Data processing device, data processing method and program

Abstract

To select content for enhancing a therapeutic effect by giving a stimulus to a patient.SOLUTION: There is provided a data processing device for selecting content given as a stimulus, which includes a selection unit 132 for selecting first content given to a user as a stimulus out of a plurality of pieces of content, and an acquisition unit 131 for acquiring sleeping data that indicates the quality of sleeping of the user given the first content selected by the selection unit 132. The selection unit 132 selects second content having an attribute different from that of the first content on the basis of relationships between an attribute of the stimulus given to the user by the first content and the quality of sleeping indicated by the sleeping data.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to a data processing device, a data processing method, and a program. [Background technology]

[0002] It is known that providing stimuli such as music or images to dementia patients can promote daytime wakefulness and nighttime sleep, thereby improving insomnia and day-night reversal, and thereby having a certain effect on improving dementia (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2020-161187 Summary of the Invention [Problem to be solved by the invention]

[0004] In conventional treatments, it has been assumed that stimuli such as predetermined music or images are provided to a person (hereinafter referred to as "user") who is to improve the quality of their sleep. However, the type of stimulus that is effective in improving sleep quality varies from user to user, and there are cases where the provided stimulus is ineffective. Therefore, there is a need to provide stimuli that are highly effective in improving the quality of sleep for each user.

[0005] The present invention has been made in consideration of these points, and aims to enhance the effect of improving the quality of sleep by providing stimulation to the user. [Means for solving the problem]

[0006] A data processing device of a first aspect of the present invention includes a selection unit that selects a first content to be provided as a stimulus to a user from a plurality of contents, and an acquisition unit that acquires sleep data indicating the quality of sleep of the user who has been given the first content selected by the selection unit, and the selection unit selects a second content having attributes different from those of the first content based on the relationship between the attributes of the first content that affect the content of the stimulus provided to the user by the first content and the quality of sleep indicated by the sleep data.

[0007] The acquisition unit may further acquire user attribute data representing attributes of the user, and the selection unit may select the first content and the second content further based on the user attribute data.

[0008] The device may include a memory unit that stores the sleep data in association with sleep days, and the selection unit may select the second content when the quality of sleep indicated by second sleep data, which is sleep data during a second sleep period including one or more sleep days after the first content is given to the user as a stimulus, is not improved compared to the quality of sleep indicated by first sleep data, which is sleep data during a first sleep period including one or more sleep days before the first content is given to the user as a stimulus.

[0009] The device may further include an evaluation unit that creates an evaluation result indicating the degree of effect of providing the first content to the user based on the relationship between the quality of sleep indicated by the first sleep data and the quality of sleep indicated by the second sleep data.

[0010] The acquisition unit may further acquire ambient environment data representing the state of the user's ambient environment, and the evaluation unit may identify the degree to which the ambient environment indicated by the ambient environment data has affected sleep quality by referring to data indicating the relationship between the ambient environment and sleep quality, and create the evaluation result by correcting the degree of effect identified based on the relationship between the sleep quality indicated by the first sleep data and the sleep quality indicated by the second sleep data based on the identified degree of influence.

[0011] The acquisition unit may further acquire activity data representing the user's activity content, and the evaluation unit may identify the degree to which the activity content indicated by the activity data has affected sleep quality by referring to data indicating the relationship between the activity content and sleep quality, and create the evaluation result by correcting the degree of effect identified based on the relationship between the sleep quality indicated by the first sleep data and the sleep quality indicated by the second sleep data based on the identified degree of influence.

[0012] The selection unit may select the second content when the quality of sleep indicated by the sleep data after the first content is given to the user as a stimulus is worse than a predetermined threshold.

[0013] The selection unit may input attribute data of the first content and the sleep data measured after the first content was provided into a trained model created by learning using training sleep data, attributes of training content provided to the training user whose training sleep data was measured, and results of improvement in the sleep quality of the training user after the training content was provided as training data, and select the second content corresponding to the attributes output by the trained model.

[0014] The selection unit may select the second content corresponding to the attributes output by a trained model created by learning using as training data the training sleep data, the attributes of the training content given to the training user whose training sleep data was measured, the attributes of the training user, and the results of improvement in the training user's sleep quality after the training content was given, by inputting the attribute data of the first content, the user attribute data of the user, and the sleep data measured after the first content was given.

[0015] The selection unit may select the first content and the second content without receiving an instruction to select content from the user.

[0016] The selection unit may select a plurality of the second contents corresponding to a plurality of time periods and having different attributes.

[0017] The acquisition unit may further acquire surrounding environment data representing the state of the user's surrounding environment, and the selection unit may select the second content to be provided to the user in the surrounding environment represented by the surrounding environment data by referring to data representing the impact that the user's surrounding environment has on the user's mind and body and data representing the content of stimuli appropriate to the user's mind and body state.

[0018] The acquisition unit may further acquire activity data representing the content of the user's activity, and the selection unit may select the second content to be given to the user who has engaged in the activity represented by the activity data by referring to data indicating the impact that the content of the user's activity has on the user's mind and body and data indicating the content of stimuli appropriate to the user's mind and body state.

[0019] A data processing device of a second aspect of the present invention includes a selection unit that selects a first content to be provided as a stimulus to a user from a plurality of contents, and an acquisition unit that acquires work data indicating the work content of a caregiver who is caring for the user to whom the first content selected by the selection unit has been provided, and the selection unit estimates the quality of sleep of the user based on the work content indicated by the work data, and selects a second content having attributes different from those of the first content based on the relationship between the attributes of the first content that affect the content of the stimulus provided to the user by the first content and the estimated quality of sleep.

