Classification Or Recognition Patents (Class 706/20)
  • Patent number: 9697236
    Abstract: Architecture that addresses page information lost as part of a selection process in a search engine framework. An aggregation process collects all page or document information from the same image cluster and uses the aggregated page information to annotate one or more selected image-page pairs within the same image cluster. Once the entire set of descriptive terms is received, the entire set of descriptive terms or only an optimum set of top N descriptive terms of the entire set is for annotation of one or more of the representative images in the cluster.
    Type: Grant
    Filed: December 5, 2014
    Date of Patent: July 4, 2017
    Assignee: Microsoft Technology Licensing, LLC
    Inventors: Pak-Ming Cheung, Bo Geng, Xin Yu, Arun Sacheti
  • Patent number: 9687199
    Abstract: A medical imaging system that processes input data (imaging and/or non imaging) having high dimensionality and few samples to learn from, by using multiple ranks of machine learning modules each dealing with a separate portion of the clinical data. The outputs of the individual machine learning modules are the combined to provide a result reflective of the complete image data set.
    Type: Grant
    Filed: September 9, 2015
    Date of Patent: June 27, 2017
    Assignee: Wisconsin Alumni Research Foundation
    Inventors: Vamsi K Ithapu, Vikas Singh, Sterling C Johnson, Ozioma C Okonkwo
  • Patent number: 9684714
    Abstract: A mechanism is provided in a data processing system for using paraphrase metrics for answering questions. The mechanism receives an input question and generating a candidate answer from a corpus of information. The candidate answer has a supporting passage from the corpus of information. The mechanism divides the input question into a first sequence of tokens and divides the supporting passage into a second sequence of tokens. The mechanism identifies a plurality of subsequences of tokens within the second sequence of tokens and applies a paraphrase metric to compare the first sequence of tokens to each of the plurality of subsequences of tokens to generate a plurality of paraphrase metric scores. The mechanism then determines a confidence score for the candidate answer based on a highest paraphrase metric score within the plurality of paraphrase metric scores.
    Type: Grant
    Filed: December 22, 2014
    Date of Patent: June 20, 2017
    Assignee: International Business Machines Corporation
    Inventors: Anthony T. Levas, James W. Murdock, IV, John M. Prager, Benjamin P. Segal, Timothy P. Winkler
  • Patent number: 9681147
    Abstract: A computer-implemented method is provided for compressing a collection of images. The method includes: receiving image data for an image ensemble composed of a plurality of images; dividing the image data into a plurality of data blocks; decomposing each data block from the plurality of data blocks into a set of rank-one tensors by estimating each rank-one tensor individually, where each rank-one tensor is represented by three one-dimensional vectors; arranging the rank-one tensors for each of the plurality of data blocks into a two-dimensional matrix, where the decomposer is implemented by computer processor; compressing the two-dimensional matrix using a compression operation; and storing the compressed two-dimensional matrix in a non-transitory data store.
    Type: Grant
    Filed: January 19, 2015
    Date of Patent: June 13, 2017
    Assignee: Board of Trustees of Michigan State University
    Inventors: Abo Talib Mahfoodh, Hayder Radha
  • Patent number: 9667575
    Abstract: A computer-implemented method for detecting webpages belonging to spam campaigns may include (1) identifying a web address of a suspicious webpage that potentially hosts a spam message, (2) capturing an image of the suspicious webpage, (3) comparing the image of the suspicious webpage to at least one spam image from a spam database, the spam image being associated with a spam campaign in the spam database, (4) determining, based on the comparison of the image of the suspicious webpage with the spam image, whether the suspicious webpage is associated with the spam campaign, and (5) updating the spam database in response to the determination of whether the suspicious webpage is associated with the spam campaign. Various other methods, systems, and computer-readable media are also disclosed.
    Type: Grant
    Filed: November 4, 2013
    Date of Patent: May 30, 2017
    Assignee: Symantec Corporation
    Inventor: Slawomir Grzonkowski
  • Patent number: 9661025
    Abstract: Methods and apparatuses for identifying and detecting threats to an enterprise or e-commerce system are disclosed, including grouping log lines belonging to one or more log line parameters from one or more enterprise or e-commerce system data sources and/or from incoming data traffic to the enterprise or e-commerce system; extracting one or more features from the grouped log lines into one or more features tables; using one or more statistical models on the one or more features tables to identify statistical outliers; labeling the statistical outliers to create one or more labeled features tables; using the one or more labeled features tables to create one or more rules for identifying threats to the enterprise or e-commerce system; and using the one or more rules on incoming enterprise or e-commerce system data traffic to detect threats to the enterprise or e-commerce system. Other embodiments are described and claimed.
