Abstract: Digital technology utilizing a cognitive model based on human naturalistic decision-making processes, including pattern recognition and episodic memory, can reduce the dependency of human-machine interactions on the abilities of a human user and can enable a machine to more closely emulate human-like responses. Such a cognitive model can enable digital technology to use cognitive capacities fundamental to human-like communication and cooperation to interact with humans.

Abstract: Systems and methods are provided for configuring software. The software may be configured with wizard-like configuration levels, including configuration levels that may overlay previous ones. In one embodiment, a computer-implemented method is provided for configuring software. The method may include prompting a user to select a scenario for configuring the software, receiving, as input, a selected scenario from the user and further prompting the user to enter configuration data based on the selected business scenario. The method may also include applying at least one template to generate at least one configuration statement for the software, wherein the at least one template comprises a set of rules for generating the configuration statement based on the configuration data entered by the user.

Abstract: Methods and apparatus for providing a comprehensive decision support system to include predictions, recommendations with consequences and optimal follow-up actions in specific situations are described. Data is obtained from multiple disparate data sources, depending on the information deemed necessary for the situation being modeled. Some embodiments perform complex systems modeling including performing massive correlative analyses of the data obtained from the multiple disparate data sources with current situational data obtained regarding the situation for which the decision support process is being utilized. The decision support system provides a prediction or predictions and a recommendation or a choice of recommendations based on the correlative analysis and/or other analyses. In some embodiments the decision support system provides possible consequences that could result from a recommendation. In other embodiments the decision support system provides a list of tasks for acting upon a recommendation.

Abstract: A method and apparatus for providing a human expert to answer a question from an agent in real time. The method includes the steps of identifying a plurality of human experts based upon a subject matter of the question, determining an availability of at least some of the plurality of identified experts and selecting an expert of the plurality of experts to answer the question in real-time based upon the subject matter and availability of the selected expert.

Abstract: A method of designing a knowledge repository includes receiving, in a computer system, a first input to form a configuration template for a knowledge repository. The configuration template describes a class of knowledge entities to be included in the knowledge repository. There is received a second input identifying a document template to be used in displaying instances of the class of knowledge entities. The document template is selected from a predefined group of document templates. The document template is associated with the configuration template in the computer system. The thus designed knowledge repository may be provided with knowledge entities accordingly. Moreover, the created knowledge repository may be used to retrieve specific knowledge.

Abstract: An action evaluation apparatus evaluates a user's action corresponding to an action desired by a system, and determines the ongoing of a story using the evaluation result. The action evaluation apparatus includes a recognizer for recognizing the user's action through a plurality of modals, and an action evaluator for evaluating the user's action based on the recognition result of the respective modals obtained by the recognizer.

Abstract: Methods, systems, and machine readable media storing machine-readable instructions for enhanced classification of marginal instances are described. In one aspect, a classification system includes a cascade of forward classification stages and a marginal instance classifier. Each of the forward classification stages is configured to classify an instance into one of a rejected class, an unconditionally accepted class, and a conditionally accepted class. The cascade is configured to reject each instance classified into the rejected class by at least one of the forward classification stages, accept each instance classified into the unconditionally accepted class by all of the forward classification stages, and conditionally accept each instance classified into the conditionally accepted class by at least one of the forward classification stages. The marginal instance classifier is configured to classify each instance conditionally accepted by the cascade into one of a rejected class and an accepted class.

Abstract: A system, apparatus, and method to dynamically allocate resources provide full connectivity between elements. A common decision execution capability (CDEC) forms a centralized command and control function able to control the elements. The CDEC is adapted to generate virtual chains, each virtual chain a combination of the elements, to achieve an objective. The CDEC can select a virtual chain and either perform the selected virtual chain or generate a plan associated with the virtual chain.

Abstract: Systems and methods for playbook automation are described. In one aspect, a user selects operational knowledge associated with an application. The user associates one or more playbook-based tasks, playbook-based views, or playbook-based links with the operational knowledge. The playbook-based tasks, playbook-based views, and playbook-based links are for diagnosing, resolving, and/or verifying a problem associated with the application. The systems and methods generate an integrated management pack from the associated operational knowledge and at least one of the one or more playbook-based tasks, playbook-based views, or playbook-based links.

Abstract: Embodiments of the invention provide methods and apparatuses for classifying electronic documents (e.g., electronic communications) as either spam electronic documents or legitimate electronic documents. In accordance with one embodiment of the invention, each of a plurality of electronic communications is reduced to a corresponding multidimensional vector based on a multi-dimensional vector space. The multi-dimensional vectors represent corresponding electronic documents that have been classified as at least one type of electronic documents. Subsequent electronic documents to be classified are reduced to a corresponding multi-dimensional vector inserted into the multi-dimensional vector space. The electronic documents corresponding to an inserted multi-dimensional vector are classified based upon the proximity of the inserted multi-dimensional vector to at least one previously classified multi-dimensional vectors of the multi-dimensional vector space.

