A method, data processing apparatus, and computer code for identifying road intersections includes providing location data obtained from at least one vehicle's trajectory, wherein the location data may include geographical data within a geographical perimeter. The method includes determining node vectors by applying a geographical descriptor model on a target location included in the geographical perimeter. The geographical descriptor model includes a plurality of multiscale node descriptors including a target multiscale descriptor and neighboring multiscale descriptors. Each of the plurality of multiscale node descriptors includes at least two shape descriptors of different geographical resolution. Each of the neighboring locations is at a respective geographical distance from the target location. The node vectors may be respectively determined for each of the plurality of multiscale node descriptors. The method includes inputting the node vectors into a trained multiscale classifier including a graph convolutional network to provide a probability of the target location being a road intersection.

Described herein are systems and techniques for privacy-preserving unsupervised learning. The disclosed system and methods can enable separate computers, operated by separate entities, to perform unsupervised learning jointly based on a pool of their respective data, while preserving privacy. The system improves efficiency and scalability, while preserving privacy and avoids leaking a cluster identification. The system can jointly compute a secure distance via privacy-preserving multiplication of respective data values x and y from the computers based on a 1-out-of-N oblivious transfer (OT). In various embodiments, N may be 2, 4, or some other number of shares. A first computer can express its data value x in base-N. A second computer can form an ×N matrix comprising random numbers m.sub.i,0 and the remaining elements m.sub.i,j=(yjN.sup.i-m.sub.i,0) mod . The first computer can receive an output vector from the OT, having components m.sub.i=(yx.sub.i N.sup.i-m.sub.i,0) mod .

Provided is an apparatus for classifying an attribute of an image object, including: a first memory configured to store target object images that are indexed; a second memory configured to store target object images that are un-indexed; and an object attribute classification module configured to perform learning on the un-indexed target object images to construct a classifier for classifying a detailed attribute of target object, and finely adjust the classifier on the basis of the indexed target object images.

Various embodiments of the present invention provide methods, apparatus, systems, computing devices, computing entities, and/or the like for performing augmented reality assistance mode functionalities. Certain embodiments utilize systems, methods, and computer program products that perform augmented reality assistance mode functionalities by using at least one of environment familiarity predictions, assistance mode triggering need determinations, and threat detection machine learning models.

A method of updating data cluster, adapted to a computing device, includes receiving update data, and calculating a first distance between the update data and an existing representative of an existing cluster, determining whether the first distance is smaller than a threshold distance, updating the existing cluster with the update data to generate an updated cluster when the first distance is smaller than the threshold distance, and performing a representative updating procedure on the updated cluster to generate an updated representative.

Finding a medical care provider that serves a patient's unique needs is a highly personalized process. Embodiments herein describe a patient provider matching system for providing a personalized experience that highlights factors that are most important for patients in choosing their medical care provider. The patient provider matching system receives a query by a patient user and identifies a list of medical providers that match the query based on provider qualifications and medical claims data. The patient provider matching system, ranks the identified list of medical providers based on patient data and provider data. The patient provider matching system, displays the ranked list of medical providers on a graphical user interface for the patient user. Further details of the patient provider matching system are provided below.

The present disclosure provides an improved approach to implement structure learning of neural networks by exploiting correlations in the data/problem the networks aim to solve. A greedy approach is described that finds bottlenecks of information gain from the bottom convolutional layers all the way to the fully connected layers. Rather than simply making the architecture deeper, additional computation and capacitance is only added where it is required.

A method and system for generating clusters for feedback data may include receiving a request for clustering feedback data, the request including one or more parameters relating to the feedback data, retrieving a plurality of vectorized feedbacks stored in a key value format, the key value format including a feedback identifier used as a key for each of the plurality of vectorized feedbacks, creating a plurality of feedback clusters based on the one or more parameters and the retrieved plurality of vectorized feedbacks, the plurality of feedback clusters categorizing at least some of the feedback data into the plurality of feedback clusters, and transmitting data relating to the plurality of feedback clusters for display on a user interface screen.

The present disclosure relates to a method and an apparatus for classifying data points using convex hulls based on centroids of clusters, and a method for classifying data points according to one embodiment of the present disclosure comprises clustering data points into a plurality of clusters; constructing a hyperplane by using a set of centroids of singular clusters having a single class label from the plurality of clusters and removing singular clusters whose centroids are not used to construct the hyperplane; generating a convex hull for a singular cluster used to construct the hyperplane; removing internal data points except for the vertices of the generated convex hull from the singular cluster whose centroid is used to construct the hyperplane; and classifying a set of remaining data points except for the removed internal data points among the plurality of clusters.

A method for enabling website access is provided. The method includes detecting an attempt to access a particular website by a computing device via a network, the particular website including one or more webpages, and accessing a particular data privacy policy for the particular website. Scores of the particular data privacy policy are determined based on text of the particular data privacy policy, and a particular multidimensional coordinate is determined based on the scores of the particular data privacy policy. A map including the particular multidimensional coordinate is displayed via the computing device. An instruction from a user is received via the computing device to enable accessing of the particular website, and the accessing by the computing device of the particular website is enabled in response to the instruction from the user.