significantly increasing the probability of receiving a high-rarity item that is considered high-quality content over a randomly determined number of turns.

[0096] From the perspective of the "mediocrity" of the assigned content elements 8, this first adjustment function can be rephrased as the server system 1100 setting the generation instruction information 10 in such a way that it reduces the probability that all of the content elements 8 will satisfy the predetermined mediocre conditions. Here, "mediocrity conditions" refer to conditions under which the quality is rated low. Specifically, this means that the settings of the constituent elements include the content element 8 having a relatively low rarity, or that there are many setting options 743 with a relatively low probability of 744.

[0097] Figure 13 is a diagram illustrating the second adjustment function related to the assignment of content elements. The server system 1100 adjusts the system so that the more opportunities an assignment is made for a user, the more likely it is to generate high-quality content elements, based on the user information registered and managed for that user.

[0098] User information includes multiple individual pieces of information. These individual pieces of information may include, for example, the user's age, the number of days since registering as a player, the number of times they have played, their ranking in terms of play performance, their player level, their game progress, and the total number of times and total amount of in-game purchases have been made. Other information may be set as appropriate depending on the game genre and game rules.

[0099] Specifically, the server system 1100 selects one of the individual pieces of user information to be used as "reference individual information" for adjustment. Then, it calculates the probability change amount ΔP by substituting the value indicated by the reference individual information (for example, the level value if it is the player level) and the cumulative number of times the grant opportunity has been granted for the user to be granted (M is 0 or a natural number) into a predetermined function f, which has these values ​​as variables.

[0100] The function f is set such that the probability change amount ΔP increases as the parameter value of the referenced individual information increases and as the cumulative number of times the opportunity is granted M increases. In the graph in Figure 13, the function f is shown as a direct proportional function as an example, but this is not the only way it can be used.

[0101] The server system 1100 then increases the probability 744 of the option that best enhances the quality among the setting options 743, by a probability change amount ΔP. It then lowers the probabilities 744 of the other setting options 743 so that the sum of all probabilities 744 in that definition data does not exceed 100%. In this case, the amount of the reduction may be set according to the degree to which the quality of the setting option 743 is enhanced (order of registration). In the example in Figure 13, the reduction amount for the setting option 743 that enhances the quality second best (e.g., Recovery Skill Lv2) may be set to be smaller than that for the setting option 743 that enhances the quality third best (e.g., Recovery Skill Lv1).

[0102] Furthermore, function f may be provided for each type of individual information that is considered as reference individual information.

[0103] Figure 14 shows an example of programs and data stored by the server system 1100. The server system 1100 stores the server program 501, the distribution client program 503, the game initial setup data 510, and the trained AI model 520 of the server-side generated AI 3 in the IC memory 1152.

[0104] Furthermore, it stores initial setting data 530 for rarity lottery probability, initial setting data 532 for component definition, template data 580 for each template 12 (see Figure 3) of the generation instruction information 10, and user information 600.

[0105] Additionally, it stores play data 700, grant opportunity control data 710, and the current date and time 900. Of course, other data may be stored as appropriate.

[0106] The server program 501 may also include a generation AI program 502 for realizing the functions of server-side generation AI3 and type generation AI4 (see Figure 1). Alternatively, the generation AI program 502 may be stored separately.

[0107] The server system 1100 performs the server program 501 and calculates the results on the CPU 1151, thereby realizing the function of a server processing unit 200s, as shown in Figure 15.

[0108] The server processing unit 200s performs various controls related to content provision. Specifically, the server processing unit 200s includes a user information management unit 202, a billing control unit 204, a generation unit 208, a generation instruction information setting unit 210, an acquisition control unit 220, a provision control unit 230, and a timing unit 290.

[0109] The User Information Management Unit 202 executes the prescribed user registration procedure and controls the registration and management of user information 600 for each user 2.