    Type: Grant
    Filed: September 7, 2016
    Date of Patent: May 23, 2017
    Assignee: PATTERNEX, INC.
    Inventors: Constantinos Bassias, Vamsi Korrapati, Uday Veeramachaneni
  • Patent number: 9652720
    Abstract: In one embodiment, network data is received at a Learning Machine (LM) in a network. It is determined whether the LM recognizes the received network data based on information available to the LM. When the LM fails to recognize the received network data: a connection to a central management node is established, a request is sent for information relating to the unrecognized network data to the central management node, and information is received from the central management node in response to the request. The received information assists the LM in recognizing the unrecognized network data.
    Type: Grant
    Filed: July 9, 2013
    Date of Patent: May 16, 2017
    Assignee: Cisco Technology, Inc.
    Inventors: Jean-Philippe Vasseur, Grégory Mermoud, Sukrit Dasgupta
  • Patent number: 9652449
    Abstract: A method, computer readable medium and apparatus for detecting a sentiment for a short message are disclosed. For example, the method receives the short message, and obtains an abstraction of the short message. The method then determines the sentiment of the short message based upon the abstraction.
    Type: Grant
    Filed: April 20, 2015
    Date of Patent: May 16, 2017
    Assignee: AT&T Intellectual Property I, L.P.
    Inventors: Luciano De Andrade Barbosa, Junlan Feng
  • Patent number: 9646226
    Abstract: Automatic selection of training images is enhanced using an instance-weighted mixture modeling framework called ARTEMIS. An optimization algorithm is derived that in addition to mixture parameter estimation learns instance-weights, essentially adapting to the noise associated with each example. The mechanism of hypothetical local mapping is evoked so that data in diverse mathematical forms or modalities can be cohesively treated as the system maintains tractability in optimization. Training examples are selected from top-ranked images of a likelihood-based image ranking. Experiments indicate that ARTEMIS exhibits higher resilience to noise than several baselines for large training data collection. The performance of ARTEMIS-trained image annotation system is comparable to using manually curated datasets.
    Type: Grant
    Filed: April 16, 2014
    Date of Patent: May 9, 2017
    Assignee: The Penn State Research Foundation
    Inventors: James Z. Wang, Neela Sawant, Jia Li
  • Patent number: 9619634
    Abstract: The invention concerns a method of identification of a person to be identified, comprising: receiving, from an input device (106A to 106C), user input data of the person to be identified and location data indicating the location of the person to be identified; and identifying by a processing device, in a user database (108) storing a plurality of user records of registered users, each record comprising user reference data and historical location information of a registered user, a record of the person to be identified based on the user input data and the location data, wherein the historical location information of at least one of the registered users includes an association of time and location data provided by a user location device associated with the registered user.
    Type: Grant
    Filed: August 29, 2014
    Date of Patent: April 11, 2017
    Assignee: ACCENTURE GLOBAL SERVICES LIMITED
    Inventors: Cyrille Bataller, Giuseppe Giordano
  • Patent number: 9619727
    Abstract: An inspection device that performs pattern matching on a searched image performs matching between a template image of an inspection object and the searched image by using: a feature region extraction process unit that extracts a feature quantity from the template image acquired for learning; a feature quantity extraction process unit that extracts a feature quantity from the searched image acquired for learning; a mutual feature quantity calculation process unit that calculates a mutual feature quantity of the template image and the searched image from the feature quantity extracted from the template image and the feature quantity extracted from the searched image; a learning process unit that calculates, using a plurality of the mutual feature quantities, a discrimination boundary surface that determines matching success or failure; a process unit that calculates a plurality of the mutual feature quantities from an image acquired from the inspection object; and the plurality of mutual feature quantities and
    Type: Grant
    Filed: July 16, 2013
    Date of Patent: April 11, 2017
    Assignee: Hitachi High-Technologies Corporation
    Inventors: Wataru Nagatomo, Yuichi Abe, Hiroyuki Ushiba
  • Patent number: 9619572
    Abstract: In embodiments of the present invention improved capabilities are described for multiple web-based content category searching for web content on a mobile communication facility comprising capturing speech presented by a user using a resident capture facility on the mobile communication facility; transmitting at least a portion of the captured speech as data through a wireless communication facility to a speech recognition facility; generating speech-to-text results for the captured speech utilizing the speech recognition facility; and transmitting the text results and a plurality of formatting rules specifying how search text may be used to form a query for a search capability on the mobile communications facility, wherein each formatting rule is associated with a category of content to be searched.
    Type: Grant
    Filed: December 15, 2014
    Date of Patent: April 11, 2017
    Assignee: Nuance Communications, Inc.
