JP7822337B2 - Information processing device, information processing method, and program - Google Patents

Description

The present invention relates to an information processing device, information processing method, and program for determining recommended lessons.

Conventionally, systems that allow users to search for lessons by setting conditions are known (see, for example, Patent Document 1).

Japanese Patent Application Laid-Open No. 2021-12618

When searching for extracurricular activities using conventional systems, it is possible to find extracurricular activities that meet certain conditions, such as the days of the week or time slots available for attendance. However, it is not always possible to find extracurricular activities that the person taking the activity (e.g., a child) can continue for a long period of time.

The present invention was made in consideration of these points, and aims to make it possible to recommend lessons that are suitable for people who take lessons.

A first aspect of the present invention is an information processing device for recommending lessons suitable for a target person, and includes an acquisition unit that acquires performance data indicating the target person's performance in viewing each of a plurality of pieces of viewing content, an identification unit that identifies one or more pieces of viewing content for which the performance data satisfies a predetermined condition, and a determination unit that determines a lesson to recommend to the target person based on the type of the one or more pieces of viewing content.

The system may further include a storage unit that stores a recommendation database in which the types of viewing content are associated with recommended lesson candidates, and the determination unit may determine the lesson to recommend to the subject by selecting the lesson candidate associated with the one or more types of viewing content identified by the identification unit in the recommendation database.

The performance data may include the length of time the subject viewed each of the plurality of viewed contents, and the identification unit may identify the one or more viewed contents that satisfy the predetermined condition that the length indicated by the performance data is relatively long.

The performance data may include the number of times the subject has viewed each of the plurality of viewing contents, and the identification unit may identify the one or more viewing contents that satisfy the predetermined condition that the number of times indicated by the performance data is relatively high.

The performance data may include the subject's level of concentration measured while the subject is viewing each of the plurality of viewing contents, and the identification unit may identify the one or more viewing contents that satisfy the predetermined condition that the level of concentration is relatively high.

The performance data may include the subject's degree of satisfaction measured while the subject is viewing each of the plurality of viewing contents, and the identification unit may identify the one or more viewing contents that satisfy the predetermined condition that the degree of satisfaction is relatively high.

The performance data may include at least one of the length of time or the number of times the subject viewed each of the plurality of viewed contents, and at least one of the subject's concentration or satisfaction measured while the subject viewed each of the plurality of viewed contents, and the identification unit may identify the one or more viewed contents that satisfy the specified condition of a relatively high level of interest determined by subtracting a level of dissatisfaction determined based on at least one of the concentration or satisfaction from a level of viewing determined based on at least one of the length or the number of times.

The information processing device may further have a memory unit that stores a recommendation database in which a person's activity tendencies are associated with recommended extracurricular activity candidates, the acquisition unit may further acquire activity tendency data indicating the activity tendencies of the subject's family, and the determination unit may determine the extracurricular activity to recommend to the subject based further on the extracurricular activity candidates associated with the activity tendencies indicated by the activity tendency data in the recommendation database.

The information processing device may further have a memory unit that stores a recommendation database in which a person's financial leeway is associated with recommended extracurricular activity candidates, the acquisition unit may further acquire financial leeway data indicating the financial leeway of the subject's family, and the determination unit may determine the extracurricular activity to recommend to the subject based on the extracurricular activity candidates associated with the financial leeway indicated by the leeway data in the recommendation database.

An information processing device according to a second aspect of the present invention is an information processing device for recommending lessons suitable for a subject, and includes: an acquisition unit that acquires performance data indicating the subject's performance in viewing each of multiple pieces of viewing content; and a determination unit that determines, as lessons to recommend to the subject, lessons that are relatively likely to be continued for a predetermined period or longer based on the content of one or more lessons and the expected duration of the lessons output from a machine learning model created by machine learning using as training data performance data for learning indicating the performance of multiple people viewing each of multiple pieces of viewing content, content data indicating the content of the lessons continued by each of the multiple people, and duration data indicating the duration of the lessons.

An information processing device according to a third aspect of the present invention is an information processing device for recommending lessons suitable for a subject, and includes: an acquisition unit that acquires performance data indicating the subject's performance in viewing each of a plurality of pieces of viewing content; and a determination unit that inputs the performance data into a machine learning model created by machine learning using training data that indicates the performance data for multiple people in viewing each of a plurality of pieces of viewing content and lesson data indicating the details of lessons that each of the multiple people has continued for a predetermined period or more, and determines, as the lesson to recommend to the subject, a lesson that has a relatively high probability of being continued for the predetermined period or more, output from the machine learning model.

A fourth aspect of the present invention is an information processing method executed by a computer for recommending lessons suitable for a target person, and includes the steps of acquiring performance data indicating the target person's performance in viewing each of a plurality of pieces of viewing content, identifying one or more pieces of viewing content for which the performance data satisfies a predetermined condition, and determining a lesson to recommend to the target person based on the type of the one or more pieces of viewing content.