[0020] A data processing method of a third aspect of the present invention includes the steps of selecting a first content to be provided as a stimulus to a user from a plurality of contents, executed by a computer; acquiring sleep data indicating the quality of sleep of the user to whom the selected first content is provided; and selecting a second content having attributes different from the first content based on the relationship between attributes of the first content that affect the content of the stimulus provided to the user by the first content and the quality of sleep indicated by the sleep data.

[0021] A fourth aspect of the program of the present invention causes a computer to perform the steps of selecting a first content from a plurality of contents to be provided as a stimulus to a user, acquiring sleep data indicating the quality of sleep of the user to whom the selected first content is provided, and selecting a second content having attributes different from those of the first content based on the relationship between attributes of the first content that affect the content of the stimulus provided to the user by the first content and the quality of sleep indicated by the sleep data. [Effects of the Invention]

[0022] According to the present invention, it is possible to select content that enhances the therapeutic effect of providing stimulation to a patient. [Brief explanation of the drawings]

[0023] [Figure 1] FIG. 1 is a diagram for explaining an overview of a data processing system S. [Figure 2] 1 is a block diagram showing a configuration of a data processing device 1A according to a first embodiment. [Figure 3] FIG. 2 is a diagram illustrating an example of a data structure of user data. [Figure 4] 10 is a flowchart showing the flow of processing in the data processing device 1A. [Figure 5] FIG. 10 is a block diagram showing the configuration of a data processing device 1B according to a modified example. [Figure 6]FIG. 10 is a block diagram showing the configuration of a data processing device 1C according to a second embodiment. [Figure 7] 10 is a flowchart showing the flow of processing in a data processing device 1C according to the second embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0024] <Outline of data processing device 1> FIG. 1 is a diagram for explaining an overview of a data processing system S. First, an overview of the data processing system S will be explained with reference to FIG. 1. The data processing system S is a system for selecting content suitable for improving the quality of a patient's sleep. The data processing system S is a system that selects first content to be given to a user as a stimulus in order to improve the user's quality of sleep, and further selects second content to be given to the user as a stimulus by receiving feedback on the quality of sleep of the user who was given the selected content. The data processing system S includes a data processing device 1, a viewing device 2, and a vital sensor 3.

[0025] The data processing device 1 is a device for selecting content to be provided to the user as a stimulus. The viewing device 2 is a device for the user to view content. The viewing device 2 is, for example, a computer, a smartphone, a tablet, a speaker, or a television. The vital sensor 3 is a device for measuring data indicating the quality of the user's sleep. The vital sensor 3 is, for example, a wearable device or a smartphone capable of acquiring vital data. The content to be provided as a stimulus is music data, image data such as movies, videos, photographs, or artworks, AR (Augmented Reality) image data, VR (Virtual Reality) image data, etc. In the following, music will be used as an example of content to be provided to the user by the data processing device 1 as a stimulus.

[0026] For example, in response to a request from the user, the data processing device 1 selects a first content to be provided to the user as a stimulus ((1) in FIG. 1). The data processing device 1 transmits content data of the selected first content to the viewing device 2 ((2) in FIG. 1). The viewing device 2 plays the transmitted first content, and the user views the selected first content. Note that if the content data transmitted by the data processing device 1 includes information for identifying the content, the viewing device 2 may obtain content corresponding to the information for identifying the content received from the data processing device 1 from a storage area of ​​the viewing device 2 or an external content server, and play the content.

[0027] After the user views the first content, the vital sensor 3 acquires sleep data indicating the quality of the user's sleep while the user is asleep and transmits the sleep data to the data processing device 1 ((3) in FIG. 1). The vital sensor 3 may transmit the sleep data to the data processing device 1 via the viewing device 2.

[0028] The data processing device 1 selects second content to be provided to the user as a stimulus based on information for identifying the first content provided to the user as a stimulus and sleep data acquired from the vital sensor 3 and indicating the quality of the user's sleep after the user is provided with the first content ((4) in FIG. 1). As will be described in detail later, if the quality of the user's sleep after the user is provided with the first content is not good, the data processing device 1 selects second content that is likely to improve the quality of the user's sleep. Then, the data processing device 1 transmits content data of the selected second content to the viewing device 2 ((5) in FIG. 1).

[0029] First Embodiment [Configuration of data processing device 1A] 2 is a diagram showing the configuration of a data processing device 1A according to the first embodiment. The data processing device 1A includes a communication unit 11, a storage unit 12, and a control unit 13. The control unit 13 includes an acquisition unit 131, a selection unit 132, and a learning unit 133.

[0030] The communication unit 11 is a communication interface for communicating with the audiovisual device 2 and the vital sensor 3 via a network.

[0031] The storage unit 12 has storage media such as a ROM (Read Only Memory), a RAM (Random Access Memory), and an SSD (Solid State Drive). The storage unit 12 stores programs executed by the control unit 13 in a non-transitory recording medium. The storage unit 12 stores user data indicating user attribute data (also referred to as user attribute data). The storage unit 12 stores data such as a user ID, name, hometown, date of birth, place of residence, and hobbies as user data.

[0032] The storage unit 12 stores a content table indicating content that can be selected by the data processing device 1A. The content table may store attribute data indicating the attributes of the content in addition to content identification information for identifying the content. The attributes of the content correspond to the properties of the content that affect the type of stimulation that the content provides to the user, such as the artist performing the music included in the content, the type of instrument used to perform the music included in the content, the frequency distribution of the sounds included in the content, the tempo, pitch, rhythm, harmony, or melody of the music included in the content, the frequency with which the user has viewed the content in the past, or the likelihood that the user has viewed the content in the past.

[0033] The storage unit 12 also stores user data in which a user ID, content identification information for identifying content provided to the user, a sleep day, and sleep data indicating the quality of the user's sleep are associated with each other. A sleep day is a day on which content corresponding to the content identification information is provided to the user. For example, if the content is provided to the user on day D and the sleep period spans day D and day D+1, the sleep day is day D. The sleep day stored in the storage unit 12 may be any day that has a certain relationship with the day on which the content corresponding to the content identification information was provided to the user, and may be the day after the content was provided to the user. Because a sleep day and a day on which the content was provided are equivalent, in the data processing device 1, data stored in the storage unit 12 as a day on which the content was provided may be used as data indicating a sleep day.