    Inventors: Michael S. Phillips, John N. Nguyen
  • Patent number: 9613296
    Abstract: Systems and methods are provided for identifying a subset of exemplar images in a set of exemplary images that may be used to generate an image classifier having one or more desirable performance characteristics. A computing system may select every possible subset of exemplar images from a set of exemplar images and may generate a classifier for each selected subset of images. The computing system may use each generated classifier to classify one or more images in order to generate performance statistics related to how each classifier performed when classifying the one or more images. The computing system may identify a particular subset of exemplar images that is associated with performance statistics of a classifier that satisfies one or more desired performance characteristics.
    Type: Grant
    Filed: October 7, 2015
    Date of Patent: April 4, 2017
    Assignee: GumGum, Inc.
    Inventors: Daniel Scott Rosen, Cambron Neil Carter
  • Patent number: 9607272
    Abstract: A predictive coding system updates a plurality of training documents for an untrained classification model based on a plurality of additional documents. The plurality of additional documents are selected from a plurality of unlabeled documents based on a decision hyperplane associated with a first trained classification model. The predictive coding system provides the updated plurality of training documents to the untrained classification model to cause the untrained classification model to be retrained and to cause a second trained classification model to be generated.
    Type: Grant
    Filed: March 15, 2013
    Date of Patent: March 28, 2017
    Assignee: Veritas Technologies LLC
    Inventors: Shengke Yu, Venkat Rangan
  • Patent number: 9600582
    Abstract: When objects are shared by one user with another user, objectionable content, if identified as such, can be blocked from being shared, while the remainder of the shared objects can be accessed by the other user. Functions that allow sharing of content are implemented so as prevent sharing of objectionable content with another user, while allowing other content to be shared. If a group of files or objects is shared, then the presence of objectionable content in one object in the group results in that objectionable content not being shared, but the remaining files or objects are still shared. A graphical user interface for accessing the storage system, whether by providers or recipients of shared content, can selectively render information about objects with objectionable content.
    Type: Grant
    Filed: May 23, 2013
    Date of Patent: March 21, 2017
    Assignee: Microsoft Technology Licensing, LLC
    Inventors: Ryan Zacher, Jorge Pozas Trevino, Christopher Wolf, Gary Anderson, Jeff Ferreira, Mark Kramer, Kathleen Ewing, Anthony Charles DiCola, Emmanuel Miranda-Steiner
  • Patent number: 9600761
    Abstract: Embodiments of the invention provide a method comprising creating a structural description for at least one neurosynaptic core circuit. Each core circuit comprises an interconnect network including plural electronic synapses for interconnecting one or more electronic neurons with one or more electronic axons. The structural description defines a desired neuronal activity for the core circuits. The desired neuronal activity is simulated by programming the core circuits with the structural description. The structural description controls routing of neuronal firing events for the core circuits.
    Type: Grant
    Filed: January 14, 2016
    Date of Patent: March 21, 2017
    Assignee: International Business Machines Corporation
    Inventor: Dharmendra S. Modha
  • Patent number: 9594817
    Abstract: The technique relates to a system and method for selecting dynamic workflow to process file data. This technique facilitates to process file data and business rules simultaneously and thus saves a lot of time. The technique involves receiving a request from File Pattern Recognition Engine (FPRE) to select rules for processing file data by Data Rule Engine (DRE) and to select business rules to be applied on the data by the Business Metrics Engine (BME). After that, DRE and BME select the possible set of rules and send the acknowledgement to FPRE. Thereafter, FPRE sends the signal to proceed to DRE and BME, and after receiving the signal DRE and BME apply at least one of the possible set of rules selected earlier.
    Type: Grant
    Filed: December 17, 2014
    Date of Patent: March 14, 2017
    Assignee: Infosys Limited
    Inventors: Rebeck Carvalho, Sriharsha Anand Pushkala, Rohit Saxena
  • Patent number: 9570088
    Abstract: A signal processor is configured to suppress a noise component contained in an input voice signal by means of coherence filtering. The processor includes an iterative coherence filtering function for repeatedly conducting the coherence filtering on a signal, which has been subjected to the coherence filtering and then input to the processor again as input signal, and performs the iteration processing on the signal obtained by the coherence filtering until a condition for terminating the iteration is satisfied, thereby preventing musical noise from generating while a noise component is suppressed.
    Type: Grant
    Filed: November 20, 2013
    Date of Patent: February 14, 2017
    Assignee: Oki Electric Industry Co., Ltd.