A fifth aspect of the present invention is a program for recommending a hobby suitable for a target person, which causes a computer to function as an acquisition unit that acquires performance data indicating the target person's performance in viewing each of a plurality of pieces of viewing content, an identification unit that identifies one or more pieces of viewing content for which the performance data satisfies a predetermined condition, and a determination unit that determines a hobby to recommend to the target person based on the type of the one or more pieces of viewing content.

The present invention has the effect of making it possible to recommend lessons suitable for people who are taking lessons.

FIG. 1 is a diagram for explaining an overview of an information processing system S. FIG. 1 is a diagram illustrating a configuration of an information processing device. FIG. 10 is a diagram illustrating an example of content data. FIG. 10 is a diagram illustrating an example of a recommendation database. FIG. 10 is a diagram illustrating an example of data showing costs required for lessons. FIG. 10 is a diagram illustrating an example of a performance database. FIG. 10 is a diagram showing an example of activity trends of user U2. 3 is a flowchart showing a flow of processing in the information processing device 1.

[Outline of Information Processing System S]
FIG. 1 is a diagram for explaining an overview of an information processing system S. The information processing system S is a system for recommending suitable lessons to people who wish to take up lessons. In this specification, the person to whom the lesson is recommended (i.e., the target person) is referred to as user U1. User U1 is, for example, an infant, but user U1 may also be a child other than an infant or an adult. Also, in this specification, a person who is considering which lesson user U1 should take is referred to as user U2. User U2 is, for example, user U1's parent, but may also be someone other than a parent.

The information processing system S comprises an information processing device 1, a viewing terminal 2, and an analysis terminal 3. The information processing device 1 is a computer, such as a server, that determines the lessons to be recommended. The information processing device 1 provides content for user U1 to view, and notifies user U2 of lessons suitable for user U1 in response to a request from user U2. A lesson suitable for user U1 is one that is relatively likely to be continued by user U1 over a long period of time (for example, a period set by user U2).

The information processing device 1 determines lessons suitable for the user U1 by referencing various data registered in the database DB. The database DB may be stored in the information processing device 1, or in an external device different from the information processing device 1.

The viewing terminal 2 is a terminal on which the user U1 views content, and is, for example, a smartphone, tablet, or personal computer. The viewing terminal 2 displays content transmitted from the information processing device 1 via a network (for example, the Internet or a mobile phone network). The content is, for example, image data of a picture book. The image data may be still image data or video data.

The analysis terminal 3 is a terminal used by user U2, such as a smartphone, tablet, or personal computer. The analysis terminal 3 accepts operations from user U2 to receive lesson recommendations, and displays information indicating lessons recommended by information processing device 1, received from information processing device 1 via the network.

The operation of the information processing system S will now be outlined with reference to Figure 1. First, the information processing device 1 transmits multiple pieces of content to the viewing terminal 2 for viewing by user U1 ((1) in Figure 1). The multiple pieces of content may be selected by user U1 or user U2, or may be randomly selected by the information processing device 1. The multiple pieces of content may be selected by the information processing device 1 from multiple pieces of content corresponding to the type of content selected by user U1 or user U2.

The viewing terminal 2 displays the content received from the information processing device 1 on its display ((2) in Figure 1). The user U1 views the content displayed by the viewing terminal 2 ((3) in Figure 1). The viewing terminal 2 detects when user U1 has started viewing the content and when user U1 has finished viewing the content, and transmits viewing data indicating the detection results to the information processing device 1 ((4) in Figure 1).

The viewing terminal 2 detects that user U1 has started viewing the content and that user U1 has finished viewing the content, for example, based on user U1's operations to start and finish viewing. The viewing terminal 2 may also capture an image of user U1 with a camera while the content is being displayed, and detect that user U1 has started viewing the content and that user U1 has finished viewing the content based on the image data obtained by capturing the image.

Based on the browsing data, the information processing device 1 selects one or more lessons suitable for user U1 from among a large number of candidate lessons. Details will be described later, but the information processing device 1 determines that a lesson is suitable for user U1 if the lesson has a relatively high relevance to the type of content that the browsing data indicates user U1 has viewed for a relatively long period of time or a relatively long number of times. The information processing device 1 may also select some of the lessons determined to be suitable for user U1 based on various conditions set by user U2.

The information processing device 1 transmits analysis data including information indicating one or more lessons determined to be suitable for user U1 to the analysis terminal 3 ((5) in Figure 1). The information indicating the lessons may be the name of the recommended lesson, or information indicating a location or institution where the recommended lesson can be taken. The analysis terminal 3 displays the received analysis data on its display ((6) in Figure 1).

User U2 can understand the recommended lessons by looking at the displayed analysis data. If multiple lessons are displayed, user U2 may select the lesson that he or she would like user U1 to learn from the multiple lessons ((7) in Figure 1). The analysis terminal 3 sends selection data indicating the selection to the information processing device 1 ((8) in Figure 1). The information processing device 1 may provide user U2 with information about places where the selected lesson can be taken, or may register the selected lesson in the database DB. The configuration and operation of the information processing device 1 are described in detail below.

