Abstract: A method for probabilistic processing of data, wherein the data is provided in form of a data set S composed of multidimensional n-tuples of the form (x1, . . . , xn), is characterized in that an n-dimensional data structure is generated by way of providing a bit matrix, providing a number K of independent hash functions Hk that are employed in order to address the bits in the matrix, and inserting the n-tuples (x1, . . . , xn) into the bit matrix by computing the hash values Hk(x) for all values x of the n-tuple for each of the number K of independent hash functions Hk, and by setting the resulting bits [Hk(x1), . . . , Hk(xn)] of the matrix. Furthermore, a respective system is disclosed.
Type:
Grant
Filed:
September 29, 2010
Date of Patent:
April 5, 2016
Assignee:
NEC EUROPE LTD.
Inventors:
Andrea Di Pietro, Felipe Huici, Saverio Niccolini
Abstract: Generating an emotional representation of received data content is provided. Data content corresponding to a user is received. In responsive to determining that the user requested an emotional representation of a predicted emotional reaction by the user to the received data content, the emotional representation of the received data content is generated based on the predicted emotional reaction by the user to the received data content.
Type:
Grant
Filed:
May 15, 2014
Date of Patent:
March 29, 2016
Assignee:
International Business Machines Corporation
Inventors:
Aleksandr Aravkin, Dimitri Kanevsky, Peter K. Malkin, Tara N. Sainath
Abstract: Systems and methods for forecasting events can be provided. A measurement database can store sensor measurements, each having been provided by a non-portable electronic device with a primary purpose unrelated to collecting measurements from a type of sensor that collected the measurement. A measurement set identifier can select a set of measurements. The electronic devices associated with the set of measurements can be in close geographical proximity relative to their geographical proximity to other devices. An inter-device correlator can access the set and collectively analyze the measurements. An event detector can determine whether an event occurred. An event forecaster can forecast a future event property. An alert engine can identify one or more entities to be alerted of the future event property, generate at least one alert identifying the future event property, and transmit the at least one alert to the identified one or more entities.
Type:
Grant
Filed:
April 2, 2015
Date of Patent:
March 15, 2016
Assignee:
Google Inc.
Inventors:
John B. Filson, Eric B. Daniels, Adam Mittleman, Sierra L. Nelmes, Yoky Matsuoka
Abstract: In an embodiment, an automaton determinization method includes: state-generating, first-transition-generating, second-transition-generating, and first-deleting. The state-generating includes generating, assigned with a first symbol, a second state newly. The first-transition-generating includes generating a second transition that leaves from the first state and enters to the second state and that is assigned with the first symbol. The second-transition-generating includes generating, regarding the first transitions, a fourth transition where a state previous to a third transition is substituted with the second state. The third transition is an outgoing transition from a next state of the first transition.
Abstract: Systems and Methods for multi-modal or multimedia image retrieval are provided. Automatic image annotation is achieved based on a probabilistic semantic model in which visual features and textual words are connected via a hidden layer comprising the semantic concepts to be discovered, to explicitly exploit the synergy between the two modalities. The association of visual features and textual words is determined in a Bayesian framework to provide confidence of the association. A hidden concept layer which connects the visual feature(s) and the words is discovered by fitting a generative model to the training image and annotation words. An Expectation-Maximization (EM) based iterative learning procedure determines the conditional probabilities of the visual features and the textual words given a hidden concept class. Based on the discovered hidden concept layer and the corresponding conditional probabilities, the image annotation and the text-to-image retrieval are performed using the Bayesian framework.
Type:
Grant
Filed:
June 1, 2012
Date of Patent:
March 8, 2016
Assignee:
The Research Foundation for The State University of New York
Abstract: Methods and systems for feature extraction of LIDAR surface manifolds. LIDAR point data with respect to one or more LIDAR surface manifolds can be generated. An AHAH-based feature extraction operation can be automatically performed on the point data for compression and processing thereof. The results of the AHAH-based feature extraction operation can be output as a compressed binary label representative of the at least one surface manifold rather than the point data to afford a high-degree of compression for transmission or further processing thereof. Additionally, one or more voxels of a LIDAR point cloud composed of the point data can be scanned in order to recover the compressed binary label, which represents prototypical surface patches with respect to the LIDAR surface manifold(s).
