Display control device, display control method, and display control program

Abstract

The display control device (100) generates a trained model by learning detection information that is obtained each time a plurality of pieces of content are displayed by the display device (210) and indicates the situation when the display was performed and the length of time the displayed content was viewed. The display control device infers the time when each of the plurality of pieces of content is viewed for each inference pattern that indicates the situation when the display was performed using the trained model, and determines the display ratio of each of the plurality of pieces of content for each inference pattern based on the inference result.

Description

[Technical Field]

[0001] The present disclosure relates to techniques for effectively displaying information on a display medium. [Background technology]

[0002] BACKGROUND ART In recent years, information has been provided in various places such as stores, train stations, airports, and hotels using electronic video media such as digital signage.

[0003] Furthermore, a technique for effectively displaying information by switching display content is known. Patent Document 1 discloses a technique for acquiring information about nearby users from camera images or the like and switching the display content.

[0004] In the technology of Patent Document 1, attributes and environments are linked to display content manually, so the effect of switching display content depends on the skill of the person who links them. Furthermore, when there are multiple display contents to be linked to attributes and environments, the display distribution of the multiple display contents must be adjusted manually. As a result, the effect of switching the display contents depends on the skill of the person making the adjustment. Also, if the display distribution of the multiple display contents is uniform, the effect cannot be sufficiently increased. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Publication No. 2019-159468 Summary of the Invention [Problem to be solved by the invention]

[0006] An object of the present disclosure is to enable the display of multiple pieces of content at a display ratio suited to the situation when the content is displayed. [Means for solving the problem]

[0007] The display control device of the present disclosure includes: a learning unit that learns detection information obtained each time each of a plurality of contents is displayed by the display device, the detection information indicating the situation when the display was performed and the length of time the displayed content was viewed, and generates a learned model; an inference unit that uses the trained model to infer the time that each of the plurality of contents is viewed for each inference pattern that indicates the situation when the display is performed, and determines the display ratio of each of the plurality of contents for each inference pattern based on the inference result; Equipped with. [Effects of the Invention]

[0008] According to the present disclosure, it is possible to determine a display ratio appropriate for each situation when content is displayed, thereby enabling multiple pieces of content to be displayed at a display ratio appropriate for the situation when the content is displayed. [Brief explanation of the drawings]

[0009] [Figure 1] FIG. 1 is a configuration diagram of an information providing system 200 according to a first embodiment. [Figure 2] FIG. 1 is a configuration diagram of a display control device 100 according to a first embodiment. [Figure 3] FIG. 1 is a functional configuration diagram of a display control device 100 according to a first embodiment. [Figure 4] FIG. 2 is a configuration diagram of a display device 210 according to the first embodiment. [Figure 5] 3 is a flowchart of a display control method according to the first embodiment. [Figure 6] 10 is a flowchart of step S110 in the first embodiment. [Figure 7] FIG. 10 is a diagram showing an example of detection information 181 according to the first embodiment. [Figure 8] FIG. 10 is a diagram showing an example of display performance information 182 according to the first embodiment. [Figure 9]FIG. 10 shows an example of content information 183 according to the first embodiment. [Figure 10] FIG. 3 shows an example of training data 184 according to the first embodiment. [Figure 11] 10 is a flowchart of step S120 in the first embodiment. [Figure 12] FIG. 10 is a diagram showing an example of inference pattern information 185 according to the first embodiment. [Figure 13] FIG. 10 is a diagram showing an example of an inference result 186 according to the first embodiment. [Figure 14] FIG. 10 is a diagram showing an example of display ratio information 187 according to the first embodiment. [Figure 15] FIG. 1 is a hardware configuration diagram of a display control device 100 according to a first embodiment. [Figure 16] FIG. 2 is a hardware configuration diagram of a display device 210 according to the first embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0010] In the embodiments and drawings, the same or corresponding elements are denoted by the same reference numerals. The description of elements denoted by the same reference numerals as those already described will be omitted or simplified as appropriate. Arrows in the drawings primarily indicate the flow of data or the flow of processing.

[0011] Embodiment 1 The information providing system 200 will be described with reference to FIGS.

[0012] ***Configuration Description*** The configuration of the information providing system 200 will be described with reference to FIG. The information providing system 200 includes a display device 210 and a display control device 100 . The display device 210 includes a display 211 and a camera 212 .

