Information processing system, information processing method, and information processing apparatus

Abstract

The present technology relates to an information processing system, an information processing method, and an information processing apparatus that enable assignment of an appropriate tag to content. An information processing system according to the present technology comprises: an emotion estimation unit that estimates the emotional state of a user on the basis of emotion value data that is time-series data of an emotion value estimated on the basis of a biological signal of a user; and a tag assignment unit that assigns, to content, a tag including an emotional state label, which is a label based on the emotional state of the user estimated on the basis of the emotion value data when the content is being presented to the user. The present technology can be applied to, for example, a content recommendation system.

Description

Information processing system, information processing method, and information processing device

[0001] The present technology relates to an information processing system, an information processing method, and an information processing device, and in particular to an information processing system, an information processing method, and an information processing device that are suitable for use when adding tags to content.

[0002] A technology has been proposed in the past that identifies a user's current emotional state and a target emotional state, and generates a music playlist to generate a trajectory of the user's emotional state from the current state to the target state (see, for example, Patent Document 1).

[0003] Special Publication No. 2023-516135

[0004] However, the invention described in Patent Document 1 does not consider adding tags to content such as music.

[0005] The present technology has been developed in light of such circumstances, and makes it possible to assign appropriate tags to content such as music, and also makes it possible to present content based on the tags, for example.

[0006] An information processing system according to a first aspect of the present technology includes an emotion estimation unit that estimates an emotional state of a user based on emotion value data, which is time-series data of emotion values ​​estimated based on a biometric signal of the user, and a tag assignment unit that assigns to the content a tag including an emotional state label, which is a label based on the emotional state of the user estimated based on the emotion value data when the content is presented to the user.

[0007] An information processing method according to a first aspect of the present technology includes an information processing system that estimates an emotional state of a user based on emotional value data, which is time-series data of emotional values ​​estimated based on a biometric signal of the user, and that assigns to the content a tag that includes an emotional state label that is based on the emotional state of the user estimated based on the emotional value data at the time the content is presented to the user.

[0008] An information processing device according to a second aspect of the present technology includes: an emotion estimation unit that estimates an emotional state of a user based on emotion value data, which is time-series data of emotion values ​​estimated based on a biometric signal of the user; and a tag assignment unit that assigns to the content a tag including an emotional state label based on the emotional state of the user estimated based on the emotion value data when the content is presented to the user.

[0009] In a first or second aspect of the present technology, a user's emotional state is estimated based on emotional value data, which is time-series data of emotional values ​​estimated based on the user's biosignals, and a tag including an emotional state label based on the user's emotional state estimated based on the emotional value data at the time the content is presented to the user is assigned to the content.

[0010] Fig. 1 is a block diagram showing an example configuration of an information processing system to which the present technology is applied; Fig. 2 is a flowchart for explaining processing of the information processing system; Fig. 3 is a graph showing an example of time series data of arousal level; Fig. 4 is a graph showing an example of time series data of emotional valence; Fig. 5 is a diagram showing an example of a method for presenting content; Fig. 6 is a diagram showing an example of a method for presenting content; Fig. 7 is a block diagram showing an example configuration of a computer;

[0011] Hereinafter, embodiments of the present technology will be described. The description will be made in the following order: 1. Embodiment 2. Modification 3. Other

[0012] <<1. Embodiment>> <Configuration Example of Information Processing System> FIG. 1 is a block diagram showing an embodiment of an information processing system 1 to which the present technology is applied.

[0013] The information processing system 1 includes a cloud system 11 , an information processing terminal 12 , and a wearable device 13 .

[0014] The cloud system 11 and the information processing terminal 12 communicate with each other via a network such as the Internet (not shown). The information processing terminal 12 and the wearable device 13 communicate with each other via wired or wireless communication.

[0015] In this example, for ease of understanding, one information processing terminal 12 and one wearable device 13 are illustrated, but the number of information processing terminals 12 and wearable devices 13 is not particularly limited.

[0016] The cloud system 11 is configured by, for example, one or more servers, etc. The cloud system 11 includes a content database 21, a property estimation unit 22, and a presentation control unit 23.

[0017] The content database 21 stores data related to content that can be presented to users (hereinafter referred to as content data). The content data includes, for example, image data used to play images (moving or still images) of the content, audio data used to play audio of the content, and content meta information. The content meta information includes, for example, tags that indicate the characteristics of the content.

[0018] The type of content is not particularly limited, and examples include video content such as movies, TV programs, music videos, and SNS (Social Networking Service) content, and audio content such as music.

[0019] The characteristic estimation unit 22 estimates the characteristics of the content based on the meta information of the content stored in the content database 21. The characteristic estimation unit 22 adds the estimated characteristics of the content to the meta information of the content.

[0020] The presentation control unit 23 controls the presentation of content to the user based on meta information of the content stored in the content database 21 .

[0021] For example, the presentation control unit 23 extracts content to be presented (e.g., recommended) to the user from the content stored in the content database 21 based on conditions (hereinafter referred to as “instructed conditions”) and meta information specified by the user on the information processing terminal 12. The presentation control unit 23 generates information about the extracted content (hereinafter referred to as “presented content information”) and transmits it to the information processing terminal 12.

[0022] For example, the presentation control unit 23 acquires content data of content to be presented to the user from the content database 21 based on the instruction conditions and meta information specified by the user on the information processing terminal 12. The presentation control unit 23 transmits the acquired content data to the information processing terminal 12.

[0023] The information processing terminal 12 is configured by, for example, a smartphone, a tablet terminal, a personal computer, a music player, a game terminal, etc. The information processing terminal 12 includes a UI (user interface) unit 31, a playback unit 32, a playback history generation unit 33, an emotion estimation unit 34, and a tag assignment unit 35.

[0024] The UI unit 31 includes various input devices and various output devices that realize a user interface of the information processing terminal 12. For example, the UI unit 31 is used to operate the information processing terminal 12 and input various types of information. For example, the UI unit 31 is used to output various types of information (for example, visual information, auditory information, and tactile information).

[0025] The playback unit 32 plays back the content data received from the presentation control unit 23 of the cloud system 11, thereby generating an image signal for outputting an image included in the content and an audio signal for outputting audio included in the content. The playback unit 32 transmits the image signal and the audio signal to the wearable device 13. The playback unit 32 also generates information about the played content (hereinafter referred to as played content information) and supplies it to the playback history generation unit 33.