    Inventor: Katsuyuki Takahashi
  • Patent number: 9563855
    Abstract: Systems and methods may provide for using one or more generic classifiers to generate self-training data based on a first plurality of events associated with a device, and training a personal classifier based on the self-training data. Additionally, the one or more generic classifiers and the personal classifier may be used to generate validation data based on a second plurality of events associated with the device. In one example, the personal classifier is substituted for the one or more generic classifiers if the validation data indicates that the personal classifier satisfies a confidence condition relative to the one or more generic classifiers.
    Type: Grant
    Filed: June 27, 2014
    Date of Patent: February 7, 2017
    Assignee: Intel Corporation
    Inventors: Xue Yang, Lakshman Krishnamurthy, Lama Nachman
  • Patent number: 9563854
    Abstract: In one embodiment, a device determines that a machine learning model is to be trained by a plurality of devices in a network. A set of training devices are identified from among the plurality of devices to train the model, with each of the training devices having a local set of training data. An instruction is then sent to each of the training devices that is configured to cause a training device to receive model parameters from a first training device in the set, use the parameters with at least a portion of the local set of training data to generate new model parameters, and forward the new model parameters to a second training device in the set. Model parameters from the training devices are also received that have been trained using a global set of training data that includes the local sets of training data on the training devices.
    Type: Grant
    Filed: January 27, 2014
    Date of Patent: February 7, 2017
    Assignee: Cisco Technology, Inc.
    Inventors: Javier Cruz Mota, Jean-Philippe Vasseur, Andrea Di Pietro
  • Patent number: 9558452
    Abstract: Technologies are generally provided far a hierarchical, feature teed statistical model that cm be used for personalized classification or predictions within a community of users. Personalization refers to learning about the habits and characteristics of individual users and adapting user experiences based on that learning. The model may be used in a communication application to predict user actions on incoming email messages and to help users triage email by making personalized suggestions based on the model predictions. A community of users associated together with the communication application may be incorporated together into a single model to enable for continuous fine-grain interaction between intelligence learned from the community of users as a whole and that learned from individual users.
    Type: Grant
    Filed: November 8, 2013
    Date of Patent: January 31, 2017
    Assignee: MICROSOFT TECHNOLOGY LICENSING, LLC
    Inventors: John Guiver, John Winn, James Edelen
  • Patent number: 9547471
    Abstract: Conversational interactions between humans and computer systems can be provided by a computer system that classifies an input by conversation type, and provides human authored responses for conversation types. The input classification can be performed using trained binary classifiers. Training can be performed by labeling inputs as either positive or negative examples of a conversation type. Conversational responses can be authored by the same individuals that label the inputs used in training the classifiers. In some cases, the process of training classifiers can result in a suggestion of a new conversation type, for which human authors can label inputs for a new classifier and write content for responses for that new conversation type.
    Type: Grant
    Filed: July 3, 2014
    Date of Patent: January 17, 2017
    Assignee: Microsoft Technology Licensing, LLC
    Inventors: Jason Williams, Geoffrey Zweig, Aparna Lakshmiratan, Carlos Garcia Jurado Suarez
  • Patent number: 9541922
    Abstract: A position measurement portion manages a measurement target area by dividing the area into a plurality of section areas, and also manages, or keeps track of, an intra-area interfering object number which is the number of interfering objects in a given section area, independently for each of the plurality of the section areas. The travel controller determines, independently for each of the plurality of section areas, if the section area is in a congested state in which the intra-area interfering object number is greater than a set upper limit number, and selects a congested area travel mode as the travel mode when the travel controller is about to cause the vehicle to travel in a section area that is determined to be in the congested state.
    Type: Grant
    Filed: April 14, 2014
    Date of Patent: January 10, 2017
    Assignee: Daifuku Co., Ltd.
    Inventors: Masanori Tsujimoto, Yoshimasa Hara, Natsuo Takagawa, Toshisada Mariyama, Koji Kubota
  • Patent number: 9542256
    Abstract: An analysis module is configured to receive data associated with an event flow. The data is received from a first analysis module (e.g., in a stack of analysis modules) or from the event flow. The analysis module is configured to execute an analysis operation on the data to generate a result. The analysis module can output the result to a second analysis module (e.g., in the stack of analysis modules) or to a user interface.
    Type: Grant
    Filed: January 22, 2015
    Date of Patent: January 10, 2017
    Assignee: Microsoft Technology Licensing, LLC
    Inventors: Olivier Nano, Ivo Jose Garcia dos Santos, Tihomir Tarnavski, Sebastien Peray, Marcel Tilly, Ramkumar Krishnan
  • Patent number: 9529777
    Abstract: A system and method is shown for identifying abnormal client behavior with respect to communications between one or more servers and one or more client devices communicatively coupled to the one or more servers. Messages are received at a server from one or more client devices communicatively coupled to the server. The plurality of messages are grouped into subsets of messages using a learn module of the server. Each subset of messages is associated with a unique client identifier, and all messages within a subset are associated with the same unique client identifier. Each message within a subset of messages is identified as belonging to a defined type of message. Sequences of the defined types of messages within each of said subsets of messages are recorded using the learn module. Time intervals between the defined types of messages are measured using the learn module.