[Configuration of information processing device 1]
2 is a diagram showing the configuration of the information processing device 1. The information processing device 1 has a communication unit 11, a storage unit 12, and a control unit 13. The control unit 13 has a content providing unit 130, an acquisition unit 131, an identification unit 132, and a determination unit 133.

The communication unit 11 includes a communication interface for sending and receiving data between the viewing terminal 2 and the analysis terminal 3 via the network. The communication unit 11 inputs data received from the viewing terminal 2 or the analysis terminal 3 to the acquisition unit 131. The communication unit 11 also transmits data input from the determination unit 133 to the viewing terminal 2 or the analysis terminal 3.

The storage unit 12 has a storage medium such as ROM (Read Only Memory), RAM (Random Access Memory), or SSD (Solid State Drive). The storage unit 12 stores programs executed by the control unit 13. The storage unit 12 also stores a database containing various data used by the control unit 13 to determine lessons to recommend. The storage unit 12 stores, for example, content data indicating the attributes of the viewing content to be viewed by user U1, along with the viewing content.

Figure 3 shows an example of content data. The content data shown in Figure 3 associates the content name, the genre to which each piece of content belongs, the length of the content, and the difficulty level of the content.

The storage unit 12 may store a recommendation database in which types of viewed content are associated with recommended hobby candidates. Figure 4 is a diagram showing an example of a recommendation database. In the recommendation database shown in Figure 4, genres of viewed content are associated with the recommendation level for each of a number of hobbies. The recommendation level is a numerical value indicating the degree of recommendation, with larger numerical values indicating a higher degree of recommendation.

The recommendation database is created, for example, based on the relationship between the past content viewing records of many users U1 and the duration of the lesson, so that the higher the average duration of the lesson for people who have viewed at least one of the specified content for a relatively long time or number of times, the higher the recommendation level. If the genre of the viewed content is baseball, baseball will have the highest recommendation level, followed by soccer, which is a ball game in common with baseball, and skating and gymnastics, which are sports in common with baseball, have higher recommendation levels than piano, chorus, and calligraphy, which are non-sports. On the other hand, if the genre of the viewed content is piano, piano will have the highest recommendation level, followed by chorus, which is a music in common with piano,

In the recommendation database, activity trends of user U1 or a person associated with user U1 (e.g., user U2) may be associated with recommended hobby candidates. Activity trends are represented, for example, by the content of competitions or concerts in which user U2 has participated, sports events they have attended, performances at concerts they have attended, television viewing history, the content of social media or websites they have accessed, or the content of emails they have sent and received.

As an example, the recommendation database is configured so that if user U1 or user U2 tends to have many opportunities to watch sports, the recommendation level for sports-related lessons will be high and the recommendation level for non-sports-related lessons will be low. The recommendation database is configured so that if user U1 or user U2 tends to have many opportunities to go to concerts, the recommendation level for music-related lessons will be relatively high.

Similarly, the recommendation database is configured so that if user U2 has a strong tendency to watch sports programs on television, a strong tendency to access sports-related social media or websites, or a strong tendency to exchange emails about sports-related topics, the recommendation level for sports-related lessons will be high and the recommendation level for non-sports-related lessons will be low.

The recommendation database may associate the financial capacity of a person associated with user U1 (e.g., user U2) with recommended extracurricular activities. Financial capacity is represented, for example, by the annual income of user U1's parents minus their annual expenses. As an example, the recommendation database is configured so that when financial capacity is high, all extracurricular activities are highly recommended, and when financial capacity is low, skating, which is a relatively expensive extracurricular activity, is less recommended.

The storage unit 12 may store data indicating the cost of each lesson associated with user U1. Figure 5 is a diagram showing an example of data indicating the cost of lessons. In Figure 5, each type of lesson is associated with annual tuition fees, transportation costs, total costs, and access types. The cost of lessons may also include the cost of equipment required for the lesson.

The storage unit 12 may also store performance data indicating the performance of user U1 in viewing the viewing content in association with user U1. The performance data may include, for example, at least one of the length of time that user U1 viewed each of the multiple viewing contents, or the number of times that user U1 viewed each of the multiple viewing contents.

The performance data may include at least one of the user U1's concentration level measured while the user U1 is viewing each of the multiple viewing contents, or the user U1's satisfaction level measured while the user U1 is viewing each of the multiple viewing contents. The performance data may include at least one of the length of time that the user U1 views each of the multiple viewing contents or the number of times the user U1 views each of the multiple viewing contents, and at least one of the user U1's concentration level or satisfaction level measured while the user U1 is viewing each of the multiple viewing contents.