[0026] The playback history generating unit 33 generates a playback history of the content based on the playback content information, and supplies the playback history of the content to the tagging unit.

[0027] The emotion estimation unit 34 estimates the emotion of the user based on the user's biological signal received from the wearable device 13. The emotion estimation unit 34 includes an emotion value estimation unit 41 and an emotional state estimation unit 42.

[0028] The emotion value estimation unit 41 estimates an emotion value that indicates an index related to the user's emotion based on the user's biosignal. The emotion value estimation unit 41 supplies time-series data of the estimated emotion value of the user (hereinafter referred to as emotion value data) to the emotional state estimation unit 42.

[0029] The emotional state estimation unit 42 estimates time-series changes in the user's emotional state based on the emotional value data. The emotional state estimation unit 42 also calculates the reliability of the estimation result of the user's emotional state (hereinafter referred to as emotional reliability) based on the emotional value data. The emotional state estimation unit 42 supplies the estimation result of the user's emotional state and information indicating the emotional reliability to the tag assignment unit 35.

[0030] The tagging unit 35 generates tags based on the user's emotional state based on the content playback history, the estimation result of the user's emotional state, and the emotional reliability. The tagging unit 35 assigns the generated tags to content data stored in the content database 21 as meta information.

[0031] The presentation control unit 23 of the cloud system 11, and the playback unit 32 and tagging unit 35 of the information processing terminal 12 present content to the user, tag the content based on biometric signals indicating the user's reaction to the presented content, and present the content based on the tagged tag, thereby forming a closed loop.

[0032] The wearable device 13 is configured by a device that a user wears and can see and hear content, such as headphones, earphones, a head-mounted display (HMD), etc. The wearable device 13 includes an output unit 51, a biosensor 52, and a signal processing unit 53.

[0033] The output unit 51 includes various output devices that output at least one of an image and an audio signal. The output unit 51 outputs an image based on an image signal received from the playback unit 32 of the information processing terminal 12, and outputs an audio signal based on an audio signal received from the information processing terminal 12.

[0034] The biosensor 52 is configured by, for example, a TWS (True Wireless Stereo) type vital sensor such as a PPG (Photoplethysmography) sensor. The biosensor 52 detects a biosignal used to estimate the emotion of the user wearing the wearable device 13. The biosensor 52 supplies the detected biosignal to the signal processing unit 53.

[0035] The signal processing unit 53 performs various signal processing on the biological signal, and transmits the processed biological signal to the information processing terminal 12 .

[0036] <Processing of Information Processing System 1> Next, processing of the information processing system 1 will be described with reference to the flowchart of FIG.

[0037] In the following, a case where music is presented to the user as content will be described as a specific example.

[0038] In step S1, the information processing system 1 presents content. For example, the playback unit 32 downloads content data from the content database 21 in accordance with a user operation on the UI unit 31. The playback unit 32 transmits at least one of an image signal and an audio signal generated by playing the downloaded content data to the output unit 51. The output unit 51 outputs at least one of an image and an audio included in the content based on at least one of the image signal and the audio signal.

[0039] For example, by repeating this process, a plurality of contents are presented to the user in succession.

[0040] The playback unit 32 also supplies information about the played content to the playback history generation unit 33. The playback history generation unit 33 generates a playback history of the content based on the information acquired from the playback unit 32. The playback history includes, for example, identification information for identifying the content, the title of the content, and the playback period of the content. In other words, the playback period of the content is the presentation period during which the content was presented to the user, and includes, for example, the date and time when playback of the content started and the date and time when playback of the content ended. The playback history generation unit 33 supplies the playback history to the tag assignment unit 35.

[0041] In step S2, the biosensor 52 acquires a biosignal of the user. For example, the biosensor 52 detects a pulse wave signal of the user while content is being presented, in other words, while the user is viewing or listening to the content, and supplies the signal to the signal processing unit 53. The signal processing unit 53 performs various signal processing on the pulse wave signal, such as noise removal, amplification, and A / D conversion. The signal processing unit 53 supplies the processed pulse wave signal to the emotion estimation unit 34.

[0042] For example, the process of step S2 is executed in parallel with the process of step S1.

[0043] In step S3, the emotion value estimation unit 41 estimates the emotion value of the user. For example, the emotion value estimation unit 41 estimates the emotion value of the user based on the user's pulse wave signal. For example, the emotion value estimation unit 41 estimates arousal and valence as the emotion values ​​of the user.

[0044] The method for estimating the emotion value is not particularly limited. For example, the emotion value is estimated using a learning model trained using machine learning.

[0045] The emotional value estimation unit 41 supplies the time series data of the user's arousal level (hereinafter referred to as arousal level data) and the time series data of the emotional valence (hereinafter referred to as emotional valence data) to the emotional state estimation unit 42.

[0046] In step S4, the emotional state estimation unit 42 estimates the emotional state of the user.

[0047] For example, the emotional state estimation unit 42 divides the emotional value data (e.g., arousal data, emotional valence data, etc.) into multiple intervals that serve as units for estimating the user's emotional state based on time-series changes in the emotional value data. For example, emotional value data has the characteristic that the emotional value does not change significantly and a stable steady state (plateau) occurs. The emotional state estimation unit 42 divides the emotional value data into multiple intervals using the steady state of the emotional value data. The emotional state estimation unit 42 estimates the user's emotional state for each interval based on at least one of the emotional value for each interval and the relative relationship of the emotional values ​​between each interval.

[0048] An example of a method for estimating the emotional state of a user will now be described with reference to FIGS.

[0049] 3 shows an example of a graph of alertness data. The horizontal axis of the graph represents time, and the vertical axis represents alertness. Below the graph, the playback order and playback duration of the songs (songs A to H) presented to the user are shown.

[0050] Specifically, for example, the emotional state estimation unit 42 extracts a plateau section (steady section) from the arousal level data. A plateau section is a section in which the emotional value (in this case, arousal level) does not change significantly and remains stable (steady state).

[0051] For example, the emotional state estimation unit 42 shifts a window of a predetermined size (e.g., X seconds) in the time direction and extracts, as a plateau section, a section in which the change in arousal level within the window is less than a predetermined threshold. Alternatively, the emotional state estimation unit 42 shifts a window of a predetermined size in the time direction and extracts, as a plateau section, a section in which the value of the first derivative of arousal level within the window is less than a predetermined threshold.

[0052] Furthermore, the emotional state estimation unit 42 determines the section between adjacent plateau sections as a transition section.

