Information processing method, information processing device, and program

Abstract

To extract, with high accuracy, content that elicits a prescribed emotion in a user.SOLUTION: An information processing method causes a processor included in an information processing device to execute the steps of: extracting prescribed content that elicits a prescribed emotion in a prescribed user from a content list on the basis of an elicitation level predicted by a content model; sequentially acquiring brainwave signals of the prescribed user stimulated by the prescribed content from a brainwave measurement device worn by the prescribed user; inputting the sequentially acquired brainwave signals to a brainwave model that predicts the elicitation level of the prescribed emotion in the prescribed user by the content during stimulation by inputting the brainwave signal of the prescribed user stimulated by the content to predict the elicitation level of the prescribed emotion caused by the prescribed content; and updating the content model that predicts the elicitation level by inputting feature amounts of the content, using the correspondence between the feature amounts of the prescribed content and the elicitation level predicted by the brainwave model as re-learning data.SELECTED DRAWING: Figure 9

Claims

1. A processor included in the information processing device The electroencephalogram measuring device attached to the predetermined user detects the electroencephalogram stimulated by the predetermined content. Sequentially acquiring electroencephalogram signals of a predetermined user; The brainwave signal of the predetermined user stimulated by the content is input, and the content being stimulated is The electroencephalogram model predicts the degree to which the content induces a predetermined emotion in the predetermined user. The electroencephalogram signals obtained in sequence are input, and the predetermined emotion is evoked by the predetermined content. To predict the degree of A correspondence relationship between the feature amount of the predetermined content and the arousal degree predicted by the electroencephalogram model. The content is re-learned by inputting the feature amount of the content and predicting the degree of attraction. Updating the model and From the content list, a content is selected based on the degree of attraction predicted by the content model. extracting predetermined content that evokes the predetermined emotion of the predetermined user; An information processing method for performing the above.

2. The processor, The feature amount of each of the plurality of candidate contents in the content list is calculated based on the feature amount of the candidate contents. The plurality of candidate contents are selected based on each of the degrees of attraction predicted by inputting the candidate contents into a model. and calculating an elicitation score for each of the samples, The extracting step includes extracting the plurality of candidate contents based on the calculated arousal scores. extracting predetermined content that evokes the predetermined emotion of the predetermined user from the content; Including, The step of obtaining the predetermined content in the order includes: sequentially acquiring electroencephalographic signals of the The information processing method according to claim 1 .

3. The calculating step includes: calculating the arousal score based on the updated content model. Including recalculating The extracting step includes extracting the plurality of candidate contexts based on the recalculated elicitation scores. re-extracting predetermined content from the content; The processor performs the recalculating, the reextracting, and the updating. Repeating and further executing The information processing method according to claim 2 .

4. The extracting step includes extracting the predetermined content having the arousal score equal to or greater than a predetermined value, Extracting at least one of the predetermined contents having the attraction score equal to or less than a predetermined value. The information processing method according to claim 2 or 3, further comprising:

5. The calculation is performed based on the training data included in the training data for the content model. The content that evokes the predetermined emotion of the predetermined user and the candidate content in the content list 3. The method of claim 2, further comprising: calculating the arousal score based on a similarity between the content and the content of the video. The information processing method described in

6. The content model is a content model that is used by the predetermined user as a content that evokes the predetermined emotion. A feature amount of at least one selected content selected by the a correspondence relationship between the evaluation of the predetermined emotion by the predetermined user stimulated by the content and the predetermined emotion; At least one non-selected content that is not selected by the predetermined user, and the predetermined emotion by the predetermined user stimulated by at least one non-selected content; The method according to any one of claims 1 to 3, which is generated based on a correspondence relationship with an evaluation regarding Information processing methods.

7. The electroencephalogram model is configured to detect the electroencephalogram of the predetermined user stimulated by the at least one selected content. a user's electroencephalogram signal and a predetermined emotion corresponding to the at least one selected content. a correspondence between the label and the at least one non-selected content; The electroencephalogram signal of the predetermined user and the predetermined content corresponding to the at least one non-selected content. The information processing method according to claim 6, wherein the information processing method is generated based on a correspondence relationship with a label related to a certain emotion. method.

8. The processor updates the predetermined user's front page based on the updated content model. Identifying a feature amount of the content that evokes the predetermined emotion; generating predetermined content that stimulates the predetermined user based on the specified characteristic amount; The information processing method according to claim 1 , further comprising:

9. The processor includes the generated content in the content list. The information processing method according to claim 8, further comprising:

10. An information processing device including a processor, The processor, The electroencephalogram measuring device attached to the predetermined user detects the electroencephalogram stimulated by the predetermined content. Sequentially acquiring electroencephalogram signals of a predetermined user; The brainwave signal of the predetermined user stimulated by the content is input, and the content being stimulated is The electroencephalogram model predicts the degree to which the content induces a predetermined emotion in the predetermined user. The electroencephalogram signals obtained in sequence are input, and the predetermined emotion is evoked by the predetermined content. To predict the degree of A correspondence relationship between the feature amount of the predetermined content and the arousal degree predicted by the electroencephalogram model. The content is re-learned by inputting the feature amount of the content and predicting the degree of attraction. Updating the model and From the content list, a content is selected based on the degree of attraction predicted by the content model. extracting predetermined content that evokes the predetermined emotion of the predetermined user; An information processing device that executes the above.

11. A processor included in the information processing device The electroencephalogram measuring device attached to the predetermined user detects the electroencephalogram stimulated by the predetermined content. Sequentially acquiring electroencephalogram signals of a predetermined user; The brainwave signal of the predetermined user stimulated by the content is input, and the content being stimulated is The electroencephalogram model predicts the degree to which the content induces a predetermined emotion in the predetermined user. The electroencephalogram signals obtained in sequence are input, and the predetermined emotion is evoked by the predetermined content. To predict the degree of A correspondence relationship between the feature amount of the predetermined content and the arousal degree predicted by the electroencephalogram model. The content is re-learned by inputting the feature amount of the content and predicting the degree of attraction. Updating the model and From the content list, a content is selected based on the degree of attraction predicted by the content model. extracting predetermined content that evokes the predetermined emotion of the predetermined user; A program that executes the following.

Images