Abstract: A technique for enriching sparse data for machine learning techniques such as supervised artificial neural network includes receiving the sparse data and enriching the received data around a deviation of the mean of the received data using a predetermined distribution. The technique further includes outputting the enriched data for unbiased and increased performance during the machine learning.

Abstract: A technique for machine learning, such as supervised artificial neural network learning includes receiving data and checking the dimensionality of the read data and reducing the dimensionality to enhance machine learning performance using Principal Component Analysis methodology. The technique further includes specifying the neural network architecture and initializing weights to establish a connection between read data including the reduced dimensionality and the predicted values. The technique also includes performing supervised machine learning using the specified neural network architecture, initialized weights, and the read data including the reduced dimensionality to predict values. Predicted values are then compared to a normalized system error threshold value and the initialized weights are revised based on the outcome of the comparison to generate a learnt neural network having a reduced error in weight space.

Abstract: Computer-implemented methods for allocating resources to items are provided. One or more assignment scores for each item/resource pair are determined by applying one or more application-specific strategies to each item/resource pair using game theory. A cost matrix is created by first summing the assignment scores for each item/resource pair and then multiplying each assignment score sum by an assignment cost associated with assignment a particular resource to a particular item. Finally, an assignment solution is found by applying a Hungarian method to the cost matrix.

Abstract: An automated customer support tool. A web server receives one or more queries regarding the product from a customer. A categorizer coupled to the web server categorizes the received queries based on a type of query, to determine whether the answers to the queries can be automatically communicated to the customer. An FAQ extractor extracts one or more corresponding product FAQs from respective product FAQ data bases. A key-word extractor extracts one or more key-words from the received queries and the extracted product FAQs. The key-word extractor further transforms the extracted key words to unique numerical representations. An analyzer coupled to the key-word extractor represents the transformed key-words into respective query and product FAQ vector forms. The analyzer further applies a convolution algorithm to each of the query vector forms, with each of the product FAQ vector forms separately and obtains one or more appropriate answers to the queries.

Abstract: Solving a Traveling Salesman Problem (TSP) by selecting a set of locations to visit, selecting a starting point and an ending point from the set of locations, applying a search method to the set of locations, and providing a route as a solution to the TSP, where the search method is a combinatoric approach to a genetic search and the search method simultaneously minimizes distance and time. The route starts and ends in different locations and completes in polynomial time, such as O(n+k), where k is a constant. The solution to the TSP has many applications, including finding distribution chains to satisfy customer demand for an Internet enterprise.

Abstract: A technique for clustering input data includes receiving data and reading a sample of the received data having a predetermined window length. The technique further includes checking the read sample of data for uncertainty and/or robustness and determining the clustering approach to be used to cluster the received data based on the outcome of the checking.

Abstract: Multiple words in a text are transformed to unique numerical representations for text mining applications. A web server receives the text, including multiple words in a natural language. A key-word extractor extracts one or more key-words from the received words. A morphologizer morphologizes the extracted key-words based on similarities of fundamental characteristics in the extracted key-words. An analyzer transforms each of the morphologized words to a unique numerical representation such that the transformed unique numerical representation does not result in multiple similar numerical representations.

Abstract: Data structures, systems, and methods are aspects of pattern recognition using observable operator models (OOMs). OOMs are more efficient than Hidden Markov Models (HMMs). A data structure for an OOM has characteristic events, an initial distribution vector, a probability transition matrix, an occurrence count matrix, and at least one observable operator. System applications include computer systems, cellular phones, wearable computers, home control systems, fire safety or security systems, PDAs, and flight systems. A method of pattern recognition comprises training OOMs, receiving unknown input, computing matching probabilities, selecting the maximum probability, and displaying the match. A method of speech recognition comprises sampling a first input stream, performing a spectral analysis, clustering, training OOMs, and recognizing speech using the OOMs.

Abstract: A system and a method for an automated intelligent information mining includes receiving product-related queries and respective product-related information from various text sources; extracting multiple key-phrases from the product-related information and received queries; generating two or more layers of contextual relation maps by mapping the extracted key-phrases to two-dimensional maps using a self organizing map, and a technique including a combination of Hessian matrix and Perturbation technique to enhance the learning process and to categorize the extracted key-phrases based on a contextual meaning. Further, the technique includes forming word clusters and constructing corresponding key phrase frequency histograms for each of the generated contextual relation maps.

Abstract: A scheduling system reschedules based on the affect such rescheduling has on cascading costs. The costs for an airline set of schedules include factors such as the affect on passengers, the affect on profit and other factors. Cascading costs include the cost of further disruptions to schedules as a result of proposed solutions to a disruption. The further disruptions are referred to as cascading disruptions since they cascade from the original disruption or solutions to the disruption. Disruptions are entered into a computer program and ranked by a common denominator such as cost by simulating the effects of the disruptions on the actual servicing of the schedules. Disruptions are prioritized based on their impact on the system if left unattended. Each disruption is then considered individually based on their rank, with the cascaded cost of proposed solutions calculated.