[0034] 3 is a diagram showing an example of the data structure of user data. The sleep data includes, for example, at least one of the following: sleep duration, time of falling asleep, time of waking up, number of times waking up, number of times turning over in sleep, and duration of light sleep. The storage unit 12 may store time-series data of at least one of the user's body movement state, sweat amount, and heart rate during sleep acquired from the vital sensor 3 in association with the sleep days.

[0035] When a plurality of different contents are provided to a user at a plurality of time periods in a day, the storage unit 12 may further store viewing history information indicating the relationship between the time periods when the user viewed the contents and the attributes of the contents. Instead of or together with the content attributes, the storage unit 12 may store content identification information for identifying the content provided to the user in association with the time periods when the content was provided to the user.

[0036] The control unit 13 executes the programs stored in the storage unit 12 to function as an acquisition unit 131, a selection unit 132, and a learning unit 133.

[0037] Acquiring unit 131 acquires sleep data indicating the quality of sleep of the user who has been given the content selected by selecting unit 132. Acquiring unit 131 acquires the sleep data indicating the quality of sleep of the user from the user's vital sensor 3 via communication unit 11. Acquiring unit 131 outputs the acquired sleep data to selecting unit 132 and stores the acquired sleep data in memory unit 12.

[0038] The acquisition unit 131 may further acquire surrounding environment data representing the state of the user's surrounding environment. The acquisition unit 131 acquires the surrounding environment data from an external server via the communication unit 11, for example. Here, the surrounding environment data is data indicating the sunshine hours, sunrise time, sunset time, weather, humidity, maximum temperature, or minimum temperature on each sleeping day. The acquisition unit 131 may acquire the surrounding environment data from a sensor device installed in the user's room. The acquisition unit 131 outputs the acquired surrounding environment data to the selection unit 132.

[0039] The acquisition unit 131 may further acquire activity data representing the user's activity details. The acquisition unit 131 acquires, for example, schedule information including activity data representing the user's activity details from an external server via the communication unit 11. The user's activity details include information indicating the user's daytime activities, such as gardening, reading, or yoga. The acquisition unit 131 outputs the acquired activity data to the selection unit 132.

[0040] The selection unit 132 selects a first content to be provided as a stimulus to the user from a plurality of contents. The selection unit 132 also selects a second content having an attribute different from that of the first content based on the relationship between the attribute of the first content that affects the type of stimulus provided to the user by the first content and the sleep quality indicated by the sleep data. For example, the second content has an attribute different from that of the first content, and is therefore capable of providing a stimulus different from that of the first content to the user. For example, the selection unit 132 selects the first content and the second content from a plurality of contents stored in advance in the storage unit 12.

[0041] The selection unit 132 may identify the attributes of the first content and the second content based on the titles or content identification information of the first content and the second content, for example, by referring to a content table that indicates the relationship between the titles or content identification information of the content and the attributes of the content stored in the storage unit 12. Furthermore, the selection unit 132 may identify, as the attributes of the first content and the second content, the degree of frequency with which the user has viewed the content in the past or the degree of probability that the user has viewed the content in the past, based on the relationship between the user's age indicated by the user's attributes and the year the content was released.

[0042] The selection unit 132, for example, presents a plurality of content candidates to the user and selects content selected by the user from the plurality of contents as the first content. The selection unit 132 may select the plurality of content candidates based on the user's attributes, or may select the first content based on the user's attributes. The selection unit 132 may select the plurality of content candidates or the first content further based on at least one of the user's surrounding environment indicated by the surrounding environment data or the user's activity indicated by the activity data.

[0043] For example, the selection unit 132 inputs the attributes of the first content and the sleep data into a trained model (described later) and selects, as the second content, content having the attributes output from the trained model or content corresponding to the content identification information output from the trained model. The selection unit 132 may select the second content different from the first content by inputting, into the trained model, multiple pieces of content previously provided to the user and multiple pieces of sleep data corresponding to the dates on which each piece of content was provided (e.g., the sleep data shown in FIG. 3). The sleep data corresponding to the date on which the content was provided is sleep data acquired during sleep that began after the content was provided.

[0044] The selection unit 132 may refer to a table that describes correspondence relationships between the attributes of the selected first content, the quality of sleep of the user who was given the first content, and the attributes of the second content to be selected, and select the second content that corresponds to the attributes of the first content and the quality of sleep of the user who was given the first content. For example, the selection unit 132 selects the second content that has attributes that correspond to the attributes of the selected first content and a condition that the sleep time of the user who gave the first content is less than a predetermined value.

[0045] If there is no content selected in the past, the selection unit 132 may refer to a table indicating the correspondence between the user's attribute data and the attributes of the content to be selected, and select, as the content to be provided to the user, content having attributes corresponding to the user's attribute data stored in the storage unit 12. The selection unit 132 may select the content to be provided to the user by inputting the user's attribute data stored in the storage unit 12 into a trained model created by training using the training user's attribute data, the attributes of the content provided to the training user, and the sleep data of the training user provided with the content as training data.

[0046] Selector 132 may also use the trained model when selecting the second content. Selector 132 inputs the attribute data of the first content and the sleep data measured after the first content is provided into a trained model created by training using, as training data, the training sleep data, the attributes of the training content provided to the training user whose training sleep data was measured, and the improvement results of the training user's sleep quality after the training content is provided. The improvement results are information indicating whether the values ​​of predetermined items have improved compared to past sleep days (e.g., the immediately preceding sleep day). Selector 132 selects the second content output from the trained model that corresponds to the attributes of the content that has the effect of improving sleep quality after the first content is provided.