Figure 6 is a diagram showing an example of an achievement database. Figure 6(a) shows achievement data for a first user U1, and Figure 6(b) shows achievement data for a second user U1. The achievement data shown in Figure 6 associates the genre of content that user U1 can view, the time spent by user U1 viewing, the number of times user U1 viewed, and the level of concentration and satisfaction while user U1 was viewing. Concentration and satisfaction are indicated on a five-point scale from 1 to 5, with 5 points representing the highest level of concentration or satisfaction and 1 point representing the lowest level of concentration or satisfaction.

The performance data in Figure 6(a) shows that soccer-related content is viewed more times and for longer periods of time than content in other genres. On the other hand, the performance data in Figure 6(b) shows that piano-related content is viewed more times and for longer periods of time than content in other genres.

The configuration and operation of the control unit 13 will be described below with reference to Figure 2. The control unit 13 has, for example, a CPU (Central Processing Unit), and functions as a content providing unit 130, an acquiring unit 131, an identifying unit 132, and a determining unit 133 by executing programs stored in the storage unit 12.

The content providing unit 130 provides viewing content. As an example, the content providing unit 130 first causes the viewing terminal 2 or the analysis terminal 3 to display a list of viewable content as shown in FIG. 3 via the communication unit 11. When the content providing unit 130 acquires instruction data from the viewing terminal 2 or the analysis terminal 3 specifying the viewing content to be displayed on the viewing terminal 2, it reads the specified viewing content from the storage unit 12 and transmits the read viewing content to the viewing terminal 2.

The content providing unit 130 may transmit to the viewing terminal 2 viewing content selected randomly without receiving a designation of viewing content, or viewing content selected based on rules specified by user U2. The content providing unit 130 terminates transmission of the viewing content when it receives an instruction to stop viewing the viewing content from the viewing terminal 2 or the analysis terminal 3.

The acquisition unit 131 acquires performance data indicating the performance of user U1 viewing each of multiple viewing contents. The acquisition unit 131 notifies the identification unit 132 of the acquired performance data. The acquisition unit 131 notifies the identification unit 132 of the performance data via the storage unit 12, for example, by associating the performance data with user U11 and storing it in the storage unit 12.

The acquisition unit 131 acquires, for example, the time from when the content providing unit 130 starts transmitting the viewing content to when the provision of the viewing content ends as performance data. The acquisition unit 131 may also acquire, as performance data, the viewing time or number of views measured by the viewing terminal 2. The acquisition unit 131 may also acquire, as performance data, the level of concentration or satisfaction of user U1 while viewing the viewing content, as identified by the viewing terminal 2.

The acquisition unit 131 may acquire performance data by determining the viewing time or number of views based on the time from when the content providing unit 130 starts transmitting the viewed content to when the provision of the viewed content ends. After the content providing unit 130 starts transmitting the viewed content, the acquisition unit 131 may increment by 1 the number of views of the viewed content of the same genre in the performance data stored in the storage unit 12.

When the viewed content includes multiple pages, the acquisition unit 131 may, for example, determine the time when a page related to content of a specific genre is viewed as the time when user U1 viewed the viewed content of the specific genre. The acquisition unit 131 may also determine the time when all pages of the viewed content including content of that genre are viewed as the viewing time when user U1 viewed the viewed content.

When user U1 views at least one page of a plurality of pages that includes content in a predetermined genre, the acquisition unit 131 increments the number of times user U1 has viewed the viewed content of that genre in the performance data by 1. The acquisition unit 131 may also increment the number of times user U1 has viewed the viewed content of that genre in the performance data by 1 each time all pages of the viewed content that includes content in that genre are viewed.

The acquisition unit 131 may acquire, as performance data, captured image data created by the viewing terminal 2 capturing an image of user U1 while the viewing content is displayed on the viewing terminal 2. The acquisition unit 131 identifies at least one of user U1's concentration level or satisfaction level by analyzing the captured image data, and stores the concentration level or satisfaction level in the storage unit 12 in association with user U1 and the viewing content. The acquisition unit 131 increases the concentration level value in the performance data the longer the percentage of time user U1 spends gazing at the screen. The acquisition unit 131 increases the satisfaction level value the longer the percentage of time user U1 is smiling.

The acquisition unit 131 may acquire activity trend data indicating the activity trends of user U1's family. The acquisition unit 131 acquires, from the analysis terminal 3, data indicating the activity trends of user U2 identified by the analysis terminal 3 based on, for example, the content of an operation performed by user U2 on the analysis terminal 3 (for example, the content of a website accessed). The acquisition unit 131 associates the acquired activity trend data with at least either user U1 or user U2 and stores it in the storage unit 12.

The acquisition unit 131 may acquire financial leeway data indicating the financial leeway of user U1's family (e.g., user U2). The acquisition unit 131 acquires, for example, leeway data entered into the analysis terminal 3 by user U2. The acquisition unit 131 associates the acquired leeway data with at least one of user U1 and user U2 and stores it in the storage unit 12.