Abstract: A system and a method for an automated intelligent information mining includes receiving product-related queries and respective product-related information from various text sources; extracting multiple key-phrases from the product-related information and received queries; generating two or more layers of contextual relation maps by mapping the extracted key-phrases to two-dimensional maps using a self organizing map, and a technique including a combination of Hessian matrix and Perturbation technique to enhance the learning process and to categorize the extracted key-phrases based on a contextual meaning. Further, the technique includes forming word clusters and constructing corresponding key phrase frequency histograms for each of the generated contextual relation maps.

Abstract: Data structures, systems, and methods are aspects of pattern recognition using observable operator models (OOMs). OOMs are more efficient than Hidden Markov Models (HMMs). A data structure for an OOM has characteristic events, an initial distribution vector, a probability transition matrix, an occurrence count matrix, and at least one observable operator. System applications include computer systems, cellular phones, wearable computers, home control systems, fire safety or security systems, PDAs, and flight systems. A method of pattern recognition comprises training OOMs, receiving unknown input, computing matching probabilities, selecting the maximum probability, and displaying the match. A method of speech recognition comprises sampling a first input stream, performing a spectral analysis, clustering, training OOMs, and recognizing speech using the OOMs.

Abstract: Solving a Traveling Salesman Problem (TSP) by selecting a set of locations to visit, selecting a starting point and an ending point from the set of locations, applying a search method to the set of locations, and providing a route as a solution to the TSP, where the search method is a combinatoric approach to a genetic search and the search method simultaneously minimizes distance and time. The route starts and ends in different locations and completes in polynomial time, such as O(n+k), where k is a constant. The solution to the TSP has many applications, including finding distribution chains to satisfy customer demand for an Internet enterprise.

Abstract: A technique for machine learning, such as supervised artificial neural network learning includes receiving data and checking the dimensionality of the read data and reducing the dimensionality to enhance machine learning performance using Principal Component Analysis methodology. The technique further includes specifying the neural network architecture and initializing weights to establish a connection between read data including the reduced dimensionality and the predicted values. The technique also includes performing supervised machine learning using the specified neural network architecture, initialized weights, and the read data including the reduced dimensionality to predict values. Predicted values are then compared to a normalized system error threshold value and the initialized weights are revised based on the outcome of the comparison to generate a learnt neural network having a reduced error in weight space.

Abstract: A technique for clustering input data includes receiving data and reading a sample of the received data having a predetermined window length. The technique further includes checking the read sample of data for uncertainty and/or robustness and determining the clustering approach to be used to cluster the received data based on the outcome of the checking.

Abstract: Systems and methods are provided through which the quantity of one or more items in a customer order are reduced in reference to a function of inverse probability of vendor profit and in reference to a reasonable margin of a target time predetermined by the customer, when the customer order cannot be produced within the margin. The reduced item quantities update the corresponding items in the customer order, and the items in the customer order are produced accordingly, such that the objective of the business goal is met or not sacrificed.

Abstract: An automated customer support tool. A web server receives one or more queries regarding the product from a customer. A categorizer coupled to the web server categorizes the received queries based on a type of query, to determine whether the answers to the queries can be automatically communicated to the customer. An FAQ extractor extracts one or more corresponding product FAQs from respective product FAQ data bases. A key-word extractor extracts one or more key-words from the received queries and the extracted product FAQs. The key-word extractor further transforms the extracted key words to unique numerical representations. An analyzer coupled to the key-word extractor represents the transformed key-words into respective query and product FAQ vector forms. The analyzer further applies a convolution algorithm to each of the query vector forms, with each of the product FAQ vector forms separately and obtains one or more appropriate answers to the queries.

Abstract: Multiple words in a text are transformed to unique numerical representations for text mining applications. A web server receives the text, including multiple words in a natural language. A key-word extractor extracts one or more key-words from the received words. A morphologizer morphologizes the extracted key-words based on similarities of fundamental characteristics in the extracted key-words. An analyzer transforms each of the morphologized words to a unique numerical representation such that the transformed unique numerical representation does not result in multiple similar numerical representations, to avoid ambiguous prediction of meaning of the translated words in the received text.

Abstract: A technique for enriching sparse data for machine learning techniques such as supervised artificial neural network includes receiving the sparse data and enriching the received data around a deviation of the mean of the received data using a predetermined distribution. The technique further includes outputting the enriched data for unbiased and increased performance during the machine learning.

Abstract: In a first embodiment, the present invention provides computer-implemented methods for allocating resources to items are provided. Methods according to the present invention first determine one or more assignment scores for each item/resource pair by applying one or more application-specific strategies to each item/resource pair using game theory. A cost matrix is created by first summing the assignment scores for each item/resource pair and then multiplying each assignment score sum by an assignment cost associated with assignment a particular resource to a particular item. Finally, an assignment solution is found by applying the Hungarian method to the cost matrix. In a second embodiment, the present invention provides a computer-readable medium having computer-executable instructions for performing a methods of the present invention.