[0013] The display 211 displays the content. The content may be, for example, an advertisement introducing a product or service, or may be information such as facility information, transportation information, or tourist information. The content displayed on the display 211 is referred to as the "display content." A person who views the displayed content is called a "viewer."

[0014] The camera 212 captures an image in front of the display 211 . The image obtained by shooting is called a "camera image." The camera image is used to detect viewers and their attributes.

[0015] The display device 210 and the display control device 100 communicate with each other via a network 201 . An example of a network 201 is the Internet.

[0016] The information providing system 200 may include multiple display devices 210 located at different locations.

[0017] The configuration of the display control device 100 will be described with reference to FIG. The display control device 100 is a computer that includes hardware such as a processor 101, a memory 102, an auxiliary storage device 103, a communication device 104, and an input / output interface 105. These pieces of hardware are connected to one another via signal lines.

[0018] Examples of computers include personal computers, stationary computers such as server computers, portable computers such as smartphones and tablet computers, microcomputers embedded in devices, and SoCs. SoC is an abbreviation for System on Chip.

[0019] The processor 101 is an IC that performs arithmetic processing and controls other hardware. For example, the processor 101 is a CPU, a DSP, or an FPGA. The processor 101 may be a single processor or a multiprocessor. IC is an abbreviation for Integrated Circuit. CPU is an abbreviation for Central Processing Unit. DSP is an abbreviation for Digital Signal Processor. FPGA is an abbreviation for Field Programmable Gate Array.

[0020] The memory 102 is a volatile or non-volatile storage device. The memory 102 is also called a primary storage device or a main memory. For example, the memory 102 is a RAM. Data stored in the memory 102 is saved in the secondary storage device 103 as needed. RAM is an abbreviation for Random Access Memory.

[0021] The auxiliary storage device 103 is a non-volatile storage device. For example, the auxiliary storage device 103 is a ROM, a HDD, an SSD, a flash memory, or a combination thereof. Data stored in the auxiliary storage device 103 is loaded into the memory 102 as needed. ROM is an abbreviation for Read Only Memory. HDD is an abbreviation for Hard Disk Drive. SSD is an abbreviation for Solid State Drive.

[0022] The communication device 104 is a receiver and a transmitter. For example, the communication device 104 is a communication chip or a NIC. The communication of the display control device 100 is performed using the communication device 104. NIC is an abbreviation for Network Interface Card.

[0023] The input / output interface 105 is a port to which an input device and an output device are connected. For example, the input / output interface 105 is a USB terminal. The input / output of the display control device 100 is performed using the input / output interface 105. USB is an abbreviation for Universal Serial Bus.

[0024] The display control device 100 includes elements such as a learning unit 110 and an inference unit 120. These elements are realized by software.

[0025] The auxiliary storage device 103 stores a display control program for causing the computer to function as the learning unit 110 and the inference unit 120. The display control program is loaded into the memory 102 and executed by the processor 101. The auxiliary storage device 103 also stores an OS. At least a part of the OS is loaded into the memory 102 and executed by the processor 101. The processor 101 executes a display control program while running the OS. OS is an abbreviation for Operating System.

[0026] The data of the display control program (input data, output data, etc.) is stored in the storage unit 190. The auxiliary storage device 103 functions as the storage unit 190. However, a storage device such as the memory 102, a register in the processor 101, or a cache memory in the processor 101 may function as the storage unit 190 instead of or together with the auxiliary storage device 103.

[0027] The display control device 100 may include a plurality of processors that replace the processor 101.

[0028] The display control program can be recorded (stored) in a computer-readable manner on a non-volatile recording medium such as an optical disk or a flash memory. The display control program may be provided as a program product.

[0029] FIG. 3 shows the functional configuration of the display control device 100. The learning unit 110 includes elements such as an acquisition unit 111 , a learning data generation unit 112 , and a learning processing unit 113 . The inference unit 120 includes elements such as an acquisition unit 121, an inference processing unit 122, a display ratio determination unit 123, and an output unit . The storage unit 190 includes a detection information storage unit 191 , a display record storage unit 192 , a content information storage unit 193 , a trained model storage unit 194 , and an inference pattern storage unit 195 .

[0030] The configuration of the display device 210 will be described with reference to FIG. The display device 210 is a computer that includes hardware such as a processor 213, a memory 214, an auxiliary storage device 215, a display 211, a camera 212, and a communication device 216. These pieces of hardware are connected to one another via signal lines.