Abstract: Described herein are systems and methods for identifying herbal ingredients effective in treating illnesses in Traditional Chinese Medicine (TCM) using an artificial neural network.
Type:
Grant
Filed:
July 28, 2011
Date of Patent:
March 1, 2016
Assignee:
HERBMINERS INFORMATICS LIMITED
Inventors:
Wilfred Wan Kei Lin, Jackei Ho Kei Wong, Allan Kang Ying Wong, Patricia Mary Hutton
Abstract: Embodiments are directed towards online content classification that includes training a machine learning system. A batch of data items may be randomly selected from unlabeled test data. The batch of data items may be communicated to a client computer enabling a user to label each data item based on the contents of each data item. These labeled data items may be employed to train the machine learning system. While a classification result score is less than a threshold value, iteration may be performed to train the machine learning system. For each iteration another batch of data items may be selected from the unlabeled test data. This batch of data items may be classified using the machine learning system. The batch of classified data items may be communicated back to the client computer to be labeled by the user.
Abstract: An illustrative data classifier device includes data storage and at least one processor configured to operate as a query engine and a passive classifier that is configured to predict classification labels for data. The processor is configured to determine a relationship between the data and training data with associated training classification labels. The processor is also configured to assign a weighted version of at least one of the training classification labels to at least one member of the data based on the determined relationship. An illustrative method of classifying data includes predicting classification labels for data by determining a relationship between the data and training data with associated training classification labels. A weighted version of at least one of the training classification labels is assigned to at least one member of the data based on the determined relationship.
Abstract: Embodiments generally relate to methods of accurately predicting seasonal fluctuations in precipitation or other approximate functionals of a climate state space, such as the number of heating or cooling degree days in a season, maximum river flow rates, water table levels and the like. In one embodiment, a method for predicting climate comprises: deriving a climate attractor from a global climate model, wherein a tuning parameter for the climate attractor comprises a value of total energy for moving air and water on the earth's surface; estimating a predictive function for each of a plurality of computational cells within the global climate model; and predicting an approximate climate functional of interest for a given specific location utilizing a combination of the predictive functions from each of the plurality of computational cells geographically proximate the location, where at all stages, predictive functions are selected in part by comparison to historical data.
Abstract: A method for selectively eliminating nondeterministic elements of NFA is disclosed. The method includes steps of: (a) determining a specific state calculated to have a highest arrival probability through a transition from a current state among all states in the NFA as a current highest probability state; (b) determining whether there exists at least one common transition between a first set of transitions including at least one transition moving the current highest probability state to a state i and a second set of transitions including at least one transition moving the current highest probability state to a state j in the NFA; and (c) excluding the at least one common transition between the first and the second sets of transitions and creating a state k which is arrived as a result of moving from the current highest probability state through the at least one common transition.
Abstract: In accordance with embodiments of the present disclosure, a system may comprise a plurality of slots each configured to receive a modular information handling system, a plurality of air movers each configured to cool at least one modular information handling system disposed in at least one of the plurality slots, and a chassis management controller communicatively coupled to the plurality of slots and the plurality of air movers and configured to display a recommended placement of modular information handling systems in the plurality of slots based on at least one of: identities of slots populated with modular information handling systems, an airflow ranking of the plurality of slots, an impedance ranking of information handling systems disposed in the slots, and a workload of each of the information handling systems disposed in the slots.
Abstract: A method for estimating artist ambiguity in a dataset is performed at a device with a processor and memory storing instructions for execution by the processor. The method includes applying a statistical classifier to a first dataset including a plurality of media items, wherein each media item is associated with one of a plurality of artist identifiers, each artist identifier identifies a real world artist, and the statistical classifier calculates a respective probability that each respective artist identifier is associated with media items from two or more different real world artists based on a respective feature vector corresponding to the respective artist identifier. The method further includes providing a report of the first dataset, including the calculated probabilities, to a user of the electronic device. Each respective feature vector includes a plurality of features that indicate likelihood of artist ambiguity.
Abstract: A method for increasing object detection rates or object recognition rates by using a classifier is disclosed. The method includes the steps of: (a) the classifier acquiring a covariance matrix by using values of at least one channel of at least some pixels included in a local block having a smaller size than detection windows of respective image samples including positive image samples and hard negative image samples while moving the local block within the detection windows; and (b) the classifier acquiring a transform matrix w for transforming at least one feature vector x of an image to be inputted later by using the covariance matrix.