[0110] The billing control unit 204 processes payments related to content. For example, it processes payments for playing content and for purchasing content elements. Once the billing control unit 204 has processed a payment, it instructs the user information management unit 202 to update the billing information contained in the user information 600.

[0111] The generation unit 208 generates given content elements to be used in the content based on the given generation instruction information 10. The generation unit 208 is implemented by a generation AI. The server-side generation AI 3 (see Figure 1) is an example of this.

[0112] The generation instruction information setting unit 210 sets the generation instruction information by determining some or all of the information in the generation instruction information 10 using a random determination method.

[0113] Specifically, the generation instruction information setting unit 210 sets the generation instruction information so that the probability of the content element 8 being acquired during the N times satisfying predetermined conditions is changed by changing the parameter values ​​related to the determination method when the generation and acquisition of content element 8 is performed N times by setting the generation instruction information 10 N times (a natural number N ≥ 2). The "predetermined conditions" may be, for example, an average condition meaning that the quality (see Figure 11) does not fall below a predetermined standard value.

[0114] More specifically, the generation instruction information setting unit 210 sets the generation instruction information in such a way that it changes the likelihood of satisfying a predetermined condition by making the parameter value related to the determination method when setting the generation instruction information 10 for the nth time (where n is a natural number) different from the parameter value when setting the generation instruction information for the (n+1)th time.

[0115] Furthermore, the generation instruction information setting unit 210 sets the generation instruction information 10 N times (a natural number N ≥ 2) to change the probability that the content element 8 obtained during the N executions will satisfy predetermined conditions when the generation and acquisition of content element 8 is performed N times.

[0116] More specifically, the generation instruction information setting unit 210 sets the generation instruction information 10 in such a way that it changes the likelihood that the generation instruction information will satisfy a predetermined condition over a period of N times.

[0117] The acquisition control unit 220 provides the generation instruction information 10 set by the generation instruction information setting unit 210 to the generation unit 208 and performs control to acquire the content element 8 to be added.

[0118] The provision control unit 230 controls the progress of content provision. Specifically, the provision control unit 230 controls the progress of game content and controls the paid sale of content elements such as items within the game. The provision control unit 230 then controls the provision of content using the content elements 8 acquired by the acquisition control unit 220.

[0119] The timing unit 290 manages the current date and time 900, measures the time, etc., using the system clock or the like.

[0120] Returning to Figure 13, the distribution client program 503 is executed by the user terminal 1500, causing the user terminal 1500 to function as a man-machine interface for various control and content provision systems 1000 as a game client.

[0121] User information 600 is prepared for each user 2 who has completed the prescribed registration procedure, and stores various information associated with that user 2. One user information 600 includes, for example, a user account 602 and multiple individual pieces of information associated with the user account, as shown in Figure 16. Individual pieces of information include, for example, play count 604, billing information 606, save data 610, and game progress data 612, player level 614, owned content element data 650, etc., which are included in the save data 610.

[0122] The billing information 606 may include, for example, the billing date and time, the billing purpose, and the billing amount. It may also include billing statistics such as the cumulative billing amount, cumulative number of billings, and billing frequency, which are automatically updated each time a billing is performed.

[0123] The game save data 610 includes game progress data 612 and owned content element data 650 for each content element assigned to the user. Each owned content element data 650 includes the assignment date and time, content element ID, content element classification, component-specific setting data (a list of only the contents set in the blank fields of the template 12 of the generation instruction information 10), and the content data itself. If the owned content was obtained through means other than the assignment opportunity, a predetermined value indicating this shall be set in the component-specific setting data.

[0124] Returning to Figure 14, the play data 700 stores various information related to the provision of the game and is updated by the game progress control unit 206. The play data 700 includes player account information, game progress information, etc. Of course, other data may also be included as appropriate.

[0125] The grant opportunity control data 710 stores various data related to the control of grant opportunities. The grant opportunity control data 710 includes, for example, grant number data 712, rarity lottery probability setting data 714, component definition data 740, and content element candidate data 750, as shown in Figure 17.