[0053] For example, the emotional state estimation unit 42 may approximate each plateau section and each transition section with a straight line, and adjust the boundary between the plateau section and the transition section based on the intersection of the approximated lines.

[0054] Furthermore, for example, the emotional state estimation unit 42 may distinguish between a plateau section and a transition section based on the extreme value of the inflection point of the graph of the arousal level data.

[0055] Next, the emotional state estimation unit 42 estimates the user's emotional state (hereinafter referred to as the arousal state) in each section based on at least one of the arousal level in each section of the arousal data and the relative relationship of the arousal levels between each section.

[0056] For example, the emotional state estimation unit 42 estimates the arousal state of each plateau section based on the relative arousal levels of adjacent plateau sections in the arousal level data. For example, the emotional state estimation unit 42 classifies the arousal state of a plateau section with a higher arousal level than the immediately preceding plateau section as an excited state. For example, the emotional state estimation unit 42 classifies the arousal state of a plateau section with a lower arousal level than the immediately preceding plateau section as a relaxed state.

[0057] Hereinafter, a period in which the wakefulness state is an active state will be referred to as an active period, and a period in which the wakefulness state is a calm period will be referred to as a calm period.

[0058] For example, the emotional state estimation unit 42 estimates the arousal state of each transition section based on the change in arousal level (e.g., the slope) in each transition section. For example, the emotional state estimation unit 42 classifies the arousal state of a transition section in which arousal level is rising, i.e., a transition section in which arousal level transitions from a calm section to an active section, as an activated state. For example, the emotional state estimation unit 42 classifies the arousal state of a transition section in which arousal level is falling, i.e., a transition section in which arousal level transitions from an active section to a calm section, as a calmed state.

[0059] Hereinafter, a section in which the wakefulness state is in an activated state will be referred to as an activated section, and a section in which the wakefulness state is in a calmed state will be referred to as a calmed section.

[0060] Next, the emotional state estimation unit 42 calculates the emotional reliability (hereinafter referred to as the arousal state reliability) for the estimation result of the arousal state for each section based on at least one of the arousal level for each section and the relative relationship of the arousal levels between each section.

[0061] For example, the reliability of the alertness state of each plateau section is calculated based on the alertness level Va of the immediately preceding plateau section and the absolute value ΔVa of the difference between the alertness level of the immediately preceding plateau section and the alertness level of the immediately preceding plateau section. Specifically, the reliability of the alertness state of each plateau section is calculated using the following formulas (1) and (2).

[0062] Rae=ΔVa×Va...(1) Rar=ΔVa×(1-Va)...(2)

[0063] Rae is the reliability of the alertness state in the active section, and Rar is the reliability of the alertness state in the calm section. Note that the alertness level Va is normalized within the range of 0 to 1.

[0064] Therefore, the higher the wakefulness Va in the immediately preceding plateau section (calm section), the higher the wakefulness state reliability Rae. This is because the higher the wakefulness Va in the immediately preceding calm section, the more difficult it is for the wakefulness Va to increase, and despite this, it is estimated that the user is actually highly likely to be in an active state in the active section reached by the increase in wakefulness Va. Furthermore, the larger the absolute value ΔVa of the difference between the wakefulness Va and the wakefulness Va in the immediately preceding plateau section (calm section), the higher the wakefulness state reliability Rae. This is because the larger the increase in wakefulness Va from the immediately preceding calm section, the higher the estimated probability that the user is actually in an active state in the subsequently reached active section.

[0065] On the other hand, the wakefulness state reliability Rar increases as the wakefulness Va in the immediately preceding plateau section (active section) decreases. This is because the lower the wakefulness Va in the immediately preceding active section, the more difficult it is for the wakefulness Va to decrease. Despite this, it is estimated that the user is likely to be in a sedated state in the calm section reached by the decrease in wakefulness Va. Furthermore, the wakefulness state reliability Rar increases as the absolute value ΔVa of the difference between the wakefulness Va and the wakefulness Va in the immediately preceding plateau section (active section) increases. This is because the greater the decrease in wakefulness Va from the immediately preceding active section, the higher the estimated probability that the user is actually in a sedated state in the calm section reached thereafter.

[0066] Furthermore, the emotional state estimation unit 42 calculates the arousal state reliability of each transition section based on the change in arousal level in each transition section, for example, at least one of the direction, width, and rate of change of arousal level. For example, the emotional state estimation unit 42 calculates the reliability of each transition section based on the absolute value of the slope (rate of change) of the arousal level graph for each transition section. For example, the larger the absolute value of the slope of the arousal level graph for the transition section, the higher the arousal state reliability, and the smaller the absolute value of the slope of the arousal level graph for the transition section, the lower the arousal state reliability.

[0067] 4 shows an example of a graph of valence data. The horizontal axis of the graph represents time, and the vertical axis represents valence. Below the graph, the playback order and playback duration of the songs (songs A to H) presented to the user are shown.

[0068] For example, the emotional state estimation unit 42 divides the valence data into a plateau section and a transition section in the same manner as for the arousal level data.

[0069] Next, the emotional state estimation unit 42 estimates the user's emotional state (hereinafter referred to as the emotional state) in each section based on at least one of the valence of each section of the valence data and the relative relationship of valence between each section.

[0070] For example, the emotional state estimation unit 42 estimates the emotional state of each plateau section based on the relative valence of adjacent plateau sections in the valence data. For example, the emotional state estimation unit 42 classifies the emotional state of a plateau section whose valence is higher than that of the previous (immediately preceding) plateau section as a pleasant state (positive). For example, the emotional state estimation unit 42 classifies the emotional state of a plateau section whose valence is lower than that of the previous (immediately preceding) plateau section as an unpleasant state (negative).

[0071] Hereinafter, a section in which the emotional state is a pleasant state will be referred to as a pleasant section, and a section in which the emotional state is an unpleasant state will be referred to as an unpleasant section.

[0072] For example, the emotional state estimation unit 42 estimates the emotional state of each transition section based on the change (e.g., slope) in valence of each transition section. For example, the emotional state estimation unit 42 classifies the emotional state of a transition section in which valence is rising, i.e., a transition section transitioning from an unpleasant section to a pleasant section, as a positive state. For example, the emotional state estimation unit 42 classifies the emotional state of a transition section in which valence is falling, i.e., a transition section transitioning from a pleasant section to an unpleasant section, as a negative state.