[0031] The processor 213 is an IC that performs arithmetic processing and controls other hardware. For example, the processor 213 is a CPU. The memory 214 is a volatile or non-volatile storage device. The memory 214 is also called a primary storage device or a main memory. For example, the memory 214 is a RAM. Data stored in the memory 214 is saved in the secondary storage device 215 as needed. The auxiliary storage device 215 is a non-volatile storage device. The auxiliary storage device 215 is also called storage. For example, the auxiliary storage device 215 is a ROM, a HDD, a flash memory, or a combination thereof. Data stored in the auxiliary storage device 215 is loaded into the memory 214 as needed. The display 211 may be a flat display or a curved display. For example, the display 211 may be a liquid crystal display or an LED display. LED is an abbreviation for Light Emission Diode. The camera 212 is an imaging device that takes pictures and outputs the images. The communication device 216 is a receiver and a transmitter. For example, the communication device 216 is a communication chip or a NIC. The communication of the display device 210 is performed using the communication device 216.

[0032] The display device 210 may include a projector instead of or in addition to the display 211 . A projector displays content by projecting it, for example, as a 3D hologram.

[0033] The display device 210 includes an element called a control unit 217. This element is realized by software.

[0034] The auxiliary storage device 215 stores a display program for causing the computer to function as the control unit 217. The display program is loaded into the memory 214 and executed by the processor 213. The auxiliary storage device 215 also stores an OS. At least a portion of the OS is loaded into the memory 214 and executed by the processor 213. The processor 213 executes the display program while running the OS.

[0035] The display program data is stored in the storage unit 218. The auxiliary storage device 215 functions as the storage unit 218. However, storage devices such as the memory 214, a register in the processor 213, and a cache memory in the processor 213 may function as the storage unit 218 instead of the auxiliary storage device 215 or together with the auxiliary storage device 215.

[0036] The display program can be recorded (stored) in a computer-readable manner on a non-volatile recording medium such as an optical disk or a flash memory.

[0037] ***Explanation of Operation*** The operation procedure of the information providing system 200 corresponds to an information providing method. The operation procedure of the information providing system 200 corresponds to a processing procedure by an information providing program. The information providing program includes a display control program and a display program.

[0038] The operation procedure of the display control device 100 corresponds to a display control method, and also corresponds to a processing procedure according to a display control program.

[0039] An overview of the display control method will be described based on FIG. In step S110, the learning unit 110 learns the detection information 181 obtained each time each of the plurality of contents is displayed by the display device 210, and generates a learned model.

[0040] The detection information 181 indicates the circumstances under which the display was performed and the length of time the displayed content was viewed.

[0041] In step S120, the inference unit 120 uses the trained model to infer, for each inference pattern, the time during which each of the multiple pieces of content is viewed. The inference pattern indicates the circumstances under which the display occurs. Then, the inference unit 120 determines the display ratio of each of the plurality of contents for each inference pattern based on the inference result.

[0042] Step S110 will be described in detail with reference to FIG. In step S111, the acquisition unit 111 acquires the detection information 181, the display record information 182, and the content information 183 from the storage unit 190. The detection information 181 is acquired from the detection information storage unit 191 . The display record information 182 is acquired from the display record storage unit 192 . The content information 183 is acquired from the content information storage unit 193 .

[0043] The detection information 181 will be described with reference to FIG. The detection information 181 indicates the display status and the "viewing time." The display status is the status when the content is displayed, and includes information on the date and time, attributes of the viewer, and the like. In FIG. 7, the detection information 181 indicates "person ID", "date and time", "gender", and "age" as display conditions. "Person ID" is an identifier that identifies the viewer. Viewers are detected using camera footage. A "Person ID" is assigned to each detected viewer. The "date and time" is the date and time when the viewer started to view the display content. The "date and time" corresponds to the capture time of the camera image in which the viewer was detected. "Gender" is the gender of the viewer. "Gender" is detected using the camera image in which the viewer is detected. "Age" is the viewer's age (or age category). "Age" is detected using the camera footage in which the viewer is detected. The "viewing time" is the length of time that the viewer views the display content. The "viewing time" corresponds to the length of time that the viewer continues to be detected. The detection information 181 is generated by the display device 210, transmitted from the display device 210, received by the acquisition unit 111, and stored in the storage unit 190, for example.