[0126] When setting up a component, the component definition data 740 is copied from the initial component definition data 532 (see Figure 14) of that component, modified as appropriate using the probability change amount ΔP (see Figure 13), and then applied.

[0127] Content element candidate data 750 is data used to manage a candidate for a content element before assigning the results generated by server-side AI3 as a content element. Content element candidate data 750 includes a candidate ID, rarity, content element classification, component-specific setting data, and the content element data itself.

[0128] Figures 18 and 19 are flowcharts illustrating the processing flow related to grant opportunities performed by the server system 1100.

[0129] As shown in Figure 18, the server system 1100 first determines the number of items to be awarded in this award opportunity (step S10), and then initializes the rarity lottery probability setting data 714 and the component definition data 740 (step S12). Specifically, it copies the rarity lottery probability initial setting data 530 to become the rarity lottery probability setting data 714, and copies the component definition initial setting data 532 to become the component definition data 740.

[0130] Next, the server system 1100 executes loop A a number of times equal to the number set in step S10 (from step S20 to step S86 in Figure 19).

[0131] In loop A, the server system 1100 creates and initializes new content element candidate data 750 (step S22). In the initialization state, the content element classification, component-specific setting data, and content data body are each set to initial values ​​indicating undetermined.

[0132] Next, the server system 1100 sets the probability change amount ΔP based on the user information (step S24). Specifically, in step S24, the server system 1100 refers to the content element data 650 owned by the user to be granted and counts the number of content elements granted during a predetermined period in the past (for example, within the past month from the start of the game, etc.) (past grant count). The number of times loop A is executed is added to this count to get the cumulative grant count M. Then, the probability change amount ΔP is calculated using a predetermined function f with the referenced individual information and the cumulative grant count M as variables.

[0133] Next, the server system 1100 changes the draw probabilities for each rarity in the rarity draw (step S26). Specifically, it changes the draw probability for the highest rarity to be increased by a probability change amount ΔP, and adjusts the draw probabilities for the other rarities so that the sum of the draw probabilities for all rarities does not exceed 100%.

[0134] Next, the server system 1100 modifies the probability 744 of the component definition data 740 (step S26). Specifically, the server system 1100 counts the number of times the target user has been granted an award in the past and modifies the probability 744 of each of the setting options 743 of the component definition data 740 to increase the probability 744 of the option that increases the quality level, based on the number of times it has been granted in the past and the probability change amount ΔP (see Figure 13).

[0135] The changes in steps S26 and S30 mean that eligible users will gradually gain better content elements the more opportunities they have to obtain them.

[0136] Here, we focus on the setting of the generation instruction information 10 for each number of grants n in this grant opportunity. Then, due to the changes in steps S26 and S30, the parameter value (lottery probability) related to the method of determining the content of the constituent elements will be different for the setting of the generation instruction information 10 for the nth time and the setting of the generation instruction information 10 for the (n+1)th time. Therefore, the possibility that the content elements 8 generated in this grant opportunity will satisfy the predetermined conditions is changed. Specifically, the possibility that all of the generated content elements 8 will satisfy the predetermined ordinary conditions is reduced.

[0137] Furthermore, as mentioned above, the probability change amount ΔP is calculated by the function f (see Figure 13), so the probability change amount ΔP has a positive relationship with the cumulative number of times the recipient has received an award opportunity M in the past predetermined period. Therefore, the changes in steps S26 and S30 reduce the probability that all generated content elements 8 will satisfy the predetermined ordinary conditions, not only in the current award opportunity but also when considering past award opportunities. The probability of satisfying the ordinary conditions decreases as the number of past award opportunities increases. Conversely, this makes it more likely that a high-quality content element 8 will be awarded somewhere in the current award opportunity based on past award opportunities.

[0138] Next, the server system 1100 performs a lottery process using the lottery probabilities indicated by the rarity lottery probability setting data 714 to determine the rarity (step S32), and then performs a component setting process (step S34).