[0073] Hereinafter, a section in which the emotional state is in a positive state will be referred to as a positive section, and a section in which the emotional state is in a negative state will be referred to as a negative section.

[0074] Next, the emotional state estimation unit 42 calculates the emotional reliability of the estimation result of the emotional state of each section (hereinafter referred to as emotional state reliability) based on at least one of the emotional valence of each section and the relative relationship of emotional valence between each section.

[0075] For example, the emotional state reliability of each plateau section is calculated based on the valence Vv of the immediately preceding plateau section and the absolute value ΔVv of the difference between the valence of the immediately preceding plateau section. Specifically, for example, the emotional state reliability of each plateau section is calculated using the following equations (3) and (4).

[0076] Rve=ΔVv×Vv...(3) Rvr=ΔVv×(1-Vv)...(4)

[0077] Rve is the emotional state reliability in the pleasant section, and Rvr is the emotional state reliability in the unpleasant section. Note that the emotional valence Vv is normalized within the range of 0 to 1.

[0078] Therefore, the emotional state reliability Rve increases as the valence Vv of the immediately preceding plateau section (unpleasant section) increases. This is because the higher the valence Vv of the immediately preceding unpleasant section, the more difficult it becomes for the valence Vv to increase, and despite this, it is estimated that the pleasant section reached by the increase in valence Vv is highly likely to indicate that the user is actually in a pleasant state. Furthermore, the emotional state reliability Rve increases as the absolute value ΔVv of the difference between the valence Vv and the immediately preceding plateau section (unpleasant section) increases. This is because the greater the increase in valence Vv from the immediately preceding unpleasant section, the higher the estimated probability that the user is actually in a pleasant state in the pleasant section reached thereafter.

[0079] On the other hand, the emotional state reliability Rvr increases as the valence Vv of the immediately preceding plateau section (comfortable section) decreases. This is because the smaller the valence Vv of the immediately preceding comfortable section, the more difficult it is for the valence Vv to decrease. Despite this, it is estimated that the unpleasant section reached by the decrease in valence Vv is likely to be an unpleasant state for the user. Furthermore, the emotional state reliability Rvr increases as the absolute value ΔVv of the difference between the valence Vv and the immediately preceding plateau section (comfortable section) increases. This is because the greater the decrease in valence Vv from the immediately preceding comfortable section, the higher the estimated probability that the user is actually in an unpleasant state in the unpleasant section reached thereafter.

[0080] Furthermore, the emotional state estimation unit 42 calculates the emotional state reliability of each transition section based on the change in valence of each transition section, for example, at least one of the direction, range, and rate of change of valence. For example, the emotional state estimation unit 42 calculates the reliability of each transition section based on the absolute value of the slope (rate of change) of the valence graph of each transition section. For example, the larger the absolute value of the slope of the valence graph of the transition section, the higher the emotional state reliability, and the smaller the absolute value of the slope of the valence graph of the transition section, the lower the emotional state reliability.

[0081] As shown in the examples of FIGS. 3 and 4, the emotional state evoked by the user in response to the same content may differ depending on factors such as the user's state, the surrounding environment, and previously played content.

[0082] For example, in the examples of Figures 3 and 4, song A is played three times. The first time song A is played, the arousal state is a calm state and the emotional state is a pleasant state. The second time song A is played, the arousal state is a calm state and the emotional state is an unpleasant state. The third time song A is played, the arousal state is a calm state and the emotional state is a positive state.

[0083] 3 and 4, song C is played twice. When song A is played the first time, the arousal state is an activated state and the emotional state is a pleasant state. When song C is played the second time, the arousal state is a calm state and the emotional state is a pleasant state.

[0084] The emotional state estimation unit 42 supplies the estimation result of the user's emotional state and information indicating the emotional reliability to the tag assignment unit 35 .

[0085] For example, the processes of steps S3 and S4 may be performed in parallel with the process of step S2, or may be performed after the process of step S2 is completed. In the latter case, for example, a biosignal of the user is recorded during the content presentation, and the processes of steps S3 and S4 are performed based on the recorded biosignal.

[0086] In step S5, the tagging unit 35 assigns a tag to the content.

[0087] For example, the tag of each piece of content includes an emotional state label indicating the user's emotion that is estimated to be caused by each piece of content, and a label confidence level indicating the confidence level of the emotional state label.

[0088] The tagging unit 35 compares the playback history of the content with the time-series changes in the estimated emotional state of the user, and assigns an emotional state label to each piece of content based on the emotional state of the user when each piece of content was presented. That is, the tagging unit 35 detects a section in the time-series changes in the user's emotional state that includes the playback period (presentation period) of each piece of content, and assigns an emotional state label to each piece of content based on the emotional state of the user estimated in the detected section.

[0089] For example, an emotional state label based on the arousal state (hereinafter referred to as an arousal state label) is assigned based on the arousal state of the user estimated during the playback period of the content. For example, the values ​​of the arousal state label include an active state, an activated state, a calmed state, and a calmed state.

[0090] For example, if the playback period of a content is included in an active section of the wakefulness data, the value of the wakefulness state label of the content is set to the active state. For example, if the playback period of a content is included in an active section of the wakefulness data, the value of the wakefulness state label of the content is set to the active state. For example, if the playback period of a content is included in a calming section of the wakefulness data, the value of the wakefulness state label of the content is set to the calming state. For example, if the playback period of a content is included in a calming section of the wakefulness data, the value of the wakefulness state label of the content is set to the calming state.

[0091] For example, an emotional state label based on the emotional state (hereinafter referred to as an emotional state label) is assigned based on the user's emotional state estimated during the playback period of the content. For example, the values ​​of the emotional state label include a pleasant state, a positive state, an unpleasant state, and a negative state.

[0092] For example, if the playback period of a content falls within a pleasant section of the valence data, the value of the emotional state label of the content is set to a pleasant state. For example, if the playback period of a content falls within a positive section of the valence data, the value of the emotional state label of the content is set to a positive state. For example, if the playback period of a content falls within an unpleasant section of the valence data, the value of the emotional state label of the content is set to an unpleasant state. For example, if the playback period of a content falls within a negative section of the valence data, the value of the emotional state label of the content is set to a negative state.

[0093] In addition, when the playback period of a content spans multiple sections of arousal data or valence data, an emotional state label is assigned to the content based on, for example, the user's emotional state estimated in one of the multiple sections that overlap with the playback period of the content according to a predetermined rule. For example, an emotional state label is assigned to the content based on the user's emotional state estimated in the section that includes the longest playback period of the content. For example, an emotional state label is assigned to the content based on the user's emotional state estimated in the section that includes the beginning of the content.