[0044] The display performance information 182 will be described with reference to FIG. Display record information 182 indicates the display record of each of a plurality of contents. In FIG. 8, the display record information 182 indicates a "content ID," a "display start date and time," a "display end date and time," and a "display time." A "content ID" is an identifier that identifies a piece of content. A "content ID" is assigned to each piece of content. The "display start date and time" is the date and time when the display of the content started. If the content is a video, the "display start date and time" corresponds to the date and time when the playback of the content started. The "display end date and time" is the date and time when the display of the content ended. If the content is a video, the "display end date and time" corresponds to the date and time when the playback of the content ended. "Display time" is the length of time that the content is displayed. "Display time" corresponds to the length of time from the display start date and time to the display end date and time. If the content is a video, "display time" corresponds to the playback time of the content. The display performance information 182 is generated by the display device 210, transmitted from the display device 210, received by the acquisition unit 111, and stored in the storage unit 190, for example.

[0045] The content information 183 will be described with reference to FIG. The content information 183 indicates various information about each of a plurality of contents. In FIG. 9, content information 183 indicates a "content ID," a "content type," a "display time," a "posting start date," and a "posting end date." "Content ID" is an identifier that identifies the content. "Content type" is the type of content. Examples of "content type" are (still) images and videos. "Playback time" is the duration of the video content. The "publication start date" is the date on which information provision by displaying the content began. The "end date of posting" is the date on which the provision of information by displaying the content will end. The content information 183 is, for example, registered in the display device 210, transmitted from the display device 210, received by the acquisition unit 111, and stored in the storage unit 190.

[0046] Returning to FIG. 6, the description continues from step S112. In step S112, the learning data generation unit 112 generates learning data 184 using the information acquired in step S111.

[0047] The training data 184 is data used for training to generate a trained model.

[0048] The learning data generating unit 112 mainly combines the detection information 181 and the display performance information 182 to generate learning data 184 .

[0049] The learning data 184 will be described with reference to FIG. The learning data 184 indicates the display situation and visual recognition information. The display status is the status when the content is displayed, and includes information on the date and time, attributes of the viewer, and the like. In FIG. 10, the learning data 184 indicates "person ID", "date and time", "gender", "age", "time period segment", and "day segment" as display conditions. The "person ID" is an identifier for identifying the viewer. The "person ID" is obtained from the detection information 181. The "date and time" is the date and time when the viewer starts to view the display content. The "date and time" is obtained from the detection information 181. "Gender" is the gender of the viewer. "Gender" is obtained from the detection information 181. “Age” is the age of the viewer. “Age” is obtained from the detection information 181. The "time period segment" is a segment of a time period during which the display content is viewed by a viewer. The time period is segmented into, for example, morning, afternoon, and night. The "time period segment" is determined based on the "date and time." The "day division" is the division of the day on which the display content is viewed by the viewer. Days are divided into weekdays and holidays, for example. The "day division" is determined based on the "date and time."

[0050] The viewing information is information relating to the viewing of the display content by the viewer. In FIG. 10, the learning data 184 indicates "viewing time", "content ID", and "time breakdown" as viewing information. The "viewing time" is the length of time that the viewer views the display content. The "viewing time" is obtained from the detection information 181. The "content ID" is an identifier that identifies the viewed display content. The "content ID" is determined based on the "date and time" and the display performance information 182. The period from the "display start date and time" to the "display end date and time" in the display performance information 182 is called the display time slot. The "content ID" in the display performance information 182 whose display time slot includes the same date and time as the "date and time" of the learning data 184 becomes the "content ID" of the learning data 184. The "time breakdown" is the "viewing time" for each display content. The "time breakdown" is determined based on the "date and time", the "viewing time", and the display performance information 182. In the learning data 184, the time period from the "date and time" to the time when the "viewing time" has elapsed is called the viewing time period. The length of the time period that overlaps between the display time period and the viewing time period is the "time breakdown".

[0051] Further information about the training data 184 will be provided. The learning data 184 is made up of information indicating how long a user viewed a particular content when that content was displayed under what attributes and in what environment. In Figure 10, viewing time is the objective variable. However, it is thought that users tend to view content with long playback times, such as videos, for longer periods of time. Therefore, when there is content with long playback times, it may be impossible to perform fair learning. Therefore, the ratio of viewing time to playback time may be used as the objective variable. In other words, a relative indicator, such as the percentage of playback time that was viewed, may be used as the objective variable.

[0052] Returning to FIG. 6, the description continues from step S113. In step S113, the learning processing unit 113 performs learning using the learning data 184. As a result, a learned model is generated.

