2363 results about "Learning machine" patented technology

A method for providing automatic, personalized information services to a computer user includes the following steps: transparently monitoring user interactions with data during normal use of the computer; updating user-specific data files including a set of user-related documents; estimating parameters of a learning machine that define a User Model specific to the user, using the user-specific data files; analyzing a document to identify its properties; estimating the probability that the user is interested in the document by applying the document properties to the parameters of the User Model; and providing personalized services based on the estimated probability. Personalized services include personalized searches that return only documents of interest to the user, personalized crawling for maintaining an index of documents of interest to the user; personalized navigation that recommends interesting documents that are hyperlinked to documents currently being viewed; and personalized news, in which a third party server customized its interaction with the user. The User Model includes continually-updated measures of user interest in words or phrases, web sites, topics, products, and product features. The measures are updated based on both positive examples, such as documents the user bookmarks, and negative examples, such as search results that the user does not follow. Users are clustered into groups of similar users by calculating the distance between User Models.

The spectral kernel machine combines kernel functions and spectral graph theory for solving problems of machine learning. The data points in the dataset are placed in the form of a matrix known as a kernel matrix, or Gram matrix, containing all pairwise kernels between the data points. The dataset is regarded as nodes of a fully connected graph. A weight equal to the kernel between the two nodes is assigned to each edge of the graph. The adjacency matrix of the graph is equivalent to the kernel matrix, also known as the Gram matrix. The eigenvectors and their corresponding eigenvalues provide information about the properties of the graph, and thus, the dataset. The second eigenvector can be thresholded to approximate the class assignment of graph nodes. Eigenvectors of the kernel matrix may be used to assign unlabeled data to clusters, merge information from labeled and unlabeled data by transduction, provide model selection information for other kernels, detect novelties or anomalies and / or clean data, and perform supervised learning tasks such as classification.

The present invention is a method and system for measuring human emotional response to visual stimulus, based on the person's facial expressions. Given a detected and tracked human face, it is accurately localized so that the facial features are correctly identified and localized. Face and facial features are localized using the geometrically specialized learning machines. Then the emotion-sensitive features, such as the shapes of the facial features or facial wrinkles, are extracted. The facial muscle actions are estimated using a learning machine trained on the emotion-sensitive features. The instantaneous facial muscle actions are projected to a point in affect space, using the relation between the facial muscle actions and the affective state (arousal, valence, and stance). The series of estimated emotional changes renders a trajectory in affect space, which is further analyzed in relation to the temporal changes in visual stimulus, to determine the response.

Digitized image data are input into a processor where a detection component identifies the areas (objects) of particular interest in the image and, by segmentation, separates those objects from the background. A feature extraction component formulates numerical values relevant to the classification task from the segmented objects. Results of the preceding analysis steps are input into a trained learning machine classifier which produces an output which may consist of an index discriminating between two possible diagnoses, or some other output in the desired output format. In one embodiment, digitized image data are input into a plurality of subsystems, each subsystem having one or more support vector machines. Pre-processing may include the use of known transformations which facilitate extraction of the useful data. Each subsystem analyzes the data relevant to a different feature or characteristic found within the image. Once each subsystem completes its analysis and classification, the output for all subsystems is input into an overall support vector machine analyzer which combines the data to make a diagnosis, decision or other action which utilizes the knowledge obtained from the image.