[0047] The selection unit 132 acquires from the storage unit 12 a trained model created by the learning unit 133 by learning, as training data, content identification information of the content provided to multiple training users by the learning unit 133, attribute data indicating one or more of the content attributes such as the artist, genre, publication year, tuning, tempo, pitch, rhythm, harmony, or melody of the content provided to the training user, and the improvement results of the training user's sleep quality after the training content was provided. The training users are people who measured their sleep quality to create the trained model.

[0048] The selection unit 132 inputs the content identification information of the first content and the results of improvement in sleep quality of the user who received the first content into the trained model, and selects content that is highly correlated with the content that is effective in improving sleep quality output from the trained model as the second content. The highly correlated content is, for example, content that has the same number of attributes as a predetermined number of the content attributes.

[0049] The selection unit 132 may determine whether the quality of sleep has improved by any method. For example, the selection unit 132 determines whether the quality of sleep has improved based on whether the value of a predetermined item has improved compared to a previous sleep day (e.g., the immediately preceding sleep day), or whether the amount of change in the value of a predetermined item included in the sleep data is equal to or greater than a threshold.

[0050] The selection unit 132 may select content that is effective for other users who have a similar relationship between the content provided to the user and the sleep quality of the user who received the content. In this case, the selection unit 132 acquires from the storage unit 12 a trained model in which the learning unit 133 has learned the relationship between the content provided to each of the multiple study users and the sleep quality of each of the multiple users on the day the content was provided. Then, the selection unit 132 identifies other users whose similarity between the content provided to the user and the improvement in sleep quality after receiving the content is higher than a threshold. The selection unit 132 may select content to provide as a stimulus to the user from one or more pieces of content that have been effective for other users with high similarity.

[0051] The selection unit 132 may select the first content and the second content without receiving a content selection instruction from the user. The selection unit 132 acquires content identification information of the first content and sleep data of the user who was given the first content at a predetermined timing without involving a content selection operation by the user. The selection unit 132 then selects the second content based on the attribute data of the first content identified based on the content identification information of the first content and the sleep data of the user who was given the first content. In this way, the selection unit 132 is configured to select content without receiving a selection operation, so that content that improves sleep quality can be provided even to users who are unfamiliar with device operations.

[0052] The selection unit 132 selects the second content when the quality of sleep indicated by the second sleep data, which is sleep data for a second sleep period including one or more sleep days after the first content is provided to the user as a stimulus, is not improved compared to the quality of sleep indicated by the first sleep data, which is sleep data for a first sleep period including one or more sleep days before the first content is provided to the user as a stimulus. The selection unit 132 compares the first sleep data, which is sleep data of the user for the first sleep period, which is a predetermined period before the first content is provided, with the second sleep data, which is sleep data of the user for the second sleep period, which is a predetermined period during which the first content is provided, to determine whether the quality of sleep has improved. The selection unit 132 selects the second content when the quality of sleep indicated by the second sleep data is not improved compared to the first sleep data. The predetermined period may be, for example, one day, three days, or one week. The lengths of the first sleep period and the second sleep period may be different.

[0053] The selection unit 132 selects the second content if the quality of sleep indicated by the sleep data after the first content is given to the user as a stimulus is worse than a predetermined threshold. The selection unit 132 selects the second content if a predetermined item among items included in the user's sleep data for a predetermined period during which the first content is given is equal to or less than the threshold. The predetermined threshold is, for example, a target value set for each user or a value obtained by adding a predetermined value to the value of an item included in the user's sleep data before the content is given. For example, if the threshold is set to be four hours or less for the user's average sleep time, the selection unit 132 may select the second content if the user's average sleep time for a predetermined period included in the first sleep data is four hours or less.

[0054] The selection unit 132 may select the first content and the second content further based on attribute data indicating the user's attributes. For example, the selection unit 132 selects the first content and the second content suitable for the user's attributes by referring to data associating the user's attributes, such as the hometown or residence, date of birth or age, or hobbies of each of a plurality of study users, with content suitable for improving the quality of sleep of users having each attribute. The selection unit 132 selects, for example, content whose lyrics include the user's hometown, content that is likely to be liked by people of the user's generation, or content related to the user's hobbies. The selection unit 132 selecting such content increases the user's satisfaction and makes it easier for the user to sleep soundly.

[0055] The selection unit 132 may select second content that is suitable for the user's attributes and that may improve the quality of sleep, based on the relationship between the attributes of the first content selected based on the user's attributes and sleep data indicating the quality of sleep of the user who was given the first content. As an example, if the quality of sleep after giving the user the first content selected based on a first attribute (e.g., age) of the user's attributes is at or above an acceptable level, the selection unit 132 selects second content that has the same first attribute as the first content but is different from the first content. By the selection unit 132 selecting such second content, the user can maintain good quality of sleep without repeatedly viewing the same content.

[0056] On the other hand, if the quality of sleep after providing the user with the first content selected based on a first attribute (e.g., age) among the user attributes is below an acceptable level, the selection unit 132 selects the second content based on a second attribute (e.g., hometown) different from the first attribute. By operating in this manner, the selection unit 132 is more likely to select the second content that can improve the user's quality of sleep.

[0057] The selection unit 132 may use a trained model created by learning using as training data the relationship between the attribute data, the plurality of contents given to each of the plurality of training users, and the improvement results of the sleep quality of each of the plurality of training users on the day the contents were given to them, to select the second content based on the sleep quality indicated by the sleep data measured after the user who was given the first content sleeps and the attributes of the user. In this case, the selection unit 132 inputs the user's attribute data, the first content given to the user, and the user's sleep data measured after the first content was given to the trained model stored in the storage unit 12. The selection unit 132 selects the content output from the trained model as the second content to be given to the user as a stimulus.