    Type: Grant
    Filed: October 26, 2012
    Date of Patent: December 27, 2016
    Assignee: ELECTRONIC ARTS INC.
    Inventors: Andrew Tjew, Wilson Chan
  • Patent number: 9519910
    Abstract: A system and method that calibrates subject data for which a relationship to a target population is not known, so that the calibrated subject data can more accurately represent the target population. In many cases the calibration will involve the use of a differential weighting scheme applied to the data at the constituent level. The system and method allows the values of the observed variables in the subject data set to be weighted so that their incidence is equivalent to that of a reference population represented by a reference data set, even if the variables used in the reference data set to make estimates for the reference population were not collected or measured for the subject data set.
    Type: Grant
    Filed: September 19, 2013
    Date of Patent: December 13, 2016
    Assignee: RENTRAK CORPORATION
    Inventors: Charles Palit, Bill Engel, Michael Vinson, Bruce Goerlich
  • Patent number: 9514390
    Abstract: Systems, methods, and non-transitory computer-readable media can receive a first image including a representation of a first user. A second image including a representation of a second user can be received. A first set of poselets associated with the first user can be detected in the first image. A second set of poselets associated with the second user can be detected in the second image. The first image including the first set of poselets can be inputted into a first instance of a neural network to generate a first multi-dimensional vector. The second image including the second set of poselets can be inputted into a second instance of the neural network to generate a second multi-dimensional vector. A first distance metric between the first multi-dimensional vector and the second multi-dimensional vector can be determined.
    Type: Grant
    Filed: December 17, 2014
    Date of Patent: December 6, 2016
    Assignee: Facebook, Inc.
    Inventors: Lubomir Bourdev, Ning Zhang, Balamanohar Paluri, Yaniv Taigman, Robert D. Fergus
  • Patent number: 9514240
    Abstract: A system to optimize a website for a search of a topic includes a processor, a memory, and instructions to cause the processor to receive at least one term related to the topic, and to receive web objects pertaining to the website to be optimized. The system is to calculate at least one metric pertaining to the at least one term and the web objects.
    Type: Grant
    Filed: March 7, 2013
    Date of Patent: December 6, 2016
    Assignee: ACCENTURE GLOBAL SERVICES LIMITED
    Inventors: Alexandre Perry Kamel, Stephen Denis Kirkby, Peter Kellett, Hikaru Phillips, Amelia Jones, Kurt Vander Bogart, Judy A. Koonce
  • Patent number: 9514388
    Abstract: Dimensionality reduction systems and methods facilitate visualization, understanding, and interpretation of high-dimensionality data sets, so long as the essential information of the data set is preserved during the dimensionality reduction process. In some of the disclosed embodiments, dimensionality reduction is accomplished using clustering, evolutionary computation of low-dimensionality coordinates for cluster kernels, particle swarm optimization of kernel positions, and training of neural networks based on the kernel mapping. The fitness function chosen for the evolutionary computation and particle swarm optimization is designed to preserve kernel distances and any other information deemed useful to the current application of the disclosed techniques, such as linear correlation with a variable that is to be predicted from future measurements. Various error measures are suitable and can be used.
    Type: Grant
    Filed: August 12, 2008
    Date of Patent: December 6, 2016
    Assignee: HALLIBURTON ENERGY SERVICES, INC.
    Inventors: Dingding Chen, Syed Hamid, Michael C. Dix
  • Patent number: 9503478
    Abstract: A method includes generating at least one access vector associated with a specified device in an industrial process control and automation system. The specified device has one of multiple device roles. The at least one access vector is generated based on one or more communication policies defining communications between one or more pairs of devices roles in the industrial process control and automation system, where each pair of device roles includes the device role of the specified device. The method also includes providing the at least one access vector to at least one of the specified device and one or more other devices in the industrial process control and automation system in order to control communications to or from the specified device.
    Type: Grant
    Filed: June 19, 2014
    Date of Patent: November 22, 2016
    Assignee: Honeywell International Inc.
    Inventors: Alexander Chernoguzov, Thomas R. Markham, Harshal S. Haridas
  • Patent number: 9483734
    Abstract: A system for translating a boosting algorithm includes an interface communicatively coupled to a processor. The interface is operable to receive a trained boosting model. The processor is operable to identify a plurality of split-node variables associated with the trained boosting model. Each of the plurality of split-node variables comprises a variable name, a cutoff point, and a weight. The processor may aggregate the split-node variables by variable name and cutoff point and then combine the weights of each of the plurality of split-node variables having the same variable name and cutoff point. The processor may then create a linear model based on the combined variables. The processor may further create a performance scorecard based on the combined weight, and create a model evaluation based on the error rate of the trained boosting model.