Abstract: Apparatus and methods for feedback in a spiking neural network. In one approach, spiking neurons receive sensory stimulus and context signal that correspond to the same context. When the stimulus provides sufficient excitation, neurons generate response. Context connections are adjusted according to inverse spike-timing dependent plasticity. When the context signal precedes the post synaptic spike, context synaptic connections are depressed. Conversely, whenever the context signal follows the post synaptic spike, the connections are potentiated. The inverse STDP connection adjustment ensures precise control of feedback-induced firing, eliminates runaway positive feedback loops, enables self-stabilizing network operation. In another aspect of the invention, the connection adjustment methodology facilitates robust context switching when processing visual information. When a context (such an object) becomes intermittently absent, prior context connection potentiation enables firing for a period of time.
Type:
Grant
Filed:
May 7, 2012
Date of Patent:
December 29, 2015
Assignee:
Brain Corporation
Inventors:
Filip Piekniewski, Eugene Izhikevich, Botond Szatmary, Csaba Petre
Abstract: A method for recommending an application includes obtaining an input model representing user interaction patterns during execution of a first application. The input model is compared to a reference model representing user interaction patterns during execution of a second application. A similarity is determined between the input model and the reference model. A recommendation of the second application is generated in response to the similarity.
Abstract: A method for classifying unclassified vehicle tracks using real-time sensor data comprises acquiring a computer database of vehicle track characteristics data for known vehicle tracks; defining vehicle track signatures based on the vehicle track characteristics data; and generating a graph based on the vehicle track signatures, the graph having state nodes representative of distinguishing vehicle track characteristics defined in the vehicle track signatures and links between the state nodes representative of relationships between distinguishing vehicle track characteristics defined in the vehicle track signatures, said graph also having reporting nodes for classifying unknown vehicle tracks when sufficient distinguishing vehicle track characteristics have been observed. The method also comprises processing with the graph, using a processor, real-time sensor data corresponding to a first unclassified vehicle track until the first unclassified vehicle track is classified.
Type:
Grant
Filed:
September 14, 2012
Date of Patent:
December 8, 2015
Assignee:
Lockheed Martin Corporation
Inventors:
Richard N. Pedersen, Mark E. Barbustiak
Abstract: Artificial intelligence methods and systems for triggering the generation of narratives are disclosed. Specific embodiments relate to real-time evaluation and automated generation of narrative stories based on received data. For example, data can be tested against data representative of a plurality of story angles to determine whether a narrative story incorporating one or more such story angles is to be automatically generated.
Type:
Grant
Filed:
November 26, 2013
Date of Patent:
December 8, 2015
Assignee:
Narrative Science Inc.
Inventors:
Nathan Nichols, Michael Justin Smathers, Lawrence Birnbaum, Kristian Hammond, Lawrence E. Adams
Abstract: Provided are systems and methods for drilling fluids expert systems using Bayesian decision networks to determine drilling fluid recommendations. A drilling fluids expert system includes a drilling fluids Bayesian decision network (BDN) model that receives inputs and outputs recommendations based on Bayesian probability determinations. The drilling fluids BDN model includes a temperature ranges uncertainty node, a formation uncertainty node, a potential hole problems uncertainty node, and a drilling fluids decision node.
Type:
Grant
Filed:
March 14, 2013
Date of Patent:
December 1, 2015
Assignees:
Saudi Arabian Oil Company, The Texas A&M University System
Inventors:
Abdullah Saleh Hussain Al-Yami, Jerome Schubert
Abstract: Systems and methods are disclosed herein for classifying records, such as product records, using a machine learning algorithm. After training a classification model according to a machine learning algorithm using an initial training set, records are classified and high confidence classifications identified. Remaining classifications are submitted to a crowdsourcing forum that validates or invalidates the classifications or marks them as to unclear to evaluate. Invalidated classifications are automatically analyzed to identify one or both of classification values and categories having a high proportion of invalidated classifications. Requests are transmitted to analysts to generate training data that is added to the training set. The process of classifying records and obtaining crowdsourced validation thereof may then repeat. High confidence classifications may be identified using an accuracy model trained to relate an accuracy percentage to a confidence score output by the classification model.