Abstract: The present invention relates to a signal processing device which learns operations made by a user without the user knowing, and which can perform processing optimal to the user based on the learning results. At a learning unit 2, operating signals supplied according to user operations are monitored, and judgment is made whether or not these can be used for learning. In the event that the operating signals are learning operating signals which can be used for learning, the learning unit 2 learns a correction norm which is the norm for correcting input signals, based on the learning operating signals. On the other hand, at a correcting unit 1, post-correction signals, wherein the input signals are corrected based on the correction norm obtained by learning, are output as output signals. The present invention can be applied to an NR (Noise Reduction) circuit which removes noise.

Abstract: A method of traversing a workflow is described. The workflow comprises a plurality of interconnected question and answer nodes, each question node being linked to a respective stored pre-determined question and each answer node being linked to a respective stored answer. The method comprises: determining the type of the current node of the workflow and, if the current node is a question node, accessing a data store, ascertaining whether the data store contains a previously determined answer to the question linked to the current question node and, a) if the data store does contain such a previously determined answer, retrieving the answer from the data store and using the retrieved answer to determine the next node of the workflow to be traversed; and b) if the data store does not contain such a previously determined answer, presenting the question linked to the current question node to a user, and using answer data obtained from the user to determine the next node of the workflow to be traversed.

Abstract: Methods and apparatus for providing a comprehensive decision support system to include predictions, recommendations with consequences and optimal follow-up actions in specific situations are described. Data is obtained from multiple disparate data sources, depending on the information deemed necessary for the situation being modeled. Some embodiments perform complex systems modeling including performing massive correlative analyses of the data obtained from the multiple disparate data sources with current situational data obtained regarding the situation for which the decision support process is being utilized. The decision support system provides a prediction or predictions and a recommendation or a choice of recommendations based on the correlative analysis and/or other analyses. In some embodiments the decision support system provides possible consequences that could result from a recommendation. In other embodiments the decision support system provides a list of tasks for acting upon a recommendation.

Abstract: The present invention tests computer systems for defects that are able to cause performance and functional problems. An agent application is first installed on a user's system. The agent application then retrieves problem data from the expert system library pertinent to the user's operating environment. The agent application tests the user's system for each problem description within a downloaded data structure from the expert system library. If a problem is discovered, a script attempts to remediate the problem and/or notifies the user, describing the issue. If an unknown problem arises, a feedback mechanism sends a notification to the expert system library that a problem was discovered. The feedback is then utilized to increase the total knowledge in the expert system library. Then when another user with a similar or the same setup is tested, the agent application will recognize the issue and will take the appropriate remedial actions.

Abstract: A terminological system automates the integration of terminological information into a built-in knowledge base. Input terminology information, which includes input terms and information that specifies relationships among at least two of the input terms, is input to the terminological system. The terminological system parses the input terminology information to generate a logical structure that depicts relationships among the input terms in a format compatible with the built-in knowledge base. Either an independent ontology, comprising the logical structure, is generated, or the knowledge base is extended by logically coupling the logical structure to a node that matches the input term. The terminological system also resolves conflicts if an input term that matches a terminological node in the knowledge base connotes a different meaning than the terminological node.

Abstract: A behavior is associated with a syndicated feed. The syndicated feed is associated with a content page, and provided from a server over a network. The syndicated feed may be an RSS feed or some other type of syndicated feed associated with a content page. The feed is accessed by an application and then associated with the behavior. The feed with the associated behavior is then provided to a user through an interface. A binding can be used to associate a behavior with a feed. The binding may include resources, scripts and other bindings, all of which may be distributed over a system or network.

Abstract: A system and method for organizing knowledge in such a way that humans can find knowledge, learn from it, and add to it as needed is disclosed. The exemplary system has four components: a knowledge base, a learning model and an associated tutor, a set of user tools, and a backend system. The invention also preferably comprises a set of application programming interfaces (APIs) that allow these components to work together, so that other people can create their own versions of each of the components. In the knowledge web a community of people with knowledge to share put knowledge in the database using the user tools. The knowledge may be in the form of documents or other media, or it may be a descriptor of a book or other physical source. Each piece of knowledge is associated with various types of meta-knowledge about what the knowledge is for, what form it is in, and so on.

Abstract: Disclosed herein is a programming tool stored on a computer-readable medium and adapted for implementation by a computer for designing an artificial neural network. The programming tool includes a network configuration module to provide a first display interface to support configuration of the artificial neural network, and a pattern data module to provide a second display interface to support establishment and modification of first and second pattern data sets for training and testing the artificial neural network, respectively.

Abstract: A count of cases in a target set classified in a class is computed using a classifier. In the method, one or more measures of behavior of the classifier are determined for a plurality of classification thresholds and at least one classification threshold for the classifier is selected based on the one or more measures of behavior. In addition, a score for a plurality of the cases in the target set using the classifier is computed and the count of the cases is computed based on the scores, the selected at least one classification threshold, and the one or more measures of behavior.

Abstract: A method, device and computer program product that monitor situations. The method includes: processing information indicative of an occurrence of events to detect, in response to a current rule set, an occurrence of at least one situation that belongs to a situation set; wherein at least one rule sub-set out of the rule set defines at least one relationship between events that should occur within a lifespan; receiving a request to update a first rule; updating the first rule to provide a first updated rule while continuing to detect an occurrence of situations not affected by the update of the first rule.