    Type: Grant
    Filed: January 5, 2016
    Date of Patent: November 1, 2016
    Assignee: Bank of America Corporation
    Inventors: Kasilingam Basker Laxmanan, Yudong Chen, Peng Song
  • Patent number: 9483463
    Abstract: A method, system, and computer program product for extracting text motifs from the electronic documents is disclosed. A user provides a largest-maximal repeat or a super-maximal repeat as a first text block. The occurrences of the first text block are detected to identify the second text blocks in the vicinity of the occurrences of the first text block on the basis of pre-defined parameters. The text motifs are determined by combining the first text block and the second text block. Finally, the text motifs are extracted from the electronic documents.
    Type: Grant
    Filed: September 10, 2012
    Date of Patent: November 1, 2016
    Assignee: Xerox Corporation
    Inventors: Matthias Galle, Jean-Michel Renders
  • Patent number: 9484745
    Abstract: A system includes power electronics inverters connected in a network. The power electronics inverters can utilize measurements at local terminals, without a need to exchange information between other power electronics inverters.
    Type: Grant
    Filed: August 4, 2014
    Date of Patent: November 1, 2016
    Assignees: RUTGERS, THE STATE UNIVERSITY OF NEW JERSEY, THE BOARD OF TRUSTEES OF THE UNIVERSITY OF ILLINOIS
    Inventors: Brian Johnson, Sairaj Dholpe, Abdullah Hamadeh, Philip Krein
  • Patent number: 9471887
    Abstract: A system and method is provided that processes a training database of human-generated requests in each of a plurality of task categories with a machine learning algorithm to develop a task classifier model that may be applied to subsequent user requests to determine the most likely one of the task categories for the subsequent user request.
    Type: Grant
    Filed: September 30, 2015
    Date of Patent: October 18, 2016
    Assignee: NTT DOCOMO Inc.
    Inventors: Hyung Sik Shin, Ronald Sujithan, Sayandev Mukherjee, Hongfeng Yin, Yang Sun, Yoshikazu Akinaga, Pero Subasic
  • Patent number: 9471881
    Abstract: Various embodiments select features from a feature space. In one embodiment, a set of training samples and a set of test samples are received. The set of training samples includes a set of features and a class value. The set of test samples includes the set of features absent the class value. A relevancy with respect to the class value is determined for each of a plurality of unselected features based on the set of training samples. A redundancy with respect to one or more of the set of features is determined for each of the plurality of unselected features in the first set of features based on the set of training samples and the set of test samples. A set of features is selected from the plurality of unselected features based on the relevancy and the redundancy determined for each of the plurality of unselected features.
    Type: Grant
    Filed: January 21, 2013
    Date of Patent: October 18, 2016
    Assignee: International Business Machines Corporation
    Inventors: David Haws, Dan He, Laxmi P. Parida, Irina Rish
  • Patent number: 9471847
    Abstract: Methods and systems for distance metric learning include generating two random projection matrices of a dataset from a d-dimensional space into an m-dimensional sub-space, where m is smaller than d. An optimization problem is solved in the m-dimensional subspace to learn a distance metric based on the random projection matrices. The distance metric is recovered in the d-dimensional space.
    Type: Grant
    Filed: October 27, 2014
    Date of Patent: October 18, 2016
    Assignee: NEC Corporation
    Inventors: Shenghuo Zhu, Yuanqing Lin, Qi Qian
  • Patent number: 9454713
    Abstract: A method to determine image similarities. The method may include obtaining a first image and a second image and determining a discrete transform difference between a first discrete transform of the first image and a second discrete transform of the second image. The method may also include determining multiple first intensity vectors for the first image and determining multiple second intensity vectors for the second image. The method may also include determining an intensity vector difference between the multiple first intensity vectors and the multiple second intensity vectors and determining a color difference between a first color histogram of the first image and a second color histogram of the second image. The method may also include determining a similarity between the first image and the second image based on the discrete transform difference, the intensity vector difference, and the color difference.
    Type: Grant
    Filed: December 30, 2014
    Date of Patent: September 27, 2016
    Assignee: EBAY INC.
    Inventors: Ming Liu, Sumit Borar, Surya Kallumadi
  • Patent number: 9451087
    Abstract: Systems and methods for routing and/or servicing contacts using video analysis of one or more video streams are provided. The systems and methods are particularly applicable to a contact center.