[0139] Figure 20 is a flowchart illustrating the flow of the component setting process. In the component setting process, the server system 1100 first sets the component to be set, which indicates which component definition data 740 to refer to, to a predetermined value, namely the component ID 741 of the "Classification of Content Elements" in the first layer (see Figure 5) (step S50).

[0140] The server system 1100 then refers to the component definition data 740 of the component to be configured. It then performs a lottery process applying the probability 744 of each configuration option 743 to select and configure one of the configuration options 743 (step S52). The result of the selection and configuration is stored as component-specific configuration data in the content element candidate data 750.

[0141] Next, the server system 1100 confirms the specification of additional items to be set in relation to the selection of setting option 743 (for example, the specification of setting parameters such as the initial ability parameter settings 746 in Figures 5 and 8) (step S54). If any are specified (YES in step S54), they are set (step S56).

[0142] Next, the server system 1100 determines whether there are any unconfigured next-layer components (step S58). Specifically, the server system 1100 determines whether there are settings for the next-layer component type 745 corresponding to the setting option 743 selected in step S52 in the component-specific setting data of the content element candidate data 750 created in step S22. If there are any next-layer component types 745 for which no settings exist, the system makes a positive determination.

[0143] Then, if there are any unconfigured next-layer components (YES in step S58), the server system 1100 sets those unconfigured next-layer components as the component to be configured (step S60), and returns to step S52.

[0144] As mentioned above, the next-level component type 745 is a lower-level component and indicates a more detailed specification. Therefore, with each step from S50 to S60, the components are defined in more detail.

[0145] When there are no more unset next-level components (NO in step S58), the server system 1100 then performs a lottery process applying probability 744 to select and set an assigned attribute (step S70). The type of assigned attribute may be randomly selected from a predetermined range of types. Alternatively, the component definition data 740 may include data specifying the types of attributes that can be set for each setting option 743, and the attribute may be selected from the specified types. The attribute types may include "no attribute".

[0146] Next, the server system 1100 refers to the component definition data 740 for component ID 741, which is "granted skill", and performs a lottery process applying probability 744 to select and set the granted skill (step S72; see Figure 8). The setting options 743 include the option "none", and if this is selected, no granted skill is set.

[0147] If a skill is assigned (YES in step S74), the server system 1100 further refers to the component definition data 740 of the complementary skill according to the specification of the next-level component type 745 of the selected setting option 743. Then, it selects and assigns complementary skills that are compatible or incompatible with the assigned skill by performing a lottery process with probability 744 applied (step S76).

[0148] Next, the server system 1100 refers to the component definition data 740 for component ID 741, which is "design theme", and performs a lottery process applying probability 744 to select and set one of the design themes (step S78). This completes the configuration of the components of the generation instruction information 10.

[0149] Moving to Figure 19, the server system 1100 provides the generation instruction information 10 to the server-side generation AI 3 to generate content element candidates (step S86).

[0150] Next, the server system 1100 determines the similarity between the newly generated content element candidates in Loop A and each of the content owned by the target user (step S88). Then, it determines the similarity between the newly generated content element candidates in Loop A and the content element candidates already generated and acquired in Loop A (step S90). The similarity is determined by similarity in, for example, the settings of the components, their appearance, color schemes, and the types and values ​​of the capability parameters. In other words, it can be said that the similarity is determined by the approximation of the settings and instructions of the components.

[0151] The server system 1100 then determines whether there are any highly similar owned content elements or content element candidates among the content elements owned by the user to be assigned and the generated content element candidates (step S92). "Highly similar" means that the similarity with the content element candidate newly generated in loop A satisfies predetermined approximation conditions.

[0152] If there are highly similar owned content elements or content element candidates (YES in step S92), the server system 1100 discards the newly generated content element candidates in loop A (step S94).