The identification unit 132 identifies one or more viewed contents for which the performance data acquired by the acquisition unit 131 satisfies a predetermined condition. The predetermined condition is, for example, that the viewing degree of user U1 is relatively high, or that the viewing degree is equal to or greater than a threshold. The viewing degree becomes a larger value, for example, the longer at least one of the time or number of times that user U1 viewed the content, as indicated by the performance data. The viewing degree may also become a larger value the higher at least one of the concentration level or satisfaction level of user U1 while viewing the content.

That is, the predetermined condition is, for example, a condition that the length of time that user U1, as indicated by the performance data, has viewed the viewed content is relatively long. In the example shown in FIG. 6, the total viewing time of soccer-related viewed content for user U1, corresponding to FIG. 6(a), is 110 minutes, which is longer than the viewing time of viewed content in other genres. Therefore, the identification unit 132 identifies the soccer-related viewed content as viewed content that satisfies the predetermined condition. On the other hand, for user U1, corresponding to FIG. 6(b), the total viewing time of piano-related viewed content is 60 minutes, which is longer than the viewing time of viewed content in other genres. Therefore, the identification unit 132 identifies the piano-related viewed content as viewed content that satisfies the predetermined condition.

The predetermined condition may be that the number of times user U1, as indicated by the performance data, has viewed the viewed content is relatively high. In the example shown in FIG. 6, user U1, corresponding to FIG. 6(a), has viewed soccer-related viewed content eight times, which is more than the number of times viewed content in other genres. Therefore, the identification unit 132 identifies the soccer-related viewed content as viewed content that satisfies the predetermined condition. On the other hand, user U1, corresponding to FIG. 6(b), has viewed piano-related viewed content six times, which is more than the number of times viewed content in other genres. Therefore, the identification unit 132 identifies the piano-related viewed content as viewed content that satisfies the predetermined condition.

The predetermined condition may be that the level of concentration while user U1, as indicated by the performance data, is relatively high while viewing the viewed content. In the example shown in FIG. 6, the average level of concentration while user U1, corresponding to FIG. 6(a), is viewing soccer-related viewed content is 4.7 points, which is higher than the average level of concentration while viewing viewed content of other genres. Therefore, the identification unit 132 identifies the soccer-related viewed content as viewed content that satisfies the predetermined condition. On the other hand, for user U1, corresponding to FIG. 6(b), the average level of concentration while viewing piano-related viewed content is 4.5 points, which is higher than the average level of concentration while viewing viewed content of other genres. Therefore, the identification unit 132 identifies the piano-related viewed content as viewed content that satisfies the predetermined condition.

The predetermined condition may be that the satisfaction level of user U1, as indicated by the performance data, while viewing the viewed content is relatively high. In the example shown in FIG. 6, the average satisfaction level of user U1, corresponding to FIG. 6(a), while viewing soccer-related viewed content is 4.3 points, which is higher than the average satisfaction level while viewing viewed content of other genres. Therefore, the identification unit 132 identifies the soccer-related viewed content as viewed content that satisfies the predetermined condition. On the other hand, for user U1, corresponding to FIG. 6(b), the average satisfaction level while viewing piano-related viewed content is 4.5 points, which is higher than the average satisfaction level while viewing viewed content of other genres. Therefore, the identification unit 132 identifies the piano-related viewed content as viewed content that satisfies the predetermined condition.

The predetermined condition may be that the degree of browsing calculated by combining multiple pieces of information among the browsing time, the number of times of browsing, the concentration level, and the satisfaction level is relatively high. As an example, the identification unit 132 calculates the degree of browsing using the following formula:
Viewing rate = (a1 x H1) + (a2 x H2) + (a3 x H3) + b x C
Here, H1, H2, and H3 are the times spent viewing each of the three pieces of content in a specific genre, a1, a2, and a3 are coefficients corresponding to the concentration or satisfaction of user U1 while viewing each of the three pieces of content, C is the number of times the content in the genre was viewed, and b is a coefficient.

The less time user U1 looks away and the less physical movement they make while viewing the viewing content, the more focused they are on the viewing content. On the other hand, when user U1 is viewing viewing content that does not interest him, he becomes restless, and the time (L) when his eyes leave the viewing content and the time (M) when his physical movement exceeds a predetermined amount are prolonged. Therefore, the acquisition unit 131 may determine user U1's level of interest in viewing content of a predetermined genre based on the degree of viewing and user U1's level of dissatisfaction.

Specifically, the identification unit 132 may identify one or more viewed contents that satisfy a predetermined condition that the degree of interest determined by subtracting the degree of dissatisfaction determined based on at least one of the concentration level and the satisfaction level from the degree of viewing determined based on at least one of the length and the number of times is relatively high. The degree of interest is calculated, for example, using the following formula:
Interest level = browsing level - dissatisfaction level = browsing level - (c x L + d x M)
Here, c and d are predetermined coefficients.