    Type: Grant
    Filed: April 16, 2012
    Date of Patent: September 20, 2016
    Assignee: Avaya Inc.
    Inventors: Luciano Godoy Fagundes, Thomas Moran, Dhaval Desai, Joylee Kohler, Paul Roller Michaelis
  • Patent number: 9449272
    Abstract: A method of frequency discrimination associated with the Doppler effect is presented. The method includes mapping a first signal to a first plurality of frequency bins and a second signal to a second plurality of frequency bins. The first signal and the second signal corresponding to different times. The method also includes firing a first plurality of neurons based on contents of the first plurality of frequency bins and firing a second plurality of neurons based on contents of the second plurality of frequency bins.
    Type: Grant
    Filed: October 29, 2013
    Date of Patent: September 20, 2016
    Assignee: QUALCOMM INCORPORATED
    Inventors: Robert William Marshall, Somdeb Majumdar
  • Patent number: 9405975
    Abstract: Object recognition apparatus and methods useful for extracting information from sensory input. In one embodiment, the input signal is representative of an element of an image, and the extracted information is encoded in a pulsed output signal. The information is encoded in one variant as a pattern of pulse latencies relative to an occurrence of a temporal event; e.g., the appearance of a new visual frame or movement of the image. The pattern of pulses advantageously is substantially insensitive to such image parameters as size, position, and orientation, so the image identity can be readily decoded. The size, position, and rotation affect the timing of occurrence of the pattern relative to the event; hence, changing the image size or position will not change the pattern of relative pulse latencies but will shift it in time, e.g., will advance or delay its occurrence.
    Type: Grant
    Filed: June 2, 2011
    Date of Patent: August 2, 2016
    Assignee: Brain Corporation
    Inventor: Eugene M. Izhikevich
  • Patent number: 9397901
    Abstract: Methods, systems, and computer readable media for classifying application traffic at a network traffic emulation device that emulates multiple application servers are disclosed. The method may include, at a network traffic emulation device, providing a plurality of different application flows for monitoring mixed application traffic received from different client applications via a device under test on the same transport layer port. For each of the different application flows that monitor the same transport layer port, parsing a payload of received packets associated with the same session in the mixed application traffic and identifying non-matching application flows based on whether a portion of each payload matches a signature associated with the application flow.
    Type: Grant
    Filed: December 18, 2012
    Date of Patent: July 19, 2016
    Assignee: Ixia
    Inventors: Partha Majumdar, Rohan Chitradurga, Konstantin Belov
  • Patent number: 9398040
    Abstract: Disclosed herein is an Intrusion Detection System (IDS) false positive detection apparatus and method. An IDS false positive detection apparatus includes a payload extraction unit for extracting payloads by dividing each packet corresponding to an IDS detection rule into a header and a payload. A false positive payload information generation unit generates false positive payload information required to identify a false positive payload by extracting a payload of a false positive packet based on results of packet analysis received from a manager. A false positive payload determination unit transmits results of a determination of whether each payload extracted by the payload extraction unit corresponds to a false positive payload, based on the false positive payload information, to the manager.
    Type: Grant
    Filed: August 27, 2014
    Date of Patent: July 19, 2016
    Assignee: ELECTRONICS AND TELECOMMUNICATIONS RESEARCH INSTITUTE
    Inventors: Taek kyu Lee, Geun Yong Kim, Seok won Lee, Myeong Ryeol Choi, Hyung Geun Oh, KiWook Sohn
  • Patent number: 9396393
    Abstract: Video sequence processing is described with various filtering rules applied to extract dominant features for content based video sequence identification. Active regions are determined in video frames of a video sequence. Video frames are selected in response to temporal statistical characteristics of the determined active regions. A two pass analysis is used to detect a set of initial interest points and interest regions in the selected video frames to reduce the effective area of images that are refined by complex filters that provide accurate region characterizations resistant to image distortion for identification of the video frames in the video sequence. Extracted features and descriptors are robust with respect to image scaling, aspect ratio change, rotation, camera viewpoint change, illumination and contrast change, video compression/decompression artifacts and noise.
    Type: Grant
    Filed: June 6, 2014
    Date of Patent: July 19, 2016
    Assignee: Gracenote, Inc.
    Inventors: Mihailo M. Stojancic, Prashant Ramanathan, Peter Wendt, Jose Pio Pereira
  • Patent number: 9390372
    Abstract: The present invention relates to unsupervised, supervised and reinforced learning via spiking computation. The neural network comprises a plurality of neural modules. Each neural module comprises multiple digital neurons such that each neuron in a neural module has a corresponding neuron in another neural module. An interconnection network comprising a plurality of edges interconnects the plurality of neural modules. Each edge interconnects a first neural module to a second neural module, and each edge comprises a weighted synaptic connection between every neuron in the first neural module and a corresponding neuron in the second neural module.
