System and method for embedded cognitive state metric system
Inventors
Le, Tan • Mackellar, Geoffrey Ross
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-11847260-B2
Publication Date
2023-12-19
Expiration Date
Abstract
An embodiment of a method for enabling content personalization for a user based on a cognitive state of the user includes providing an interface configured to enable a third party to request cognitive state data of the user as the user interacts with a content-providing source; establishing bioelectrical contact between a biosignal detector and the user; automatically collecting a dataset from the user; generating a cognitive state metric; receiving a request from the third party for cognitive state data; transmitting the cognitive state data to the third party device; and automatically collecting a dataset from the user as the user engaged tailored content.
Core Innovation
The disclosed invention relates to an embedded cognitive-state metric system in which bioelectrical contact is established between a head-mounted bioelectrical sensor and a user at a plurality of head regions. While the user engages stimuli from a content provider and a third party, the system automatically collects a bioelectrical dataset measured from the plurality of head regions and determines cognitive state metric data using a first model.
The system generates a set of user groups based on the stimuli, the cognitive state metric, and the second cognitive state metric. Using a second model that comprises a machine learning model, the system selects a user group associated with an individual who is not within the set of users, determines personalized content for the individual based on the selected user group, and presents the personalized content to the individual.
In training, the system measures bioelectrical signals for training users and determines cognitive state data using a first model. The system determines user groups by classifying a cognitive state of each training user so the classifications are used to group the training users into user groups, and trains a second machine learning model to output a user group associated with a training user.
Claims Coverage
The partial content provides two independent methods. Across the independent claims, the inventive features are focused on head-mounted multi-region bioelectrical contact, model-based cognitive state determination, user-group generation or classification, machine learning-based user-group selection, and presenting personalized content or tailored stimuli to a user interacting with third-party stimuli.
Head-mounted multi-region bioelectrical contact and dataset collection
Establishing bioelectrical contact between a head-mounted bioelectrical sensor and the user at a plurality of head regions; automatically collecting a bioelectrical dataset from the user as the user engages a first stimulus and a second stimulus, wherein the bioelectrical dataset comprises bioelectrical data measured from the plurality of head regions.
Model-based cognitive state metric determination
Determining a cognitive state metric based on the bioelectrical dataset using a first model; determining a second cognitive state metric based on the second bioelectrical dataset using the first model.
User-group generation and machine learning-based selection for an individual not in the monitored users
Generating a set of user groups based on the first stimulus, the cognitive state metric, the second stimulus, and the second cognitive state metric; using a second model that comprises a machine learning model to select a user group from the set of user groups based on a cognitive state metric of an individual not within the set of users.
Personalized content presenting based on selected user group
Determining personalized content for the individual based on the user group; presenting the personalized content to the individual.
Training with classified cognitive states and a trained machine learning user-group selector
Determining user groups from the set of training users by classifying a cognitive state of each training user so the classifications are used to group the training users into user groups; training a second model to output a user group associated with a training user based on cognitive state data, wherein the second model comprises a machine learning model.
Cognitive state data determined from bioelectrical signals associated with brain activity from brain lobes
Establishing bioelectrical contact between a biosignal detector and a user, wherein the biosignal detector is coupled to a plurality of head regions corresponding to a plurality of brain lobes; automatically collecting a bioelectrical signal dataset from the user as the user interacts with stimuli provided by a third party; using the first model to determine cognitive state data indicative of a cognitive state of the user from the bioelectrical signal dataset, wherein the cognitive state data is associated with brain activity from the plurality of brain lobes and is determined using a machine learning method trained using a training dataset generated by measuring bioelectrical signals and cognitive state data associated with a training group of users interacting with training stimuli from the third party.
Tailored stimuli presenting based on selected user group and association
Using the second model, selecting a user group based on the cognitive state data associated with the user; presenting tailored stimuli generated based on an analysis of the selected user group, wherein the tailored stimuli is further determined based on the stimuli and the association between the cognitive state data for the user and the stimuli.
Across both independent claims, the invention centers on multi-region head-mounted bioelectrical contact and automatic bioelectrical dataset collection during third-party stimuli interaction, model-based cognitive state metric or data determination, forming or classifying user groups based on cognitive state metrics or data, selecting a user group via a machine learning model, and presenting personalized content or tailored stimuli based on the selected user group and its association with the user’s cognitive state data.
Stated Advantages
Documented Applications
No documented applications found
Interested in licensing this patent?