The determination unit 133 determines the extracurricular activity to recommend to user U1 by selecting, in the recommendation database, candidate extracurricular activities associated with one or more types of viewed content identified by the identification unit 132. The determination unit 133 notifies the analysis terminal 3 of the determined extracurricular activity via the communication unit 11. When the type of viewed content identified by the identification unit 132 is "baseball," the determination unit 133 determines to recommend "baseball" and "soccer," which are candidate extracurricular activities whose recommendation levels for the content genre "baseball" shown in FIG. 4 are equal to or greater than a predetermined threshold (e.g., 5). The predetermined threshold may be a value set by user U2 via the analysis terminal 3, a value set by a business providing services using the information processing device 1, or a value set when the information processing device 1 was manufactured.

The determination unit 133 may notify the analysis terminal 3 of multiple recommended lessons in association with the lesson recommendation level. The determination unit 133 may notify the analysis terminal 3 of multiple recommended lessons in a state where the lessons are arranged in descending order of the lesson recommendation level.

However, the length of time that user U1 can continue a lesson depends on factors other than user U1's level of interest in the content viewed. The length of time that user U1 can continue a lesson depends, for example, on the behavioral or preference tendencies of user U2, the relationship between the cost of the lesson and user U2's financial leeway, the distance required to travel to take the lesson, etc. Therefore, the determination unit 133 may determine the lesson to recommend based on these factors as well.

The determination unit 133 determines the lesson to recommend to user U1 based, for example, on the candidate lessons associated with the activity tendencies indicated by the activity tendency data in the recommendation database. For example, if user U2 tends to watch sports frequently, the determination unit 133 will preferentially recommend a lesson related to sports. If user U2 tends to go to concerts frequently, the determination unit 133 will preferentially recommend a lesson related to music. Similarly, if user U2 tends to watch sports programs on television, if he tends to access sports-related social media or websites, or if he tends to exchange emails about sports-related topics, the determination unit 133 will preferentially recommend a lesson related to sports.

Figure 7 is a diagram showing an example of the activity trends of user U2. Circles in Figure 7 indicate activities that the father or mother engage in relatively frequently. The father shown in Figure 7(a) has a strong interest in soccer, and the mother has a strong interest in skating. In such a case, the decision unit 133 decides to preferentially recommend soccer or skating. The father shown in Figure 7(b) has a strong interest in choral singing, and the mother has a strong interest in piano and choral singing. In such a case, the decision unit 133 decides to preferentially recommend piano or choral singing.

The activity tendency may be calculated as a parent activity tendency value by integrating the activity tendencies of the father and the mother. As an example, the determination unit 133 calculates the parent activity tendency value for each extracurricular activity category using the following formula:
Parents' activity tendency score = e x father's activity level + f x mother's activity level + g x father's past activity level + h x mother's past activity level

In the above formula, e, f, g, and h are coefficients. For example, the activity level in Figure 7 is given a score of 1 when there is a circle and a score of 0 when there is no circle, but it may also be expressed by multiple scores depending on the activity level. The decision unit 133 decides to preferentially recommend lessons in genres with high parent activity tendency values.

The determination unit 133 may determine the lesson to recommend to user U1 based on the lesson candidate associated with the financial leeway indicated by the leeway data in the recommendation database. The determination unit 133 tentatively determines multiple lesson candidate based on, for example, one or more types of viewing content identified by the identification unit 132. In such a case, the determination unit 133 does not recommend any of the tentatively determined lesson candidate candidates that do not match user U2's financial leeway.

As an example, suppose the determination unit 133 tentatively determines soccer and skating as candidate extracurricular activities based on the type of one or more viewed content identified by the identification unit 132. As shown in FIG. 5, the annual cost of learning soccer is 100,000 yen, while the annual cost of learning skating is 660,000 yen. If user U2 has a high level of financial leeway and the amount obtained by subtracting living expenses from annual income is greater than 660,000 yen, the determination unit 133 decides to recommend both soccer and skating. If user U2 has a low level of financial leeway and the amount obtained by subtracting living expenses from annual income is less than 660,000 yen, the determination unit 133 decides to recommend soccer and not to recommend skating. By operating in this manner, the determination unit 133 can prevent the recommendation of extracurricular activities that user U2 will have difficulty continuing due to a lack of financial leeway.

However, if the lesson location is far from the user's home and requires a long commute, it may be difficult to continue the lesson. Therefore, the determination unit 133 may determine the lesson to recommend based on the distance from the user U1's home to the lesson location or the time required for the commute. As an example, the determination unit 133 may tentatively determine soccer, skating, piano, and singing as lesson candidate options based on the type of one or more viewing content identified by the identification unit 132. As shown in FIG. 5, soccer and singing are lessons that can be attended on foot. In such a case, the determination unit 133 preferentially recommends soccer and singing.

The decision unit 133 may also determine which lessons to recommend based on the quality of the environment, such as whether there is a suitable coach in a location where the lesson can be taken within a threshold distance or travel time from user U1. As an example, the decision unit 133 may recommend a lesson if the distance or travel time from user U1 is within a first threshold and the quality of the environment is equal to or greater than a second threshold.