[0058] Incidentally, optimizing the time at which content is provided to the user may further improve the quality of sleep. Therefore, the selection unit 132 may select multiple second contents having different attributes corresponding to multiple time periods. Specifically, the selection unit 132 selects multiple second contents such that a combination of attributes of the multiple second contents provided to the user for each of the multiple time periods is suitable for improving sleep quality. For example, the selection unit 132 selects second content with a fast tempo during the daytime and second content with a slow tempo during the nighttime. The multiple time periods may be time periods corresponding to user activities, such as upon waking up, after breakfast, after lunch, after dinner, or before bedtime, or may be time periods determined by the time of day.

[0059] The selection unit 132 may select second content to be provided to the user in each of a plurality of time periods. For example, for a user who tends to fall asleep during the day, the selection unit 132 selects second content with a relatively fast tempo during the daytime, and for a user who tends to be restless during the day, selects second content with a relatively slow tempo during the daytime. By configuring the selection unit 132 in this way, second content with attributes suitable for the user's attributes can be provided to each time period, which makes it easier to improve the quality of sleep.

[0060] The selection unit 132 may use the trained model to select multiple pieces of second content corresponding to multiple time periods. In this case, the selection unit 132 acquires from the storage unit 12 a trained model created by learning, using as training data, attribute data of the content provided to the study user by the learning unit 133, the sleep data of the study user, information indicating the time period during which the content was provided to the study user, and the improvement in the study user's sleep quality after the study content was provided. The selection unit 132 inputs the attribute data of the content provided to the user and the sleep data of the user who provided the content into the trained model. The selection unit 132 acquires information output from the trained model regarding multiple time periods during which content effective in improving sleep quality is provided. The selection unit 132 selects one or more pieces of second content to provide to the user for one or more time periods indicated by the acquired information.

[0061] The selection unit 132 may select second content to be provided to the user in the surrounding environment indicated by the surrounding environment data by referring to data indicating the influence of the user's surrounding environment on the user's mind and body and data indicating the content of stimuli suitable for the user's mental and physical state. The selection unit 132 refers to a table storing the relationship between the data indicating the influence of the user's surrounding environment on the user's mind and body, which is stored in the storage unit 12, and the data indicating the content of stimuli suitable for the user's mental and physical state, and identifies the content of stimuli suitable for the user's mental and physical state corresponding to the surrounding environment data acquired from the acquisition unit 131. The selection unit 132 selects content corresponding to the content of stimuli suitable for the user's mental and physical state as the second content.

[0062] As an example, if the ambient environment data indicates that the temperature is high, it is expected that the user will be easily fatigued, and therefore "music with a slow tempo" is determined to be a stimulus suitable for the user's mental and physical state. The selection unit 132 selects content corresponding to "music with a slow tempo." Furthermore, if the ambient environment data indicates that it is rainy, it is expected that the user will be easily depressed, and therefore "music with a fast tempo" is determined to be a stimulus suitable for the user's mental and physical state. The selection unit 132 selects content corresponding to "music with a fast tempo."

[0063] The selection unit 132 may use the trained model to select content suitable for the surrounding environment. In this case, the selection unit 132 acquires from the storage unit 12 a trained model created by the learning unit 133 learning using, as training data, attribute data of the study content provided to each of the multiple study users, the surrounding environment on the day the multiple study users received the study content, and the improvement results of the sleep quality of each of the multiple users on the day the content was provided. Then, the selection unit 132 inputs the attribute data of the first content provided to the user, the user's sleep data measured on the day the first content was provided, and the surrounding environment data into the trained model. The selection unit 132 selects the content output from the trained model as second content to be provided to the user as a stimulus in the user's surrounding environment.

[0064] The selection unit 132 may select second content to be provided to a user who has performed an activity indicated by the activity data, by referring to data indicating the influence of the user's activity content on the user's mind and body and data indicating the content of a stimulus suitable for the user's mental and physical state. The selection unit 132 refers to a table stored in the storage unit 12 that stores the relationship between the data indicating the influence of the user's activity content on the user's mind and body and the data indicating the content of a stimulus suitable for the user's mental and physical state, and acquires data indicating the content of a stimulus suitable for the mental and physical state of the user who performed an activity indicated by the activity data acquired from the acquisition unit 131. The selection unit 132 selects content corresponding to the data indicating the content of a stimulus suitable for the user's mental and physical state as the second content.

[0065] As an example, when the activity data indicates that the user has performed strenuous exercise, the selection unit 132 assumes that the user is likely to be affected by fatigue and therefore determines that the content of the stimulus suitable for the user's mental and physical state is "music with a slow tempo." The selection unit 132 selects content corresponding to "music with a slow tempo." Furthermore, when the activity data indicates that the user has not gone out much, the selection unit 132 assumes that the user is likely to be affected by depression and therefore determines that the content of the stimulus suitable for the user's mental and physical state is "music with a fast tempo." The selection unit 132 selects content corresponding to "music with a fast tempo."

[0066] The selection unit 132 may use the trained model to select content suitable for the activity details of the user. In this case, the selection unit 132 acquires from the storage unit 12 a trained model created by the learning unit 133 learning using as training data attribute data of the first content and the second content given to each of the multiple study users, the activity details performed by the multiple study users before being given the second content, and the results of improvement in sleep quality for each of the multiple study users on the day they were given the second content after the first content. Then, the selection unit 132 inputs the attribute data of the first content given to the user, the user's sleep data measured on the day the first content was given, and activity detail data indicating the activity details performed by the user before being given the second content into the trained model. The selection unit 132 selects the content output from the trained model as the second content to be given to the user who performed the activity indicated by the activity detail.

[0067] The learning unit 133 generates the various trained models described above, including a trained model that has learned the relationship between the content provided to the training user and the quality of sleep of the training user to whom the content was provided, and stores the generated trained models in the memory unit 12.