[0094] Furthermore, the tagging unit 35 calculates the label reliability for the emotional state label assigned to each piece of content based on the emotional reliability of the section that is the basis for assigning the emotional state label. For example, the tagging unit 35 may use the emotional reliability directly as the label reliability, or may calculate the label reliability based on the emotional reliability using a predetermined formula or the like.

[0095] For example, the tag assigning unit 35 calculates the label reliability of the emotional state label assigned to each piece of content (hereinafter referred to as the arousal state label reliability) based on the arousal state reliability of the section that is the basis for assigning the emotional state label. For example, the tag assigning unit 35 calculates the label reliability of the emotional state label assigned to each piece of content (hereinafter referred to as the emotional state label reliability) based on the emotional state reliability of the section that is the basis for assigning the emotional state label.

[0096] The tagging unit 35 assigns a tag to each piece of content. For example, the tagging unit 35 generates a tag for each piece of content and assigns the tag to the content data stored in the content database 21 as meta information.

[0097] The tag includes, for example, identification information for identifying the content (e.g., a content ID), identification information for identifying the user (e.g., a user ID), an alertness state label, an alertness state reliability, an emotional state label, an emotional state reliability, and the date and time the tag was assigned.

[0098] It should be noted that multiple tags can be assigned to each piece of content.

[0099] For example, when multiple users play the same content, multiple tags generated when each user plays the content are assigned to the content, respectively. That is, tags corresponding to each user are assigned to each piece of content.

[0100] For example, when one user plays the same content multiple times, multiple tags generated at each playback are assigned to the content.

[0101] For example, if a user plays the same content multiple times, the number of tags assigned to the content corresponding to the user may be limited to one. In this case, for example, the tags assigned to the content for the user may be updated based on the context (e.g., time, period, situation, etc.) when the content was played (presented). For example, the tag generated during the most recent playback may overwrite the previous tag. For example, a tag including an emotional state label (e.g., a label indicating the average emotional state of the user) based on the emotional state label included in the previous tag and the emotional state label included in the tag generated during the most recent playback may be newly assigned to the content. In this way, the tags assigned to each content are personalized and optimized for each user.

[0102] In step S6, the characteristic estimation unit 22 estimates the characteristics of the content based on the tag.

[0103] For example, the characteristic estimation unit 22 estimates the emotional state and level that each piece of content evokes in the user based on the tag attached to each piece of content. Specifically, for example, the characteristic estimation unit 22 estimates an emotional state score that indicates the emotional state and level that each piece of content evokes in the user based on the tag attached to each piece of content.

[0104] For example, the characteristic estimation unit 22 estimates the activity level and pleasantness level as the emotional state score of the content to be evaluated (hereinafter referred to as the content to be evaluated) based on a tag corresponding to the user to be evaluated (hereinafter referred to as the user to be evaluated) that is attached to the content to be evaluated (hereinafter referred to as the content to be evaluated).

[0105] Here, the tag corresponding to the user to be presumed is a tag that is assigned when content to be presumed is presented to the user to be presumed.

[0106] The activity level is set to a value between −100 and +100, for example.

[0107] For example, the higher the degree to which the estimation target content activates (excites) the user's emotions, the higher the activation level (the larger the value in the positive direction). For example, the higher the activation level of the estimation target content, the higher the probability that the estimation target content will put the estimation target user in an activated state or activated state, or the higher the level of the activation state.

[0108] Conversely, the greater the degree to which the inference target content calms (relaxes) the user's emotions, the lower the activity level (the larger the value in the negative direction). For example, the lower the activity level of the inference target content, the higher the probability that the inference target content will put the inference target user into a calm or relaxed state, or the higher the level of the relaxed state.

[0109] For example, the activity level increases as the number or ratio of arousal state labels having values ​​of the active state or the activated state among the arousal state labels assigned to the content to be estimated increases. For example, the activity level increases as the arousal state label reliability for the arousal state labels having values ​​of the active state or the activated state among the arousal state labels assigned to the content to be estimated increases.

[0110] For example, the activity level decreases as the number or ratio of arousal state labels having a value of a sedated state or a calmed state among the arousal state labels assigned to the estimation target content increases. For example, the activity level decreases as the arousal state label reliability for arousal state labels having a value of a sedated state or a calmed state among the arousal state labels assigned to the estimation target content increases.

[0111] The emotional state score obtained by inverting the sign of the activity level is the calmness level.

[0112] The comfort level is set to a value between -100 and +100, for example.

[0113] For example, the more the inferred content positively changes the user's emotions, the higher the pleasantness rating (the larger the value in the positive direction). For example, the higher the pleasantness rating of the inferred content, the higher the probability that the inferred content will put the inferred user in a pleasant state or a positive state, or the higher the level of the pleasant state.

[0114] Conversely, the more the inference target content negatively affects the user's emotions, the lower the pleasantness level (the larger the value in the negative direction). For example, the lower the pleasantness level of the inference target content, the higher the probability that the inference target content will put the inference target user in an unpleasant or negative state, or the higher the level of the unpleasant state.

[0115] For example, the pleasantness level increases as the number or ratio of emotional state labels having values ​​of a pleasant state or a positive state among the emotional state labels assigned to the content to be estimated increases. For example, the pleasantness level increases as the emotional state label reliability for the emotional state labels having values ​​of a pleasant state or a positive state among the emotional state labels assigned to the content to be estimated increases.

[0116] For example, the pleasantness level decreases as the number or ratio of emotional state labels having values ​​of unpleasant or negative states among the emotional state labels assigned to the content to be inferred increases. For example, the pleasantness level decreases as the emotional state label reliability for the emotional state labels having values ​​of unpleasant or negative states among the emotional state labels assigned to the content to be inferred increases.

[0117] The emotional state score obtained by reversing the sign of the pleasantness rating is the unpleasantness rating.

[0118] The characteristic estimation unit 22 assigns activity levels and pleasantness levels as meta information to the estimation target contents stored in the content database 21 .

[0119] The number of users to be estimated may be one or more.

[0120] For example, if the number of users to be estimated is one, the activity level and pleasantness level of the (personalized) content to be estimated for each individual are estimated.