[0153] The server system 1100 then references the component-specific setting data of the highly similar owned content element data 650 and the highly similar content element candidate data 750. The server system 1100 then makes a change to the component definition data 740 that reduces the probability 744 corresponding to the setting option 743 set for each component in the referenced data (step S96), and returns to step S34.

[0154] Steps S88 to S96 ensure that the recipient user is not assigned content elements that they already own or content elements that are highly similar to content element candidates already generated in Loop A. When the component settings are configured again in step S96, it becomes less likely that content element candidates highly similar to owned content will be generated. In other words, the probability that all of the content elements 8 generated and assigned to the user during this assignment opportunity will satisfy the predetermined approximation conditions is reduced. It can also be said that the probability that all of the components of the generation instruction information 10 related to this assignment opportunity will satisfy the predetermined instruction content approximation conditions is reduced.

[0155] If there are no content elements that are highly similar to the newly generated content element candidate among the content elements owned by the user to be assigned and the content element candidates already generated in Loop A (NO in step S92), Loop A is terminated (step S96).

[0156] Next, if the number of items awarded this time is equal to or greater than a predetermined threshold value (for example, "10") (YES in step S110), the server system 1100 executes the rare item confirmation process (step S112).

[0157] Figure 21 is a flowchart illustrating the process for determining the rarity of an item. In the rare item determination process, the server system 1100 first determines the quality of each content element candidate (step S130; see Figure 11).

[0158] If at least one determined quality level is found to be a "good content element candidate" that meets the predetermined quality conditions (YES in step S132), the rarity confirmation process ends. The "quality conditions" may be defined as a content element candidate with a rarity of "A" or "S". In addition, the condition of having a predetermined skill with high rarity may be included as a quality condition.

[0159] However, if there are no content element candidates that meet the excellent criteria (NO in step S132), the server system 1100 sets the number of replacements to a value less than or equal to the number of assigned elements (step S134). Then, it discards the number of replacements from the current content element candidates (step S136) and executes loop B a number of times equal to the number of replacements (steps S150 to S158).

[0160] In loop B, the server system 1100 changes the probability of drawing rarity so that it increases with higher rarity (step S152), and determines the rarity by lottery (step S154). Then, it generates additional content element candidates (step S156). In other words, it executes the processes corresponding to steps S34 to S94 and terminates loop B (step S158).

[0161] The rarity confirmation process ensures that even if none of the initially generated content element candidates meet the criteria for a "high-quality content element," at least one of them can be designated as such. In other words, it is effectively guaranteed that the recipient user will receive a high-quality content element.

[0162] Returning to Figure 19, the server system 1100 executes control for displaying the grant animation and executes control for granting the prepared content element candidates to the target user (step S260), and then terminates the series of processes.

[0163] As described above, this embodiment provides a technology that allows for randomness in the content elements to be provided when the content generation unit generates content elements to be given to the user. The content provision system 1000 can have the generation AI generate content elements 8 while maintaining randomness by setting the generation instruction information 10 given to the server-side generation AI 3, which is capable of generating content elements 8, to have randomness. Then, the generated content elements 8 can be provided to the target users.

[0164] [Variation] Although examples of embodiments to which the present invention is applied have been described above, the forms to which the present invention can be applied are not limited to the above forms, and components can be added, omitted, or modified as appropriate.

[0165] (Variation 1) For example, the content provision system 1000 may be implemented using a P2P (Peer to Peer) architecture with multiple user terminals 1500. In this case, programs and data corresponding to the functional division are stored in the user terminals 1500, and the functions corresponding to the server processing unit 200s in the above embodiment are distributed and implemented by the user terminals 1500, which act as P2P nodes. The same effects as in the above embodiment can be obtained with this configuration as well.

[0166] (Variation 2) Furthermore, in the above embodiment, the content provision system 1000 may be implemented not as a client-server type, but as a single computer system that was designated as the user terminal 1500 in the above embodiment.