[0068] [Processing flow in data processing device 1A] 4 is a flowchart showing the flow of processing in the data processing device 1 A. The processing in this flowchart starts, for example, when the user registration process is completed and content selection becomes possible.

[0069] The selection unit 132 acquires user attribute data stored in the storage unit 12 (S01). Next, the selection unit 132 inputs the user attribute data into the trained model acquired from the storage unit 12, and selects the output content as the first content to be provided as a stimulus to the user (S02).

[0070] The acquisition unit 131 acquires sleep data of the user who has received the first content as a stimulus from the vital sensor 3 (S03). The selection unit 132 determines whether the quality of sleep indicated by the acquired sleep data has improved (S04). Next, the selection unit 132 determines whether an end condition is met (S05). The end condition is, for example, when the user has achieved a desired sleep quality. If the end condition is met (YES in S05), the data processing device 1 ends the content selection process. If the end condition is not met (NO in S05), the process proceeds to S06.

[0071] If the sleep quality indicated by the sleep data has improved (YES in S06), proceed to S03. If the sleep quality indicated by the sleep data has not improved (NO in S06), the selection unit 132 inputs information identifying the first content and the sleep quality of the user who was given the first content into the trained model acquired from the storage unit 12, and selects the content output from the trained model as the second content (S07). Thereafter, the data processing device 1 provides the selected second content to the user as the first content, and repeats the content selection process until the termination condition is satisfied.

[0072] [Advantages of the first embodiment] As described above, in data processing device 1A, selection unit 132 selects first content to be provided as a stimulus to the user from a plurality of contents. Next, acquisition unit 131 acquires sleep data indicating the quality of sleep of the user who has been provided with the first content selected by selection unit 132. Then, selection unit 132 selects second content having attributes different from those of the first content based on the relationship between the attributes of the stimulus provided to the user by the first content and the quality of sleep indicated by the sleep data. By configuring data processing device 1A in this way, data processing device 1A can select content that has a therapeutic effect on the user.

[0073] [Variations] Even if improvement in sleep quality is confirmed, it may be that the improvement in sleep quality is due to factors other than the content, such as exercise, weather, etc. Therefore, the data processing device 1 may calculate a sleep score indicating sleep quality, evaluate the influence of factors other than the content, and correct the sleep score based on the influence of factors other than the content.

[0074] FIG. 5 is a diagram showing the configuration of a data processing device 1B according to a modified example. Functional units equivalent to those already described in FIG. 5 are assigned the same reference numerals, and their descriptions are omitted. Data processing device 1B further includes an evaluation unit 134. Evaluation unit 134 generates an evaluation result indicating the degree of effect of providing the first content to the user based on the relationship between the sleep quality indicated by the first sleep data and the sleep quality indicated by the second sleep data. Here, the first sleep data is the user's sleep data during the first sleep period, which is a predetermined period before the first content is provided. Furthermore, the second sleep data is the user's sleep data during the second sleep period, which is a predetermined period during which the first content is provided.

[0075] The evaluation unit 134 calculates a sleep score indicating the quality of sleep based on, for example, the time of falling asleep, the time of waking up, the duration of sleep, and the amount of body movement or heart rate during sleep. Then, by subtracting the second sleep score calculated from the second sleep data from the sleep score calculated from the first sleep data, an evaluation result indicating the degree of effect that the content had on improving the sleep data is calculated. The evaluation unit 134 may calculate a difference rate as the evaluation result, by dividing the difference between the first sleep score and the second sleep score by the first sleep score. Alternatively, the evaluation unit 134 may use average values ​​of sleep data acquired over a predetermined period as the first sleep data and the second sleep data. The predetermined period is, for example, three days or one week.

[0076] The evaluation unit 134 may correct the evaluation result using the surrounding environment data. For example, the evaluation unit 134 identifies the degree to which the surrounding environment indicated by the surrounding environment data affected sleep quality by referring to data indicating the relationship between the surrounding environment and sleep quality. The evaluation unit 134 creates the evaluation result by correcting the degree of effect identified based on the relationship between the sleep quality indicated by the first sleep data and the sleep quality indicated by the second sleep data based on the identified degree of influence.

[0077] The evaluation unit 134 refers to a table that describes the relationship between the surrounding environment data acquired by the acquisition unit 131 and the correction values ​​of the evaluation results, and acquires the correction values ​​of the evaluation results that correspond to the surrounding environment data on the day for which the sleep score is to be calculated, which are input from the acquisition unit 131. The evaluation unit 134 corrects the first sleep score and the second sleep score by adding the acquired correction values ​​to the calculated sleep scores.

[0078] The evaluation unit 134 may correct the evaluation result based on the content of the user's daytime activities. The evaluation unit 134 identifies the degree to which the activity content indicated by the activity data affected the quality of sleep by referring to data indicating the relationship between the activity content and the quality of sleep. The evaluation unit 134 creates the evaluation result by correcting the degree of effect identified based on the relationship between the quality of sleep indicated by the first sleep data and the quality of sleep indicated by the second sleep data based on the identified degree of influence.

[0079] The evaluation unit 134 acquires information indicating the user's activity details included in the schedule information acquired by the acquisition unit 131. The evaluation unit 134 references a table that describes the relationship between the user's activity details and correction values ​​of the evaluation results, and acquires a correction value corresponding to the user's activity details on the day for which the sleep score is to be calculated. The evaluation unit 134 corrects the first sleep score and the second sleep score by adding the acquired correction value to the sleep score.

[0080] In the data processing device 1B, the selection unit 132 calculates the degree of improvement in sleep quality using the corrected sleep scores acquired from the evaluation unit 134. As an example, the selection unit 132 calculates the degree of improvement in sleep quality as the value obtained by subtracting the corrected second sleep score from the corrected first sleep score.