[0121] For example, if the number of users to be estimated is multiple, the average activity level and pleasantness level of the content to be estimated are estimated for a group consisting of the multiple users to be estimated. For example, if the users to be estimated are all users, the average activity level and pleasantness level of the content to be estimated are estimated for all users.

[0122] In step S7, the information processing system 1 presents the content.

[0123] Specifically, the presentation control unit 23 of the cloud system 11 selects content to be presented to a user (hereinafter referred to as a presentation target user).

[0124] At this time, the presentation control unit 23 may select content based on conditions (hereinafter referred to as specified conditions) specified by the target user via the UI unit 31, or may select content regardless of the specified conditions.

[0125] The specified condition is not particularly limited. For example, the title of a specific piece of content may be specified as the specified condition, or a condition based on the characteristics of the content may be specified as the specified condition. In the latter case, for example, the presentation target user inputs a specified condition such as exciting content, relaxing content, or comfortable content. The UI unit 31 transmits information indicating the input specified condition to the presentation control unit 23 of the cloud system 11.

[0126] For example, when exciting content is set as a specified condition, the presentation control unit 23 extracts content with a high activity level for the presentation target user. For example, the presentation control unit 23 extracts a predetermined number of pieces of content in descending order of activity level for the presentation target user. Alternatively, for example, the presentation control unit 23 extracts content with an activity level for the presentation target user equal to or greater than a predetermined threshold.

[0127] For example, when relaxing content is set as a specified condition, the presentation control unit 23 extracts content with a low activity level (high calmness level) relative to the presentation target user. For example, the presentation control unit 23 extracts a predetermined number of contents in descending order of activity level (high calmness level) relative to the presentation target user. Alternatively, for example, the presentation control unit 23 extracts content with an activity level relative to the presentation target user that is equal to or lower than a predetermined threshold (a calmness level equal to or higher than a predetermined threshold).

[0128] For example, when "pleasant content" is set as a specified condition, the presentation control unit 23 extracts content having a high pleasantness rating for the presentation target user. For example, the presentation control unit 23 extracts a predetermined number of pieces of content in descending order of pleasantness rating for the presentation target user. Alternatively, for example, the presentation control unit 23 extracts content having a pleasantness rating for the presentation target user equal to or greater than a predetermined threshold.

[0129] For example, the presentation control unit 23 may extract content based on the emotional state score for a user group including multiple users (for example, all users), rather than the emotional state score for the presentation target user.

[0130] The presentation control unit 23 generates information related to the content to be presented (hereinafter referred to as presented content information). The presented content information includes, for example, meta information of the content to be presented. The meta information of the content to be presented includes the above-mentioned tags and characteristics. The presentation control unit 23 controls the presentation of the content on the information processing terminal 12 by transmitting the information to the information processing terminal 12.

[0131] In response, the UI unit 31 of the information processing terminal 12 receives the presented content information. The UI unit 31 presents content to the presentation target user based on the presented content information. For example, the selected content may be presented individually, or a list of the selected content may be presented.

[0132] 5 to 7 show examples of content presentation.

[0133] 5 shows an example of music presentation. Specifically, the title of the music, the cover photo of the music, and the relaxation level of the music are displayed. The relaxation level of the music may be, for example, the calmness level of the music.

[0134] Figure 6 shows an example of a content distribution based on valence and arousal. The horizontal axis of the chart in Figure 6 represents valence, and the horizontal axis represents arousal. Jacket photos representing each song are arranged within the chart based on the valence and arousal of each song.

[0135] 7 shows an example of a playlist containing the top 10 songs by activity level. The cover art, title, artist name, and activity level of each song in the playlist are displayed. The activity levels are displayed in a pie chart.

[0136] In this way, it is possible to assign appropriate tags to content based on the user's emotional response to the content, and it is also possible to present appropriate content to the user based on the assigned tags.

[0137] As a result, it becomes possible to provide a service that is highly satisfying and reflects the user's psychological sensations, such as their emotional state.

[0138] <<2. Modifications>> Modifications of the above-described embodiments of the present technology will now be described.

[0139] <Modifications of the Configuration of Information Processing System 1> The configuration of the information processing system 1 described above is one example, and can be modified as appropriate.

[0140] For example, the cloud system 11 may be configured to execute part or all of the processing of the information processing terminal 12. For example, the cloud system 11 may be configured to execute part or all of the processing of the playback history generation unit 33, the emotion estimation unit 34, and the tag assignment unit 35.

[0141] For example, the information processing terminal 12 may be configured to execute some or all of the processing of the cloud system 11. For example, if the information processing terminal 12 is configured to execute all of the processing of the cloud system 11, a private system closed to an individual user is configured.

[0142] For example, the information processing terminal 12 may be configured to execute part of the processing of the wearable device 13. For example, the information processing terminal 12 may be configured to execute all of the processing of the biosensor 52 and the signal processing unit 53. In other words, the information processing terminal 12 may be configured to detect a biosignal of the user.

[0143] For example, the wearable device 13 may be configured to execute part of the processing of the information processing terminal 12. For example, the wearable device 13 may be configured to execute part or all of the processing of the playback unit 32 and the playback history generation unit 33. For example, the wearable device 13 may be configured to execute part or all of the processing of the emotion estimation unit 34 and the tag assignment unit 35.

[0144] For example, the wearable device 13 and the information processing terminal 12 may be integrated. In this case, for example, the integrated device may execute part or all of the processing of the cloud system 11.

[0145] For example, the output unit 51 of the wearable device 13 may be configured as different devices from the biosensor 52 and the signal processing unit 53.

[0146] <Modifications Related to Emotion Value> For example, only one of the arousal level or the emotional valence may be used.

[0147] For example, affective values ​​other than arousal and valence may be used.

[0148] For example, the user's emotional state may be estimated using Russell's circumplex model or the like by combining arousal and emotional valence.

[0149] <Modifications Related to Biological Signals> For example, a biological signal other than a pulse wave signal may be used to estimate an emotion value.

[0150] For example, a biological signal indicating a biological index of the autonomic nervous system or central nervous system, such as an electroencephalogram or a heart rate, may be used.

[0151] For example, multiple biological signals may be used.

[0152] <Modifications Regarding Content Presentation Method> For example, tags (e.g., emotional state labels, etc.) may be presented together with the content.

[0153] For example, content to be presented may be selected based on tags (e.g., emotional state labels and label confidence levels), such as content tagged with a specified condition.

[0154] In this case, for example, the expression of the presented emotional state label may be changed as appropriate. For example, expressions such as excited or focused may be used instead of active state. For example, expressions such as relaxed may be used instead of calm state.