[0053] The generated trained model estimates the amount of time each piece of content is viewed for each display situation.

[0054] For example, a trained model is a supervised model that is generated by learning the relationship between objective variables such as viewing time and number of views, using the viewer's attributes, environmental information (such as date and time), and content as explanatory variables.

[0055] A trained model can be generated using, for example, a bandit algorithm, a type of reinforcement learning. In this case, attribute information and environmental information can be used as context information, display content can be selected, and viewing time and number of viewings can be used as rewards.

[0056] The trained model can be generated using a regression method such as multiple regression. In this case, the learning processing unit 113 uses the multiple regression equation of equation (1) to optimize the constant term γ by the least squares method or the like, and generates a learned model. "y" is the response variable. "x1~x i " indicates attribute information, environmental information, and display content. "α1~α i " and "β1~β j " is a correction coefficient. The correction coefficient α is a parameter of the model. y = α1x1+α2x2+···+α i x i +γ (1)

[0057] Further information on generating trained models. The trained model may be generated using any learning algorithm. For example, known statistical methods or known machine learning algorithms can be used. Examples of known statistical methods include simple regression and multiple regression. Examples of known machine learning algorithms include neural networks, support vector machines (SVMs), gradient boosting decision trees (GBDTs), clustering, and reinforcement learning. Alternatively, a contextual bandit algorithm, a type of reinforcement learning, can be used. A contextual bandit algorithm uses three concepts: "options," "context," and "reward" to determine which content to display. "Options" are content IDs. "Context" is attribute information and environmental information. "Reward" is the viewing time and number of viewings. This allows the optimal content to be selected for the input "context" based on past "options" and "rewards." Examples of contextual bandit algorithms include Linear Upper Confidence found and Linear Thompson Sampling.

[0058] In step S114, the learning processing unit 113 stores the learned model in the learned model storage unit 194.

[0059] Step S120 will be described in detail with reference to FIG. 11. In step S121, the acquisition unit 121 acquires the content information 183 and the trained model and inference pattern information 185 from the storage unit 190. The content information 183 is acquired from the content information storage unit 193 . The trained model is obtained from the trained model storage unit 194. The inference pattern information 185 is acquired from the inference pattern storage unit 195 .

[0060] The inference pattern information 185 will be described with reference to FIG. The inference pattern information 185 indicates an inference pattern. In FIG. 12, the inference pattern information 185 indicates an inference pattern by a set of "pattern No.", "gender", "age", "time period division", and "day division". "Pattern No." is an identifier that identifies an inference pattern. "Gender" is the gender of the viewer. "Age category" is the age category of the viewer. The "time period segment" is a segment of the time period during which the display content was viewed by the viewer. The "day division" is the division of the day when the display content was viewed by the viewer.

[0061] Each inference pattern is predefined. By defining many inference patterns, it becomes possible to optimize the content display ratio under various conditions (situations).

[0062] Returning to FIG. 11, the description continues from step S122. In step S122, the inference processing unit 122 executes inference for each inference pattern using the trained model, thereby obtaining an inference result 186.

[0063] Specifically, the inference processing unit 122 receives the inference pattern information 185 as an input and uses a trained model to infer objective variables such as viewing time and viewing count for the content information 183. The inference processing unit 122 may refer to the content information 183 to determine which content has passed its publication end date and is no longer being advertised, and may not perform inference on that content.

[0064] The inference result 186 will be explained based on FIG. The inference result 186 indicates, for each inference pattern, an estimated value of the time that each piece of content will be viewed when displayed on the display 211. In FIG. 13, the inference result 186 indicates the "inference pattern" and the "viewing time." The “inference pattern” is the same as the inference pattern indicated in the inference pattern information 185 . "Viewing time" is an estimate of the time each piece of content is viewed.

[0065] Returning to FIG. 11, the description continues from step S123. In step S123, the display ratio determination unit 123 determines the display ratio of each piece of content for each inference pattern based on the inference result 186. Information indicating the determined display ratio is referred to as display ratio information 187.

[0066] For example, the display ratio determination unit 123 calculates the display ratio for each piece of content using a softmax function. The softmax function is expressed by equation (2). "i" indicates the content ID. "P i " indicates the display rate (probability) of content whose content ID is "i". "V" indicates the inferred value. "n" indicates the number of contents.