    Type: Grant
    Filed: September 23, 2014
    Date of Patent: July 12, 2016
    Assignee: International Business Machines Corporation
    Inventor: Dharmendra S. Modha
  • Patent number: 9386931
    Abstract: Systems, methods, and devices for determining whether a medical sensor has been properly applied to a patient are provided. In one embodiment, a patient monitor having such capabilities may include a medical sensor interface and data processing circuitry. The medical sensor interface may receive physiological data from a medical sensor applied to a patient. The data processing circuitry may be capable of being trained, using a learning-based algorithm, to determine whether the received physiological data indicates that the medical sensor has been properly applied to the patient.
    Type: Grant
    Filed: March 31, 2010
    Date of Patent: July 12, 2016
    Assignee: COVIDIEN LP
    Inventor: Edward M. McKenna
  • Patent number: 9378466
    Abstract: A set S is initialized. Initially, S is empty; but, as the disclosed process is performed, items are added to it. It may contain one or more samples (e.g., items) from each class. One or more labeled samples for one or more classes may be obtained. A series of operations may be performed, iteratively, until a stopping criterion is reach to obtain the reduced set. For each class of the one or more classes, a point may be generated based on at least one sample in the class having a nearest neighbor in a set S with a different class label than the sample. The point may be added to the set S. The process may be repeated unless a stopping criterion is reached. A nearest neighbor for a submitted point in the set S may be identified and a candidate nearest neighbor may be output for the submitted point.
    Type: Grant
    Filed: December 31, 2013
    Date of Patent: June 28, 2016
    Assignee: Google Inc.
    Inventors: Sanjiv Kumar, Henry Allan Rowley, Francisco Jose Claude Faust
  • Patent number: 9367818
    Abstract: Various embodiments select features from a feature space. In one embodiment, a set of training samples and a set of test samples are received. A first centered Gram matrix of a given dimension is determined for each of a set of feature vectors that include at least one of the set of training samples and at least one of the set of test samples. A second centered Gram matrix of the given dimension is determined for a target value vector that includes target values from the set of training samples. A set of columns and rows associated with the at least one of the test samples in the second centered Gram matrix is set to 0. A subset of features is selected from a set of features based on the first and second centered Gram matrices.
    Type: Grant
    Filed: September 18, 2013
    Date of Patent: June 14, 2016
    Assignee: International Business Machines Corporation
    Inventors: Dan He, Laxmi P. Parida, Irina Rish
  • Patent number: 9369369
    Abstract: Systems and methods are disclosed herein to provide communication test systems for the testing of packet data communication devices, systems and networks. According to one aspect of the subject matter described herein, a test system containing an integrated traffic, protocol and waveform analyzer is disclosed that includes a traffic generator/analyzer and a protocol engine that triggers a waveform analyzer to analyze the signal waveform of a device under test at a predetermined point. Such a test system may offer improved capabilities such as a more selective and accurate measurement of complex signal waveforms, more automated measurements of waveforms pertaining to wireless communication data streams, and more rapid identification and measurement of waveforms corresponding to errored packets.
    Type: Grant
    Filed: March 5, 2013
    Date of Patent: June 14, 2016
    Assignee: IXIA
    Inventors: Lester Noel Stott, David Jennings Wicker, Jr., Jeffrey Dean Marker, Alexander David Rekow, Timothy Earl Bennington-Davis, Thomas Alexander
  • Patent number: 9345412
    Abstract: Systems and methods are described herein for analyzing neural response data that can be assigned to multiple classes. The systems and methods begin with a set of training data from which optimal weight factors are derived. The derived weight factors are used in a classifier which is then applied to test data from test subjects. The classifier filters out the effects of less relevant data in the test data and provides a result in the form of probabilities associated with classes for the test data.
    Type: Grant
    Filed: October 2, 2012
    Date of Patent: May 24, 2016
    Assignee: Leidos, Inc.
    Inventor: James Herbert Horne
  • Patent number: 9349093
    Abstract: Methods and systems to reduce the number of factors or variables that need to be considered in generating a function such as a prediction function. The methods and systems may involve receiving a first set of factors and a data set associated with the first set of factors, calculating an importance value of each factor in the first set of factors, and selecting, based on the importance values, a second set of factors. In some embodiments, the methods and systems also include generating the prediction function based on the data set and the second set of factors.
    Type: Grant
    Filed: August 27, 2010
    Date of Patent: May 24, 2016
    Assignee: PayPal, Inc.
    Inventors: Rogene Eichler West, Stephen Severance