[0167] Specifically, as shown in Figure 22, for example, the user terminal 1500 stores all the data stored by the server system 1100 of the above embodiment (see Figure 14). However, instead of the server program 501 and the distribution client program 503, a content provision program 505 is provided as an application program for the user terminal 1500.

[0168] In the above embodiment, the content provision program 505 implements all of the functional parts of the server system 1100 (see Figure 15) as a terminal processing unit 200t on the user terminal 1500, as shown in Figure 23, for example. In this modified example, the content provision program is executed on the user terminal 1500. Then, the processing flow described in the above embodiment (see Figures 18 to 21) can be read as the execution entity being the user terminal 1500 instead of the server system 1100.

[0169] (Variation 3) Furthermore, although the content provision system 1000 is a client-server type, it is also possible to configure it so that the user terminal 1500 functions as a terminal processing unit 200t (see Figure 23) through the distribution client program 503 (see Figure 14).

[0170] (Variation 4) The above embodiments and modifications may be modified to include (1) a function to present generation instruction information 10 to the user and accept changes, and (2) a function to accept user evaluations of content elements 8. Furthermore, the following may be added: (3) a function to further train the server-side generation AI 3 based on the evaluation, and (4) a function to reacquire content elements 8 based on the evaluation.

[0171] Specifically, in addition to the above embodiment, as shown in Figure 24, the server system 1100 has a server processing unit 200s which includes a generation instruction information presentation control unit 232, an evaluation reception unit 234, an additional learning control unit 236, and a reacquisition execution control unit 238.

[0172] The generation instruction information presentation control unit 232 controls the presentation of some or all of the generation instruction information 10 set by the generation instruction information setting unit 210 to the user. In this configuration, the generation instruction information setting unit 210 controls the updating of some or all of the information presented by the generation instruction information presentation control unit 232 based on user operations.

[0173] The evaluation reception unit 234 receives user evaluations of the content elements 8 acquired by the acquisition control unit 220, specifically the content elements 8 assigned to the target user.

[0174] The additional learning control unit 236 controls the additional learning of the generation unit 208. Specifically, it temporarily stores the evaluations received by the evaluation reception unit 234 and performs the learning process of the server-side generated AI 3 based on those evaluations at a given timing. If the server-side generated AI 3 is managed and controlled by an external device, the additional learning control unit 236 provides the received evaluations to the external device as additional learning material. The provision procedure may be carried out using an API (Application Programming Interface) separately provided for the external device.

[0175] The reacquisition execution control unit 238, when the evaluation received by the evaluation reception unit 234 satisfies predetermined reacquisition conditions, causes the generation instruction information setting unit 210 to set the generation instruction information 10 and the acquisition control unit 220 to acquire the content element 8 again.

[0176] In this modified example, the server system 1100 executes the generation instruction information change acceptance process shown in Figure 25 before executing step S86.

[0177] In the process of receiving a change in the generation instruction information, the server system 1100 determines whether the user to be granted the information is eligible (step S200).

[0178] Specifically, the system refers to the user information 600 of the user to whom the grant is granted, and determines whether the specified individual information meets the specified eligibility conditions. For example, if one or more of the following conditions are met, such as "the player level 614 has reached the specified standard," "the information on billing history, such as the cumulative billing amount and billing frequency obtained from the billing information 606, has reached the specified standard value," or "the progress shown in the game progress data 612 has reached the specified standard," the eligibility conditions are considered met.

[0179] If the recipient user is eligible (YES in step S200), the server system 1100 presents part or all of the generation instruction information 10 to the recipient user (step S202). When presenting, the current settings and other setting options 743 are presented as change candidates for each component of the generation instruction information 10, and the user is made able to change the current settings to one of the change candidates. When presenting, the server system 1100 also makes it possible to accept the operation of approving the presented generation instruction information 10.

[0180] Then, if a change operation is performed (YES0 in step S204), the server system 1100 changes the configuration settings of the generated instruction information 10 presented in accordance with that change operation (step S206).