[0067]

number

[0068] If the trained model is not a model that performs point-wise prediction but a model that performs interval prediction such as Gaussian process regression, the predicted value and upper or lower bound may be used when determining the display percentage. For example, the upper bound is used, and the inference results for three pieces of content are revised to 17±3 seconds, 7±3 seconds, and 9±1 seconds within a 95% confidence interval. In this case, the upper bounds of 20 seconds, 10 seconds, and 10 seconds are used, and the display percentages for the three pieces of content are determined to be 50%, 25%, and 25%.

[0069] When a trained model is generated using Linear Thompson Sampling, the trained model holds, as an internal parameter, a probability distribution indicating the probability with which content should be displayed during training. Therefore, the display ratio may be determined using the probability distribution held within the trained model as is.

[0070] If the learning data 184 is not sufficiently accumulated and a learned model has not been constructed, the display ratio determination unit 123 determines the display ratio of each content in an equal distribution.

[0071] The display ratio information 187 will be described with reference to FIG. The display ratio information 187 indicates the display ratio of each content for each inference pattern. In FIG. 14, the display ratio information 187 indicates an "inference pattern" and a "display ratio." The “inference pattern” is the same as the inference pattern indicated in the inference pattern information 185 . "Display ratio" is the display ratio of each content.

[0072] Returning to FIG. 11, step S124 will be described. In step S124, the output unit 124 outputs the display ratio information 187.

[0073] Specifically, the output unit 124 transmits the display ratio information 187 to the display device 210. The control unit 217 receives the display ratio information 187 and stores it in the storage unit 218. Thereafter, the control unit 217 uses the display ratio information 187 to execute content display as follows. First, the control unit 217 determines the situation when content is displayed. For example, the control unit 217 uses camera images to detect the "gender" and "age" of a person in front of the display 211. The control unit 217 also determines the "time period division" and "day division" based on the current date and time. Next, the control unit 217 selects an “inference pattern” that matches the determined situation from the display ratio information 187, and obtains from the display ratio information 187 the “display ratio” associated with the selected “inference pattern”. Then, the control unit 217 displays each content on the display 211 according to the acquired "display ratio."

[0074] By periodically executing the flow of the display control method and periodically delivering the display ratio information 187 to the display device 210, the display device 210 can display the optimum content according to the attributes and environment.

[0075] ***Effects of the First Embodiment*** The first embodiment relates to an apparatus and method for effectively displaying information on a display medium.

[0076] If the model for determining the content display ratio is placed on the server side, the server must be queried to determine the content display ratio, which can cause a time lag before the content is displayed, potentially causing people to walk away from the signage. When a model for determining the display ratio of content is placed on the edge side, computational resources for model learning are required on the edge side, which increases costs on the edge side. When distributing a model from the server side to the edge side, issues arise regarding the communication cost and communication load for distributing the model. In addition, the edge side requires computational resources to determine the display ratio using the model, which increases costs on the edge side.

[0077] The first embodiment has been made in consideration of these circumstances. The first embodiment aims to display content at an optimal display ratio according to attribute information and environmental information while reducing the time lag until content is displayed by distributing display ratio information generated in advance to offline video media such as signage.

[0078] The display control device 100 constructs a model for determining the optimal display ratio based on accumulated past data, and distributes information on the display ratio determined based on the inference results of the trained model.

[0079] According to the first embodiment, it is possible to display content at an optimal display ratio according to attributes and environment while reducing the time lag until content is displayed on offline video media such as signage. In addition, models can be centrally placed on the server side, making data management and model learning more efficient. Furthermore, even if connection to the server is temporarily lost due to poor network conditions, the edge signage can switch the display of content that is optimal for each combination of attributes and environment based on display ratio information distributed in advance.

[0080] ***Supplement to the first embodiment*** The hardware configuration of the display control device 100 will be described with reference to FIG. The display control device 100 includes a processing circuit 109 . The processing circuit 109 is hardware that realizes the learning unit 110 and the inference unit 120 . The processing circuitry 109 may be dedicated hardware, or may be a processor 101 that executes a program stored in memory 102 .

[0081] When processing circuitry 109 is dedicated hardware, processing circuitry 109 may be, for example, a single circuit, a multiple circuit, a programmed processor, parallel programmed processors, an ASIC, an FPGA, or a combination thereof. ASIC is an abbreviation for Application Specific Integrated Circuit. FPGA is an abbreviation for Field Programmable Gate Array.