[0081] The learning unit 133 generates a trained model that is trained using the sleep data of the training user acquired by the acquisition unit 131, the training content selected by the selection unit 132, and the corrected sleep score of the training user evaluated by the evaluation unit 134 as training data, and stores the trained model in the memory unit 12.

[0082] By configuring the data processing device 1B to include an evaluation unit 134, it is possible to correct for the influence of factors other than the content on the sleep score, determine with greater accuracy whether the content has contributed to improving the quality of sleep, and improve the accuracy of content selection.

[0083] <Second embodiment> In the first embodiment, the quality of the user's sleep was measured using sleep data, but the quality of the user's sleep may also be determined using data indicating the caregiver's workload. Fig. 6 is a diagram showing the device configuration of a data processing device 1C according to a second embodiment. In Fig. 6, functional units equivalent to those already described are assigned the same reference numerals, and their description will be omitted.

[0084] The data processing device 1C includes a task data acquisition unit 135. The task data acquisition unit 135 acquires task data indicating the task content of a caregiver who is caring for the user to whom the first content selected by the selection unit 132 has been given. The task data acquisition unit 135 acquires task data of the caregiver who is caring for the user. The task data is, for example, daily report information in which the task content of the caregiver is recorded.

[0085] The selection unit 132 estimates the quality of the user's sleep based on the caregiver's work content indicated by the work data, and selects second content having attributes different from those of the first content based on the relationship between the attributes of the stimuli provided to the user by the first content and the estimated quality of sleep. If the work data is text data, the selection unit 132 extracts information indicating the caregiver's burden described in the daily report by performing natural language processing on the work record data. Information indicating the caregiver's burden is, for example, information indicating the patient's symptoms associated with dementia-related symptoms. As an example, the information indicating the caregiver's burden is information indicating whether or not the caregiver has engaged in violence against the caregiver, information indicating whether the user has shouted or acted violently, information indicating whether the user has complained of hallucinations, or information indicating the number of times the caregiver has been called for nighttime toileting or the number of times the caregiver has assisted with toileting.

[0086] The selection unit 132 estimates the quality of sleep of the user who has been given the first content, for example, by referring to data indicating the relationship between the work content of the caregiver and the quality of sleep of the user (care recipient), which is stored in the storage unit 12. The selection unit 132 may estimate the quality of sleep of the user by inputting information indicating the load of the caregiver to a trained model created by learning information indicating the load of the caregiver of the learning user and the quality of sleep of the learning user as training data.

[0087] The selection unit 132 selects second content suitable for a user who, after receiving the first content, needs a caregiver to perform the task indicated by the task data using the above-described method. The selection unit 132 acquires from the storage unit 12 a trained model created by the learning unit 133 using, as training data, the first and second content provided to each of the multiple study users, task data indicating the task performed by the caregiver after the multiple study users were given the first content, and the results of improvement in sleep quality for each of the multiple users on the day the first content was given to the second content. The selection unit 132 then inputs, into the trained model, attribute data of the first content provided to the user and the task data indicating the task performed by the caregiver after the first content was given to the user. The selection unit 132 selects the content output from the trained model as the second content.

[0088] [Processing flow in data processing device 1C] 7 is a flowchart showing the flow of processing in the data processing device 1C. The processing in this flowchart starts, for example, when the user registration process is completed and content selection becomes possible.

[0089] The selection unit 132 acquires user attribute data stored in the storage unit 12 (S11). Next, the selection unit 132 inputs the user attribute data into the trained model acquired from the storage unit 12, and selects the output content as first content to be provided as a stimulus to the user (S12).

[0090] The acquisition unit 131 acquires task data of a caregiver who is caring for a user (S13). The selection unit 132 extracts information indicating the workload of the caregiver from the acquired task data, and estimates the user's sleep quality based on the information indicating the workload (S14). The selection unit 132 determines whether the estimated sleep quality has improved (S15).

[0091] Next, the selection unit 132 determines whether a termination condition is satisfied (S16). The termination condition is, for example, when the user has reached a desired sleep quality. If the termination condition is satisfied (YES in S16), the data processing device 1C terminates the content selection process. If the termination condition is not satisfied (NO in S16), the process proceeds to S17.

[0092] If the sleep quality indicated by the sleep data has improved (YES in S17), proceed to S13. If the sleep quality indicated by the sleep data has not improved (NO in S17), input information identifying the first content and the sleep quality of the user who was given the first content into the trained model obtained from the storage unit 12, and select the output content as the second content (S18). Thereafter, the data processing device 1C provides the selected second content to the user as the first content, and repeats the content selection process until the termination condition is satisfied.

[0093] [Effects of Data Processing Device 1C] As described above, in data processing device 1C, selection unit 132 selects first content from multiple contents to provide as a stimulus to the user. Next, activity data acquisition unit 135 acquires activity data indicating the activity details of the caregiver caring for the user to whom the first content selected by selection unit 132 has been provided. Selection unit 132 then estimates the user's sleep quality based on the activity details indicated by the activity data, and selects second content having attributes different from those of the first content based on the relationship between the attributes of the first content that affect the content of the stimulus provided to the user and the estimated sleep quality. This configuration of data processing device 1C enables data processing device 1C to select content that has a therapeutic effect even for users for whom sleep data cannot be acquired.

[0094] 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 changes are possible within the scope of the gist of the present 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 the new embodiments resulting from the combination also have the effects of the original embodiments. [Explanation of symbols]

[0095] 1 Data Processing Device 2. Viewing devices 3 Vital Sensors 11 Communications Department 12 Storage section 13 Control Unit 131 Acquisition Department 132 Selection section 133 Learning Department 134 Evaluation Department 135 Work data acquisition unit

Claims

1. a selection unit that selects a first content to be given as a stimulus to the care recipient from a plurality of contents; a work data acquisition unit that acquires work data in which work details of a caregiver who is caring for the care recipient are recorded; a storage unit that stores relationship data indicating a relationship between the work content and the quality of sleep of the care recipient; and the selection unit estimates the quality of sleep of the care recipient when the task recorded in the task data occurs by referring to the relationship data, and selects second content having attributes different from those of the first content to be provided to the care recipient after the first content, based on the relationship between the attribute of the first content that influences the stimulus provided to the care recipient by the first content and the estimated quality of sleep. Data processing device.