[0155] For example, tags and characteristics may be combined to select content for presentation.

[0156] For example, content may be presented based on user preferences, etc., using at least one of the tags or characteristics and other meta information (for example, the number of times the content has been played, the playback time, etc.).

[0157] For example, tags (e.g., emotional state labels) attached to content may be presented to the user, and the user may refer to the tags to create a playlist or the like of the content themselves.

[0158] <Other Modifications> The present technology can be applied to, for example, systems, devices, services, etc. that present content to users. For example, the present technology can be applied to systems that recommend content, content streaming services, etc.

[0159] <<3. Others>> <Example of Computer Configuration> The above-described series of processes can be executed by hardware or software. When the series of processes is executed by software, the programs that make up the software are installed on a computer. Here, the term "computer" includes computers built into dedicated hardware, and general-purpose personal computers, for example, that can execute various functions by installing various programs.

[0160] FIG. 8 is a block diagram showing an example of the hardware configuration of a computer that executes the above-described series of processes by a program.

[0161] In the computer 1000 , a CPU (Central Processing Unit) 1001 , a ROM (Read Only Memory) 1002 , and a RAM (Random Access Memory) 1003 are interconnected by a bus 1004 .

[0162] An input / output interface 1005 is further connected to the bus 1004. An input unit 1006, an output unit 1007, a storage unit 1008, a communication unit 1009, and a drive 1010 are connected to the input / output interface 1005.

[0163] The input unit 1006 includes input switches, buttons, a microphone, an image sensor, etc. The output unit 1007 includes a display, a speaker, etc. The storage unit 1008 includes a hard disk, a non-volatile memory, etc. The communication unit 1009 includes a network interface, etc. The drive 1010 drives removable media 1011 such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory.

[0164] In the computer 1000 configured as described above, the CPU 1001 performs the above-described series of processes by, for example, loading a program recorded in the memory unit 1008 into the RAM 1003 via the input / output interface 1005 and the bus 1004 and executing it.

[0165] The program executed by the computer 1000 (CPU 1001) can be provided by being recorded on a removable medium 1011 such as a package medium, for example. The program can also be provided via a wired or wireless transmission medium such as a local area network, the Internet, or digital satellite broadcasting.

[0166] In the computer 1000, the program can be installed in the storage unit 1008 via the input / output interface 1005 by inserting the removable medium 1011 into the drive 1010. The program can also be received by the communication unit 1009 via a wired or wireless transmission medium and installed in the storage unit 1008. Alternatively, the program can be installed in the ROM 1002 or the storage unit 1008 in advance.

[0167] The program executed by the computer may be a program that processes in chronological order according to the order described in this specification, or may be a program that processes in parallel or at the required timing, such as when called.

[0168] In this specification, a system refers to a collection of multiple components (devices, modules (components), etc.), regardless of whether all of the components are housed in the same housing. Therefore, multiple devices housed in separate housings and connected via a network, and a single device housed in a single housing with multiple modules, are both systems.

[0169] Furthermore, the embodiments of the present technology are not limited to the above-described embodiments, and various modifications are possible within the scope of the gist of the present technology.

[0170] For example, the present technology can be configured as a cloud computing system in which a single function is shared and processed collaboratively by a plurality of devices via a network.

[0171] Furthermore, each step described in the above flowchart can be executed by one device, or can be shared and executed by a plurality of devices.

[0172] Furthermore, when one step includes multiple processes, the multiple processes included in that one step can be executed by one device or can be shared and executed by multiple devices.

[0173] <Examples of Combinations of Configurations> The present technology can also have the following configurations.

[0174] (1) An information processing system comprising: an emotion estimation unit that estimates an emotional state of a user based on emotion value data that is time-series data of emotion values ​​estimated based on a biosignal of the user; and a tag assignment unit that assigns to the content a tag including an emotional state label that is a label based on the emotional state of the user estimated based on the emotion value data when the content is presented to the user. (2) The information processing system described in (1), wherein the emotion estimation unit divides the emotion value data into a plurality of intervals and estimates the emotional state of the user for each interval based on at least one of the emotion value of each interval and the relative relationship of the emotion values ​​between each interval. (3) The information processing system according to (2), wherein the emotion estimation unit extracts a steady section from the emotion value data, where the emotion value is stable, and estimates the user's emotional state during a target steady section based on a relative relationship between the emotion value of the target steady section, where the steady section is the section targeted for estimation of the user's emotional state, and the emotion value of a immediately preceding steady section, where the steady section immediately preceding the target steady section. (4) The information processing system according to (3), wherein the emotion estimation unit calculates emotion reliability, which is a reliability of the estimation result of the user's emotional state during the target steady section, based on the emotion value of the immediately preceding steady section and the relative relationship between the emotion value of the immediately preceding steady section and the emotion value of the target steady section. (5) The information processing system according to (3) or (4), wherein the emotion estimation unit estimates the user's emotional state during a transition section based on a change in the emotion value during a transition section, where the transition section is a section between adjacent steady sections. (6) The information processing system according to any one of (2) to (6), wherein the emotion estimation unit calculates an emotion reliability, which is a reliability of an estimation result of the emotional state of the user during the transition period, based on at least one of a direction of fluctuation, a fluctuation range, and a fluctuation rate of the emotion value during the transition period. (7) The information processing system according to any one of (2) to (6), wherein the tag assignment unit assigns the emotional state label to the content based on an estimation result of the emotional state of the user during a period overlapping with a presentation period of the content.(8) The information processing system according to (7), wherein the emotion estimation unit calculates emotion reliability, which is the reliability of the estimation result of the user's emotional state for each interval, based on at least one of the emotion value for each interval and the relative relationship of the emotion values ​​between each interval, and the tag assignment unit calculates label reliability, which is the reliability of the emotional state label, based on the emotion reliability for the interval overlapping with the presentation period of the content. (9) The information processing system according to (8), further comprising: a characteristic estimation unit that estimates characteristics of the content based on the emotional state label and the label reliability assigned to the content. (10) The information processing system according to (9), further comprising: a presentation control unit that controls presentation of the content to a user based on the estimation result of the content characteristics. (11) The information processing system according to (10), wherein the presentation control unit further controls presentation of the content characteristics. (12) The information processing system according to any one of (9) to (11), wherein the characteristic estimation unit estimates the state and level of the emotion caused by the content as the characteristic of the content based on the emotional state label and the label reliability assigned to the content. (13) The information processing system according to any one of (9) to (12), wherein the characteristic estimation unit estimates the characteristic of the content based on a plurality of the emotional state labels and the label reliability assigned to the content when presented to a plurality of users. (14) The information processing system according to any one of (1) to (13), wherein the tag assignment unit assigns the tag corresponding to each user to the content. (15) The information processing system according to (14), wherein, when the same content is presented to the same user, the tag assignment unit updates the tag assigned to the content corresponding to the user based on the context when the content was presented. (16) The information processing system according to any one of (1) to (15), further comprising a presentation control unit that controls presentation of the content to a user based on the emotional state label assigned to the content.(17) The information processing system according to (16), wherein the presentation control unit controls presentation of the content to users based on a plurality of emotional state labels respectively assigned to the content when presented to each of a plurality of users. (18) The information processing system according to any of (1) to (17), wherein the emotional value includes at least one of arousal and valence. (19) An information processing method, wherein an information processing system estimates an emotional state of a user based on emotion value data that is time-series data of emotion values ​​estimated based on a biosignal of the user, and assigns to the content a tag including an emotional state label based on the emotional state of the user estimated based on the emotion value data when the content is presented to the user. (20) An information processing device comprising: an emotion estimation unit that estimates an emotional state of a user based on emotion value data that is time-series data of emotion values ​​estimated based on a biosignal of the user; and a tag assignment unit that assigns to the content a tag including an emotional state label based on the emotional state of the user estimated based on the emotion value data when the content is presented to the user.

