A method for employing a differentiable ranking based graph sparsification (DRGS) network to use supervision signals from downstream tasks to guide graph sparsification is presented. The method includes, in a training phase, generating node representations by neighborhood aggregation operators, generating sparsified subgraphs by top-k neighbor sampling from a learned neighborhood ranking distribution, feeding the sparsified subgraphs to a task, generating a prediction, and collecting a prediction error to update parameters in the generating and feeding steps to minimize an error, and, in a testing phase, generating node representations by neighborhood aggregation operators related to testing data, generating sparsified subgraphs by top-k neighbor sampling from a learned neighborhood ranking distribution related to the testing data, feeding the sparsified subgraphs related to the testing data to a task, and outputting prediction results to a visualization device.

There is described a neural network system implemented by one or more computers for determining graph similarity. The neural network system comprises one or more neural networks configured to process an input graph to generate a node state representation vector for each node of the input graph and an edge representation vector for each edge of the input graph; and process the node state representation vectors and the edge representation vectors to generate a vector representation of the input graph. The neural network system further comprises one or more processors configured to: receive a first graph; receive a second graph; generate a vector representation of the first graph; generate a vector representation of the second graph; determine a similarity score for the first graph and the second graph based upon the vector representations of the first graph and the second graph.

Various embodiments are generally directed to systems for summarizing data visualizations (i.e., images of data visualizations), such as a graph image, for instance. Some embodiments are particularly directed to a personalized graph summarizer that analyzes a data visualization, or image, to detect pre-defined patterns within the data visualization, and produces a textual summary of the data visualization based on the pre-defined patterns detected within the data visualization. In various embodiments, the personalized graph summarizer may include features to adapt to the preferences of a user for generating an automated, personalized computer-generated narrative. For instance, additional pre-defined patterns may be created for detection and/or the textual summary may be tailored based on user preferences. In some such instances, one or more of the user preferences may be automatically determined by the personalized graph summarizer without requiring the user to explicitly indicate them. Embodiments may integrate machine learning and computer vision concepts.

A method, an electronic device and computer readable medium for entity-relationship embeddings using automatically generated entity graphs instead of a traditional knowledge graph are provided. The method includes receiving, by a processor, an input text. The method also includes identifying a primary entity, a secondary entity and a context from the input text, wherein the context comprises a relationship between the primary entity and the secondary entity. The method additionally includes generating, by the processor, an entity context graph based on the primary entity, the secondary entity, and the context by: extracting, from the context, one or more text segments comprising a plurality of words describing one or more additional relationships between the primary entity and the secondary entity, and generating a plurality of context triples from the one or more text segments, each of the plurality of context triples defining a respective relationship between primary entity and the secondary entity.

The invention relates to a computer-implemented method for automatically extracting data from a scatter plot. The method comprises receiving a digital image of a scatter plot; analyzing the received digital image for identifying a plurality of pixel sets, each pixel set being a group of adjacent pixels; analyzing the pixel sets in the received image or in a derivative of the received image for generating a plurality of templates; comparing the templates with pixels of a target image for identifying matching templates; identifying data points for the identified similar templates; assigning to each identified data point a data series; and returning the identified data points.

An image comparison unit (81) compares a query image with a registered image to detect, in the registered image, a region corresponding to the query image. An action information determining unit (82), on the basis of intermediate information in which sub-region information identifying sub-regions in the registered image and action information representing information processing to be executed by a target device are associated with each other, identifies sub-regions on the basis of the sub-region information, chooses a sub-region having the highest degree of matching with the detected region among the identified sub-regions, and identifies action information corresponding to the chosen sub-region. An action information execution unit (83) causes the target device to execute information processing corresponding to the action information.

A pattern extraction apparatus for extracting a pattern of event occurrence from event time-series data generates an adjacent event graph by fetching adjacent events from the event time-series data, representing each of the adjacent events as a node, connecting the nodes by a directed link having a transition direction between the adjacent events and a weight, representing identical events as a single node, and, if there are a plurality of directed links between identical adjacent events, accumulating weights of the directed links into a single directed link. The pattern extraction apparatus cuts a directed link having an evaluation value smaller than or equal to a predetermined value in the generated adjacent event graph, the expected value being obtained based on the weight of the directed link.

Described herein are methods and systems for closed-form 3D model generation of non-rigid complex objects from scans with large holes. A computing device receives (i) a partial scan of a non-rigid complex object captured by a sensor coupled to the computing device; (ii) a partial 3D model corresponding to the object, and (iii) a whole 3D model corresponding to the object, wherein the partial 3D scan and the partial 3D model each includes one or more large holes. The device performs a rough match on the partial 3D model and changes the whole 3D model using the rough match to generate a deformed 3D model. The device refines the deformed 3D model using a deformation graph, reshapes the refined deformed 3D model to have greater detail, and adjusts the whole 3D model according to the reshaped 3D model to generate a closed-form 3D model that closes holes in the scan.

Automated methods and systems described herein are directed to identifying potential root causes of a problem in a data center. Methods and systems receipt an alert or other notification of a problem occurring in a data center and a time when the problem was noticed. A search window is created based on the time and a stream of log messages generated in the search window is converted into a time dependent metric. An anomaly detection technique is applied to the metric to determine a start time of a problem. Logging events and key phrases in the log messages are identified in the search window and presented as potential root causes of the problem. The potential root cause may then be used by system administrators and/or tenants to diagnose the problem and execute remedial measures to correct the problem.

Examples relate to simultaneous localization and calibration. An example implementation may involve receiving sensor data indicative of markers detected by a sensor on a vehicle located at vehicle poses within an environment, and determining a pose graph representing the vehicle poses and the markers. For instance, the pose graph may include edges associated with a cost function representing a distance measurement between matching marker detections at different vehicle poses. The distance measurement may incorporate the different vehicle poses and a sensor pose on the vehicle. The implementation may further involve determining a sensor pose transform representing the sensor pose on the vehicle that optimizes the cost function associated with the edges in the pose graph, and providing the sensor pose transform. In further examples, motion model parameters of the vehicle may be optimized as part of a graph-based system as well or instead of sensor calibration.