[0181] If the prescribed approval operation is performed (YES in step S208), the server system 1100 terminates the generation instruction information change acceptance process and executes step S86.

[0182] Furthermore, in this modified example, the server system 1100 executes the user evaluation acceptance process shown in Figure 26 before executing step S160.

[0183] In the user evaluation reception process, the server system 1100 presents the candidate content elements to be assigned in step S160 to the target user and accepts user evaluations (e.g., OK / NG, evaluation score, etc.) (step S220). If there are multiple assignments in this assignment opportunity, user evaluations will be accepted for each candidate content element.

[0184] Next, the server system 1100 performs additional training processing on the server-side generated AI3 based on the received evaluation results (step S222).

[0185] Specifically, if the time required for the additional training process is sufficiently short, the additional training process will be executed at this time. If the time required is expected to exceed the time available for this opportunity, the additional training process may be performed by separately storing information such as the evaluation results as material data for additional training. In this case, additional training using the material data for additional training will be performed at a later time.

[0186] Alternatively, the generated AI may not be exclusively for the server system 1100, but may be used as an externally provided service by accessing an external device that can communicate via the network 99. In that case, the additional learning process may consist of creating additional learning material data and sending it to the external device.

[0187] Next, the server system 1100 determines whether the received evaluation result meets the predetermined acceptance criteria (step S224).

[0188] If the passing criteria are not met and the result is a failure, the server system 1100 considers that the predetermined reacquisition conditions have been met (YES in step S224). Then, a description to increase "Temperature," one of the generation conditions, is added to the generation instruction information 10 when the content element candidate was generated (step S230), and loop A is executed (step S232).

[0189] Next, Loop A is executed to retrieve the content element candidates again, and any unsuccessful content element candidates that meet the retrieval conditions are replaced (step S234), and the process proceeds to step S160.

[0190] In addition, while the user evaluation acceptance process accepts evaluations from eligible users, it is also possible to configure the system so that a computer automatically performs the evaluation separately. For example, the generation purpose could be set to "evaluation of the evaluation target," and generation instruction information could be created specifying evaluation criteria as generation conditions, along with data of candidate content elements to be evaluated as reference information. This information could then be provided to the server-side generation AI3, which would then perform the evaluation. [Explanation of Symbols]

[0191] 2…User 3…Server-side generated AI 7…Virtual Lottery Machine 8…Content elements 10…Generation instruction information 12…Template 200s... Server Processing Unit 202...User Information Management Department 204... Billing Control Unit 206...Game Progression Control Unit 208...Generation section 210…Generation instruction information setting section 220... Acquisition Control Unit 230...Providing Control Unit 232...Generation Instruction Information Presentation Control Unit 234... Evaluation Reception Department 236... Additional Learning Control Unit 238...Reacquisition Execution Control Unit 501…Server Program 510...Game initial setup data 520... Pre-trained AI models 530... Initial setting data for rarity lottery probability 532...Component definition initial settings data 580…Template data 600... User Information 612…Game progress data 650…Owned content element data 700... Play data 710... Granting opportunity control data 712...Number of assignments data 714... Rarity draw probability setting data 740...Component definition data 743…Settings Options 744... Probability 745...Types of next-level component elements 746...Initial ability parameter settings 747...Data format reference 748...Reference image list 750...Content element candidate data 1000... Content delivery system 1100... Server System 1500... User terminal

Claims

1. A computer system that controls the provision of content to users, A generation instruction information setting means sets the generation instruction information by determining some or all of the information of the generation instruction information to be given to a generation unit that generates a given content element to be used in the said content based on given generation instruction information, using a random determination method. An acquisition control means that controls the acquisition of content elements by providing the generation instruction information set by the generation instruction information setting means to the generation unit, A provision control means that performs control to provide the content using the content elements acquired by the acquisition control means, A computer system equipped with the following features.