[Processing flow in information processing device 1]
Fig. 8 is a flowchart showing the flow of processing in the information processing device 1. The flowchart shown in Fig. 8 starts after the content providing unit 130 provides browsing content to the user U1.

The acquisition unit 131 acquires from the viewing terminal 2 performance data indicating that user U1 has started or finished viewing viewing content (S1). The identification unit 132 identifies viewing content that satisfies a predetermined condition, such as user U1's level of interest being equal to or greater than threshold Ta1, based on at least one of the length of viewing time, number of times, level of concentration during viewing, and satisfaction level indicated in the performance data (S2).

The determination unit 133 refers to the recommendation database and provisionally selects one candidate lesson to recommend based on the type of viewing content identified by the identification unit 132 (S3). The determination unit 133 determines whether user U2's financial leeway is equal to or greater than threshold Tb1 for the provisionally selected lesson (S4). If the financial leeway is equal to or greater than threshold Tb1 (YES in S4), the determination unit 133 proceeds to S7.

If the financial leeway is less than threshold Tb1 and greater than or equal to threshold Tb2 (YES in S5), the determination unit 133 determines whether the interest level is greater than or equal to threshold Ta2, which is greater than threshold Ta1 (S6). If the determination unit 133 determines that the interest level is greater than or equal to threshold Ta2 (YES in S6), the determination unit 133 proceeds to S7 because the interest level is sufficiently high even if the financial leeway is not very high.

In S7, the decision unit 133 determines whether the candidate lesson location is easy to commute to based on the distance or travel time from the home to the lesson location (S7). If the decision unit 133 determines that the location is easy to commute to (YES in S7), it decides to recommend the provisionally selected lesson candidate (S8). If the decision unit 133 determines NO in S5, S6, and S7, it decides not to recommend the provisionally selected lesson candidate (S9).

If there are other candidate lessons (YES in S10), the determination unit 133 repeats the processes from S4 to S9. If there are no other candidate lessons (NO in S10), the determination unit 133 ends the process.

[Modification]
In the above explanation, the control unit 13 determines the lessons to recommend based on various factors related to user U1 or user U2 by referring to the recommendation table, but the control unit 13 may also determine the lessons to recommend by using a machine learning model.

The decision unit 133 may, for example, input performance data into the machine learning model and, based on the content of one or more lessons and the expected duration of the lessons output from the machine learning model, decide to recommend lessons to user U1 that are relatively likely to be continued for a predetermined period of time or longer. The performance data may, for example, associate the content viewed by user U1 with the viewing time and the number of views. The performance data may include the level of concentration or satisfaction. The performance data may also be captured image data capturing the facial expression of user U1 while he or she is viewing the content.

This machine learning model is created by machine learning using learning achievement data indicating the achievements of multiple people viewing each of multiple pieces of viewing content, content data indicating the details of the lessons continued by each of the multiple people, and duration data indicating the period for which the lesson was continued as training data as training data. When achievement data is input, this type of machine learning model can output a likely period for which another user U1, whose achievements have a similar tendency to the input achievement data, continued the lesson. Therefore, by using this type of machine learning model, the determination unit 133 can more easily select lessons that user U1 is likely to continue for a long time.

The machine learning model may be created by learning using, in addition to performance data, activity tendency data, financial leeway, ease of commuting to a place where lessons can be taken (i.e., distance or travel time), etc. as training data. In this case, the determination unit 133 inputs, in addition to performance data, the activity tendency data, financial leeway, and ease of commuting to a place where lessons can be taken into the machine learning model. By using such a machine learning model, the determination unit 133 further increases the probability that user U1 will be able to select a lesson that they can continue for a long time.

The machine learning model may be created by machine learning using learning performance data indicating the viewing history of multiple people for each of multiple viewing contents and extracurricular activity data indicating the details of extracurricular activities that each of the multiple people has continued for a predetermined period of time as training data. The predetermined period is a period set by user U2 or a period determined by a provider that provides services using information processing device 1, such as three years.

The decision unit 133 may input performance data into this machine learning model and determine, as the lesson to recommend to user U1, the lesson that is output from the machine learning model and that is relatively likely to be continued by user U1 for a predetermined period or longer. By using such a machine learning model, the decision unit 133 increases the probability of selecting a lesson that user U1 can continue for a predetermined period or longer. Note that this machine learning model may also be created by learning using, in addition to performance data, activity tendency data, financial availability, accessibility to a place where the lesson can be taken (i.e., distance or travel time), etc. as training data.

[Effects of information processing device 1]
As described above, in the information processing device 1, the identification unit 132 identifies one or more pieces of viewing content for which performance data indicating the performance of user U1 viewing the viewing content satisfies a predetermined condition, and the determination unit 133 determines a lesson to recommend to user U1 based on the type of the one or more pieces of viewing content. By configuring the information processing device 1 in this way, even if user U1 is an infant who is unable to determine what lessons interest him or her, the information processing device 1 is more likely to be able to recommend a lesson that is suitable for user U1.