[0175] The effects described in this specification are merely examples and are not limiting, and other effects may also be present.

[0176] REFERENCE SIGNS LIST 1 Information processing system, 11 Cloud system, 12 Information processing terminal, 13 Wearable device, 21 Content database, 22 Characteristics estimation unit, 23 Presentation control unit, 31 UI unit, 32 Playback unit, 33 Playback history generation unit, 34 Emotion estimation unit, 35 Tag assignment unit, 41 Emotion value estimation unit, 42 Emotion state estimation unit, 51 Output unit, 52 Biometric sensor, 53 Signal processing unit

Claims

1. An information processing system comprising: an emotion estimation unit that estimates a user's emotional state based on emotion value data, which is time-series data of emotion values ​​estimated based on the user's biometric signals; and a tag assignment unit that assigns to the content a tag including an emotional state label, which is a label based on the user's emotional state estimated based on the emotion value data when the content is presented to the user.

2. The information processing system according to claim 1, wherein the emotion estimation unit divides the emotion value data into a plurality of intervals and estimates the user's emotional state for each interval based on at least one of the emotion value for each interval and the relative relationship of the emotion values ​​between the intervals.

3. The information processing system according to claim 2, wherein the emotion estimation unit extracts a steady section from the emotion value data, where the emotion value is stable, and estimates the user's emotional state in the target steady section based on the relative relationship between the emotion value of the target steady section, which is the steady section targeted for estimation of the user's emotional state, and the emotion value of the immediately preceding steady section, which is the steady section immediately preceding the target steady section.

4. The information processing system according to claim 3, wherein the emotion estimation unit calculates emotion reliability, which is the reliability of the estimation result of the user's emotional state in the target steady section, based on the emotion value of the immediately preceding steady section and the relative relationship between the emotion value of the immediately preceding steady section and the emotion value of the target steady section.

5. The information processing system according to claim 3, wherein the emotion estimation unit estimates the user's emotional state in a transition section, which is a section between adjacent steady sections, based on a change in the emotion value in the transition section.

6. The information processing system according to claim 5, wherein the emotion estimation unit calculates emotion reliability, which is the reliability of the estimation result of the user's emotional state during the transition section, based on at least one of the direction of change, the range of change, and the rate of change of the emotion value during the transition section.

7. The information processing system according to claim 2, wherein the tagging unit assigns the tag including the emotional state label to the content based on the estimation result of the user's emotional state during the period overlapping with the presentation period of the content.

8. The information processing system according to claim 7, wherein the emotion estimation unit calculates emotion reliability, which is the reliability of the estimation result of the user's emotional state for each interval, based on at least one of the emotion value for each interval and the relative relationship of the emotion values ​​between each interval; and the tag assignment unit calculates label reliability, which is the reliability of the emotional state label, based on the emotion reliability for the interval overlapping with the presentation period of the content.

9. The information processing system according to claim 8, further comprising a characteristic estimation unit that estimates the characteristics of the content based on the emotional state label and the label reliability assigned to the content.

10. The information processing system according to claim 9, further comprising a presentation control unit that controls the presentation of said content to a user based on the estimation result of the characteristics of said content.

11. The information processing system according to claim 10, wherein the presentation control unit further controls the presentation of the characteristics of the content.

12. The information processing system according to claim 9, wherein the characteristic estimation unit estimates the emotional state and level caused by the content as the characteristic of the content based on the emotional state label and the label reliability assigned to the content.

13. The information processing system according to claim 9, wherein the characteristic estimation unit estimates the characteristics of the content based on the plurality of emotional state labels and the label reliability assigned to the content when presented to each of the plurality of users.

14. The information processing system according to claim 1, wherein the tagging unit assigns the tag corresponding to each user to each piece of content.

15. The information processing system of claim 14, wherein when the same content is presented to the same user, the tag assignment unit updates the tag assigned to the content corresponding to the user based on the context when the content was presented.

16. The information processing system according to claim 1, further comprising a presentation control unit that controls presentation of the content to the user based on the emotional state label assigned to the content.

17. The information processing system according to claim 16, wherein the presentation control unit controls the presentation of the content to the users based on the plurality of emotional state labels assigned to the content when presented to each of the plurality of users.

18. The information processing system of claim 1, wherein the emotional value includes at least one of arousal and emotional valence.

19. An information processing method in which an information processing system estimates a user's emotional state based on emotional value data, which is time-series data of emotional values ​​estimated based on the user's biometric signals, and assigns to the content a tag including an emotional state label based on the user's emotional state estimated based on the emotional value data when the content is presented to the user.

20. An information processing device comprising: an emotion estimation unit that estimates a user's emotional state based on emotion value data, which is time-series data of emotion values ​​estimated based on the user's biometric signals; and a tag assignment unit that assigns to the content a tag including an emotional state label based on the user's emotional state estimated based on the emotion value data when the content is presented to the user.

Images