Abstract: An automated method comprises receiving training data representing an initial data set including text representing at least one concept embodied by the data set, using the training data in order to generate a set of atoms, each atom comprising at least one word that represents one or more concepts of the initial data set, wherein generating a set of atoms comprises minimizing a cost function using an iterative process to identify one or more atoms.

Abstract: A method for matching a user to a conversation includes, for each of a plurality of documents, mapping the document to one or more of a plurality of topics. The document is associated with a user. For each user, the user is mapped to each of the plurality of topics that are mapped to documents with which the user is associated. One of the users is matched to a conversation according to the mappings of the users to those of the plurality of topics that are also mapped to the conversation.

Abstract: Method and system embodiments of the present invention are directed to restoration of corrupted images using spatial-domain image-processing methods that can effectively employ spatial-domain information both in order to avoid various types of artifacts and distortions produced by frequency-domain image-processing operations and to achieve computational efficiency. The various method and system embodiments of the present invention employ a family of penalty functions to constrain iterative restoration images corrupted by both deterministic corruptions, such as motion-induced blur and blurring due to optical misalignment, incorrect positioning of optical components, and defective optical components as well as essentially non-deterministic noise corruption introduced at various stages of image acquisition, image encoding, image storage, and image transmission.

Abstract: A method for pattern recognition performed by a physical computing system includes, with the physical computing system, structuring data as a bi-partite graph, a set of instance nodes within the graph representing instances within the data and a set of property nodes within the graph representing properties of the instances, edges between the instance nodes and the property nodes representing values of the properties, assigning a transition probability function to each of the instance nodes and to each of the property nodes, and applying a random walker to the graph, the random walker utilizing the transition probability functions.

Abstract: A signal processing system adapted for sparse representation of signals is provided, comprising: (i)one or more training signals; (ii) a dictionary containing signal-atoms; (iii) a representation of each training signal using a linear combination of said dictionary's signal-atoms; (iv) means for updating the representation of the training signal; (v) means for updating the dictionary one group of atoms at a time, wherein each atom update may include all representations referring to said updated atom; and (vi) means for iterating (iv) and (v) until a stopping rule is fulfilled. The system uses the K-SVD algorithm for designing dictionaries for sparse representation of signals.

Abstract: An automated method of processing computer system event logs comprises receiving event messages associated with one or more system event logs, each event message including event text, determining a set of message clusters, each comprising a template text, representative of the event messages across the one or more event logs, and assigning each event message to a message cluster of the set, according to a measure of similarity between the respective event text of the event message and the template text of the message cluster.

Abstract: A method of accurately printing a target colour is provided. The method comprises: in a first step measuring the light spectrum (106) of a first printed colour sample (104), which has been printed with a first ink coverage; and from the measured light spectrum calculating a colour discrepancy between the first printed colour sample and the target colour. The method further comprises in a second step calculating an improved ink coverage (112), and printing a second colour sample with the improved ink coverage (104). The aforementioned steps are repeated until a predetermined condition for accuracy of the target colour has been reached (110).

Abstract: A signal processing system adapted for sparse representation of signals is provided, comprising: (i) one or more training signals; (ii) a dictionary containing signal-atoms; (iii) a representation of each training signal using a linear combination of said dictionary's signal-atoms; (iv) means for updating the representation of the training signal; (v) means for updating the dictionary one group of atoms at a time, wherein each atom update may include all representations referring to said updated atom; and (vi) means for iterating (iv) and (v) until a stopping rule is fulfilled. The system uses the K-SVD algorithm for designing dictionaries for sparse representation of signals.

Abstract: Embodiments of the present invention are directed to determining the spectral vector of electromagnetic radiation reflected from, transmitted through, or emitted from a sample using a set of n intensity measurements. In general, the spectral vector has a dimension k that is greater than the number of measured intensities n. However, in many cases, the physical and chemical constraints of a system, when properly identified and modeled, effectively reduce the number of unknowns, generally the k components of the spectral vector, to an extent that allows for the spectral vector to be characterized from a relatively small number n of measured intensities.

Abstract: Event analysis methods and apparatus in which sequences (44, 46) of one or more events are identified based on event records (20) describing the events. Respective distance values (28) representing distances between ones of the sequences (44, 46) are determined. A configuration of points in a target metric space is constructed based on the distance values (28), where each of the points represents a respective one of the sequences (44, 46). A visual representation (38) of the configuration is presented on a display (34).

Abstract: An automated method comprises receiving training data representing an initial data set including text representing at least one concept embodied by the data set, using the training data in order to generate a set of atoms, each atom comprising at least one word that represents one or more concepts of the initial data set, wherein generating a set of atoms comprises minimising a cost function using an iterative process to identify one or more atoms.

Abstract: An automated method of processing computer system event logs comprises receiving event messages associated with one or more system event logs, each event message including event text, determining a set of message clusters, each comprising a template text, representative of the event messages across the one or more event logs, and assigning each event message to a message cluster of the set, according to a measure of similarity between the respective event text of the event message and the template text of the message cluster.

Abstract: A business service model includes a description of a topology of interconnections between configuration items that implement a business service. Each of the configuration items is associated with a respective vulnerability score and a respective type classification. Based on the vulnerability scores and the type classifications, the following values are determined for each of the configuration items: a respective activity level value indicating a probability of the configuration item being active in the business process, a respective vulnerability probability value indicating a probability of the configuration items being compromised and damaged in the business process, and a respective business process risk value indicating a probability of a failure of the business process resulting from damage of the configuration item. The business process is scored based on the activity level values, the vulnerability values, and the business process risk values.

Abstract: Method and system embodiments of the present invention are directed to restoration of corrupted images using spatial-domain image-processing methods that can effectively employ spatial-domain information both in order to avoid various types of artifacts and distortions produced by frequency-domain image-processing operations and to achieve computational efficiency. The various method and system embodiments of the present invention employ a family of penalty functions to constrain iterative restoration images corrupted by both deterministic corruptions, such as motion-induced blur and blurring due to optical misalignment, incorrect positioning of optical components, and defective optical components as well as essentially non-deterministic noise corruption introduced at various stages of image acquisition, image encoding, image storage, and image transmission.

Abstract: A signal processing system adapted for sparse representation of signals is provided, comprising: (i) one or more training signals; (ii) a dictionary containing signal-atoms; (iii) a representation of each training signal using a linear combination of said dictionary's signal-atoms; (iv) means for updating the representation of the training signal; (v) means for updating the dictionary one group of atoms at a time, wherein each atom update may include all representations referring to said updated atom; and (vi) means for iterating (iv) and (v) until a stopping rule is fulfilled. The system uses the K-SVD algorithm for designing dictionaries for sparse representation of signals.