[0082] The display control device 100 may include a plurality of processing circuits that replace the processing circuit 109.

[0083] In the processing circuit 109, some functions may be realized by dedicated hardware, and the remaining functions may be realized by software or firmware.

[0084] In this way, the functions of the display control device 100 can be realized by hardware, software, firmware, or a combination of these.

[0085] The hardware configuration of the display device 210 will be described with reference to FIG. The display device 210 includes a processing circuit 219 . The processing circuit 219 is hardware that realizes the control unit 217 . The processing circuitry 219 may be dedicated hardware or may be a processor 213 that executes a program stored in memory 214 .

[0086] When processing circuitry 219 is dedicated hardware, processing circuitry 219 may be, for example, a single circuit, multiple circuits, a programmed processor, parallel programmed processors, an ASIC, an FPGA, or a combination thereof.

[0087] The display device 210 may include multiple processing circuits that replace the processing circuit 219.

[0088] In the processing circuit 219, some functions may be realized by dedicated hardware, and the remaining functions may be realized by software or firmware.

[0089] Thus, the functionality of display device 210 can be implemented in hardware, software, firmware, or a combination thereof.

[0090] The first embodiment is an example of a preferred embodiment and is not intended to limit the technical scope of the present disclosure. The first embodiment may be implemented in part or in combination with other embodiments. The procedures described using flowcharts, etc. may be modified as appropriate.

[0091] The "unit" of each element of the display control device 100 and the display device 210 may be read as "processing," "step," "circuit," or "circuitry." [Explanation of symbols]

[0092] 100 Display control device, 101 Processor, 102 Memory, 103 Auxiliary storage device, 104 Communication device, 105 Input / output interface, 109 Processing circuit, 110 Learning unit, 111 Acquisition unit, 112 Learning data generation unit, 113 Learning processing unit, 120 Inference unit, 121 Acquisition unit, 122 Inference processing unit, 123 Display ratio determination unit, 124 Output unit, 181 Detection information, 182 Display performance information, 183 Content information, 184 Learning data, 185 Inference pattern information, 186 Inference result, 187 Display ratio information, 190 Memory unit, 191 Detection information memory unit, 192 Display performance memory unit, 193 Content information memory unit, 194 Learned model memory unit, 195 Inference pattern memory unit, 200 Information provision system, 201 Network, 210 Display device, 211 Display, 212 camera, 213 processor, 214 memory, 215 auxiliary storage device, 216 communication device, 217 control unit, 218 storage unit, 219 processing circuit.

Claims

1. a learning unit that learns detection information obtained each time each of a plurality of contents is displayed by the display device, the detection information indicating the situation when the display was performed and the length of time the displayed content was viewed, and generates a learned model; an inference pattern storage unit in which a plurality of inference patterns indicating different situations when a display is made are predefined and stored; an inference unit that uses the trained model and the plurality of inference patterns to infer the time that each of the plurality of contents will be viewed for each of the inference patterns, and determines the display ratio of each of the plurality of contents for each of the inference patterns based on the inference result; A display control device comprising:

2. The inference unit transmits display ratio information indicating the determined display ratio to a display device that displays each of the plurality of contents at a display ratio corresponding to an inference pattern that matches the situation when the content is displayed. The display control device according to claim 1 .

3. A display control method using a display control device having an inference pattern storage unit in which a plurality of inference patterns showing different situations when a display is performed are predefined and stored, The display control device, generating a trained model by learning detection information obtained each time each of the plurality of contents is displayed by the display device, the detection information indicating the situation when the display was performed and the length of time the displayed content was viewed; Inferring the time when each of the plurality of contents will be viewed for each of the plurality of inference patterns using the trained model and the plurality of inference patterns, and determining the display ratio of each of the plurality of contents for each of the inference patterns based on the inference results. Display control method.

4. A display control program for a computer having an inference pattern storage unit in which a plurality of inference patterns showing different situations when a display is performed are predefined and stored, a learning process for learning detection information obtained each time each of a plurality of contents is displayed by the display device, the detection information indicating the situation when the display was performed and the length of time the displayed content was viewed, to generate a learned model; and an inference process for inferring the time when each of the plurality of contents will be viewed for each of the plurality of inference patterns using the trained model and the plurality of inference patterns, and determining the display ratio of each of the plurality of contents for each of the plurality of inference patterns based on the inference result; A display control program for causing the computer to execute the above.

Images