Systems and methods for machine learning enhanced by human measurements
AssigneePresident and Fellows of Harvard College
InventorsDavid Cox, Walter Scheirer, Samuel Anthony, Ken Nakayama
Life of this patent
Estimating the term…
Abstract
In various embodiments, training objects are classified by human annotators, psychometric data characterizing the annotation of the training objects is acquired, a human-weighted loss function based at least in part on the classification data and the psychometric data is computationally derived, and one or more features of a query object are computationally classified based at least its part on the human-weighted loss function.
Claims30
A computer-implemented method for data classification, the method comprising:
- providing, over a computer network, data corresponding to a plurality of training objects to a plurality of human annotators for annotation of the training objects by the human annotators;
- receiving, over the computer network, classification data associated with the training objects and based at least in part on the annotation of the training objects by the human annotators;
- computationally deriving a human-weighted loss function based at least in part on the classification data, the human-weighted loss function comprising penalties for misclassification, magnitudes of the penalties increasing with increasing deviation from the classification data;
- receiving, by a classification device, data corresponding to a query object;
- and
- thereafter, by a computer processor, computationally classifying the query object based at least in part on the human-weighted loss function.
A computer-implemented method for data classification, the method comprising:
- providing, over a computer network, data corresponding to a plurality of training objects to a plurality of human annotators for annotation of the training objects by the human annotators;
- receiving, over the computer network, classification data associated with the training objects and based at least in part on the annotation of the training objects by the human annotators;
- computationally training a supervised learning model based at least in part on the classification data;
- receiving, by a classification device, data corresponding to a query object;
- and
- thereafter, by a computer processor, computationally classifying the query object based at least in part on the supervised learning model.
Dependencies are inferred from the claim text.