2. The selection unit extracts information indicating the caregiver's burden by natural language processing the work data, and estimates the quality of sleep of the care recipient by inputting the information indicating the caregiver's burden into a trained model created by learning information indicating the caregiver's burden of the learning user and the learning user's sleep quality as training data.

2. The data processing device according to claim 1.

3. The storage unit stores sleep data indicating the quality of sleep of the care recipient in association with a sleep day; the selection unit selects the second content when a quality of sleep indicated by second sleep data, which is sleep data during a second sleep period including one or more sleep days after the first content is given as a stimulus to the care recipient, is not improved compared to a quality of sleep indicated by first sleep data, which is sleep data during a first sleep period including one or more sleep days before the first content is given as a stimulus to the care recipient.

3. A data processing device according to claim 1 or 2.

4. an evaluation unit that generates an evaluation result indicating a degree of effect of providing the first content to the care recipient based on a relationship between a sleep quality indicated by the first sleep data and a sleep quality indicated by the second sleep data; 4. The data processing device according to claim 3.

5. The apparatus further includes an acquisition unit for acquiring ambient environment data indicating humidity, maximum temperature, or minimum temperature measured in the room where the care recipient is present; the evaluation unit identifies the degree to which the surrounding environment indicated by the surrounding environment data affected the quality of sleep by referring to data indicating the relationship between the surrounding environment and the quality of sleep, and creates the evaluation result by correcting the degree of effect identified based on the relationship between the quality of sleep indicated by the first sleep data and the quality of sleep indicated by the second sleep data based on the identified degree of influence.

5. The data processing device according to claim 4.

6. The method further includes an acquisition unit that acquires activity data representing the activity content of the care recipient indicated by schedule information acquired from an external server, The evaluation unit identifies the degree to which the activity content indicated by the activity data affected the quality of sleep by referring to data indicating the relationship between the activity content and the quality of sleep, and creates the evaluation result by correcting the degree of effect identified based on the relationship between the quality of sleep indicated by the first sleep data and the quality of sleep indicated by the second sleep data based on the identified degree of influence.

5. The data processing device according to claim 4.

7. the selection unit selects the second content when sleep quality corresponding to sleep data indicating sleep quality of the care recipient after the first content is given as a stimulus to the care recipient is worse than a predetermined threshold.

3. A data processing device according to claim 1 or 2.

8. the selection unit inputs attribute data of the first content and sleep data indicating the quality of sleep of the care recipient measured after the first content is provided into a trained model created by learning using, as teacher data, the training sleep data, attributes of the training content provided to the training user whose training sleep data was measured, and an improvement result of the sleep quality of the training user after the training content is provided, and selects the second content corresponding to the attributes output by the trained model. A data processing device according to any one of claims 1 to 7.

9. the selection unit inputs attribute data of the first content, user attribute data of the care recipient, and sleep data indicating the quality of sleep of the care recipient measured after the first content is provided into a trained model created by learning using, as teacher data, the training sleep data, attributes of the training content provided to the training user whose training sleep data was measured, attributes of the training user, and an improvement result of the sleep quality of the training user after the training content is provided, thereby selecting the second content corresponding to the attributes output by the trained model; A data processing device according to any one of claims 1 to 7.

10. the selection unit selects the first content and the second content without receiving a content selection instruction from the care receiver.

10. A data processing device according to any one of claims 1 to 9.

11. the selection unit selects a plurality of second contents having different attributes corresponding to a plurality of time periods, A data processing device according to any one of claims 1 to 10.

12. Further comprising an acquisition unit for acquiring surrounding environment data representing the situation of the surrounding environment of the care recipient, the selection unit selects the second content to be provided to the care recipient in the surrounding environment indicated by the surrounding environment data by referring to data indicating an influence that the surrounding environment of the care recipient has on the mind and body of the care recipient and data indicating a stimulus content suitable for the mind and body state of the care recipient.

5. A data processing device according to claim 1.

13. An acquisition unit that acquires activity data representing activity details of the care recipient, the selection unit selects the second content to be provided to the care recipient who has performed the activity indicated by the activity data by referring to data indicating an influence that the activity content of the care recipient has on the mind and body of the care recipient and data indicating a stimulus content appropriate for the mind and body state of the care recipient.

5. A data processing device according to claim 1.

14. The computer executes selecting a first content to be given as a stimulus to the care recipient from a plurality of contents; acquiring work data recording work details of a caregiver who is caring for the care recipient; estimating the quality of sleep of the care recipient when the task recorded in the task data occurs by referring to relationship data indicating a relationship between the task content and the quality of sleep of the care recipient; selecting second content having attributes different from those of the first content to be provided to the care recipient after the first content, based on a relationship between an attribute of the first content that influences the type of stimulation provided to the care recipient by the first content and the estimated quality of sleep; A data processing method comprising:

15. On the computer, selecting a first content to be given as a stimulus to the care recipient from a plurality of contents; acquiring work data recording work details of a caregiver who is caring for the care recipient; estimating the quality of sleep of the care recipient when the task recorded in the task data occurs by referring to relationship data indicating a relationship between the task content and the quality of sleep of the care recipient; selecting second content having attributes different from those of the first content to be provided to the care recipient after the first content, based on a relationship between an attribute of the first content that influences the type of stimulation provided to the care recipient by the first content and the estimated quality of sleep; A program to execute.

Images