Furthermore, this invention will contribute to Goal 9 of the United Nations' Sustainable Development Goals (SDGs), "Build resilient infrastructure, promote inclusive and sustainable industrialization, and promote innovation and resilience."

The present invention has been described above using embodiments, but the technical scope of the present invention is not limited to the scope described in the above embodiments, and various modifications and alterations are possible within the spirit of the invention. For example, all or part of the device can be configured by functionally or physically distributing or integrating any unit. Furthermore, new embodiments resulting from any combination of multiple embodiments are also included in the embodiments of the present invention. The effects of new embodiments resulting from the combination will also have the effects of the original embodiments.

REFERENCE SIGNS LIST 1 Information processing device 2 Viewing terminal 3 Analysis terminal 11 Communication unit 12 Storage unit 13 Control unit 130 Content providing unit 131 Acquisition unit 132 Identification unit 133 Decision unit

Claims (8)

An information processing device for recommending lessons suitable for a target person,
an acquisition unit that acquires performance data indicating the performance of the target person in browsing each of a plurality of browsing contents and the financial allowance of a related person associated with the target person ;
a storage unit that stores a recommendation database in which the type of the content to be viewed, a recommended lesson candidate, and the financial availability of the related person are associated with each other;
an identification unit that identifies one or more of the viewed contents, based on the performance data , that satisfy a predetermined condition that the degree of interest of the subject is equal to or greater than a threshold ;
a determination unit that selects tentative candidates for lessons to be recommended to the subject based on the type of the one or more viewing contents, and determines, from the selected tentative candidates, a lesson candidate that is associated with the financial allowance of the related person acquired by the acquisition unit in the recommendation database and that is determined to be easy to attend based on the distance or travel time from the subject's home to the place where the lesson is to be taken, as the lesson to be recommended to the subject ;
An information processing device having the above. the performance data includes a length of time that the subject viewed each of the plurality of viewing contents;
the identifying unit identifies the one or more pieces of browsing content that satisfy the predetermined condition that the length indicated by the performance data is relatively long;
The information processing device according to claim 1 . the performance data includes the number of times the subject has viewed each of the plurality of viewing contents;
the identification unit identifies the one or more pieces of browsing content that satisfy the predetermined condition that the number of times indicated by the performance data is relatively large.
The information processing device according to claim 1 . the performance data includes a concentration level of the subject measured while the subject is viewing each of the plurality of viewing contents;
the identification unit identifies the one or more pieces of browsing content that satisfy the predetermined condition that the degree of concentration is relatively high.
The information processing device according to claim 1 . the performance data includes a satisfaction level of the subject measured while the subject is viewing each of the plurality of viewing contents;
the specifying unit specifies the one or more pieces of browsing content that satisfy the predetermined condition that the satisfaction level is relatively high.
The information processing device according to claim 1 . the performance data includes at least one of a length of time that the subject viewed each of the plurality of viewing contents or a number of times the subject viewed each of the plurality of viewing contents, and at least one of a concentration level or a satisfaction level of the subject measured while the subject viewed each of the plurality of viewing contents;
the identification unit identifies the one or more viewed contents that satisfy the predetermined condition that the degree of interest determined by subtracting a degree of dissatisfaction determined based on at least one of the concentration degree or the satisfaction degree from a degree of viewing determined based on at least one of the length or the number of times is relatively high.
The information processing device according to claim 1 . An information processing method for recommending lessons suitable for a target person, executed by a computer, comprising:
acquiring performance data indicating the performance of the target person in browsing each of a plurality of browsing contents and the financial allowance of related persons associated with the target person ;
Identifying one or more of the viewed contents , in which the performance data satisfies a predetermined condition that the interest level of the subject is equal to or greater than a threshold ;
selecting tentative candidates for lessons to be recommended to the subject based on the type of the one or more viewed contents, and determining, from the tentative candidates selected, the lesson candidate that is associated with the acquired financial allowance of the related person in a recommendation database in which the type of viewed content, the recommended lesson candidate, and the related person's financial allowance are associated, and that is determined to be easy to attend based on the distance or travel time from the subject's home to the place where the lesson will be held, as the lesson to be recommended to the subject ;
An information processing method comprising: A program for recommending suitable lessons to target individuals,
On the computer,
acquiring performance data indicating the performance of the target person in browsing each of a plurality of browsing contents and the financial allowance of related persons associated with the target person;
Identifying one or more of the viewed contents, in which the performance data satisfies a predetermined condition that the interest level of the subject is equal to or greater than a threshold;
selecting tentative candidates for lessons to be recommended to the subject based on the type of the one or more viewed contents, and determining, from the tentative candidates selected, the lesson candidate that is associated with the acquired financial allowance of the related person in a recommendation database in which the type of viewed content, the recommended lesson candidate, and the related person's financial allowance are associated, and that is determined to be easy to attend based on the distance or travel time from the subject's home to the place where the lesson will be held, as the lesson to be recommended to the subject;
A program to execute .