2. The generation instruction information setting means sets the generation instruction information N times (N ≥ 2) so that when the acquisition of the content element by the acquisition control means is performed N times, the probability that the content element acquired during the N times will satisfy a predetermined condition is changed. The computer system according to claim 1.

3. The generation instruction information setting means sets the generation instruction information in such a way that it changes the likelihood of satisfying the predetermined conditions by changing the parameter values ​​related to the determination method. The computer system according to claim 2.

4. The generation instruction information setting means sets the generation instruction information in such a way that it changes the likelihood of satisfying the predetermined conditions by making the parameter value related to the determination method when setting the generation instruction information for the nth time different from the parameter value when setting the generation instruction information for the (n+1)th time. The computer system according to claim 2.

5. The generation instruction information setting means sets the generation instruction information N times (N ≥ 2) so that when the acquisition of the content element by the acquisition control means is performed N times, the probability that the content element acquired during the N times will satisfy a predetermined condition is changed. The computer system according to claim 1.

6. The generation instruction information setting means sets the generation instruction information such that the probability of the generation instruction information satisfying the predetermined conditions over N times changes. The computer system according to claim 5.

7. A generation instruction information presentation control means that controls the presentation of some or all of the generation instruction information set by the generation instruction information setting means to the user. The computer system according to claim 1, further comprising:

8. The generation instruction information setting means controls the updating of part or all of the information presented by the generation instruction information presentation control means based on the user's operation. The computer system according to claim 7.

9. The generation instruction information includes subject identification information for identifying the subject of the content element and setting value specification information for specifying the setting parameter values ​​related to the subject. The generation instruction information setting means sets the generation instruction information by determining the subject identified by the subject identification information and / or the setting parameter value specified by the setting value specification information using the determination method. The computer system according to claim 1.

10. The aforementioned subject identification information is structured in a hierarchical manner, with options for identifying the details of each subject for each type of subject. The computer system according to claim 9.

11. The aforementioned content is a game, The aforementioned entity is at least one of the characters, items, game stages, and music that appear in the game. The computer system according to claim 9.

12. The generation instruction information setting means makes the determination using the determination method based on the acquisition results of the content elements previously acquired by the acquisition control means, and sets the generation instruction information. The computer system according to claim 9.

13. The generation instruction information setting means performs the determination using the determination method, which modifies the parameter value specifying the randomness related to the determination method based on the acquired result. The computer system according to claim 12.

14. The generation instruction information setting means modifies the parameter value based on the user information of the user. The computer system according to claim 13.

15. The aforementioned content is a game, The user information includes at least information indicating the game's progress, The generation instruction information setting means changes the parameter value based on the information indicating the progress. The computer system according to claim 14.

16. The aforementioned user information includes multiple individual pieces of information, The generation instruction information setting means selects from the plurality of individual pieces of information to be used for changing the parameter value. The computer system according to claim 14.

17. The generation unit generates the content elements using artificial intelligence (AI), An evaluation receiving means for receiving user evaluations of the content elements acquired by the acquisition control means, Based on the received evaluation, the AI ​​generation process is performed. The computer system according to claim 1.

18. An evaluation receiving means for receiving user evaluations of the content elements acquired by the acquisition control means, If the received evaluation satisfies predetermined reacquisition conditions, the reacquisition execution control means causes the setting of the generation instruction information by the generation instruction information setting means and the acquisition of the content element by the acquisition control means to be executed again. The computer system according to claim 1, further comprising:

19. A program for causing a computer system to control the provision of content to users, A generation instruction information setting means that sets the generation instruction information by determining some or all of the information of the generation instruction information to be given to a generation unit that generates a given content element to be used in the said content based on given generation instruction information, using a random determination method. Acquisition control means that controls the acquisition of content elements by providing the generation instruction information set by the generation instruction information setting means to the generation unit, A provisioning control means that performs control to provide the content using the content elements acquired by the acquisition control means, A program for causing the aforementioned computer system to function.

Images