Abstract: A method, system, and process for classifying a dataset using a decision tree is discussed. The decision tree may be stored in a novel format on a distributed architecture, and the dataset classification may be performed in parallel on the distributed architecture.

Abstract: Aspects of the invention collect data or information from a dynamic and adaptive target group selection. Sets of experts are selected. Responses to a survey are used to expand the set of experts by adding experts noted in answer referrals. The survey is updated by removing satisfied questions, and expert selections, answers analysis and survey updating and resending steps are iteratively repeated until each question is satisfied, wherein the expert sets may be dynamically revised in each iteration.

Abstract: A computationally efficient method and apparatus is disclosed for machine learning an unknown, n-dimensional ?-margin halfspace to accuracy 1?? in the presence of malicious noise, when the noise rate is allowed to be as high as ? ? ( ?? ? log ? ( 1 / ? ) ) . A smooth boosting module repeatedly calls a weak learner module that generates candidate classifiers based on a majority vote of randomly-generated classifiers.

Abstract: The present description is directed to a technique to store one or more tasks, each task including one or more knowledge actions (KAs) that includes an action of one of a plurality of KA types being performed, determine a degree of similarity between each of one or more of the stored tasks and a current task, identify one of the stored tasks that most closely matches the current task based on the degree of similarity for each of the one or more stored tasks, identify one or more information objects that were open for the identified stored task, identify one or more information objects that are currently open for the current task, determine an additional information object that was open for the identified stored task but is not currently open for the current task, and provide a recommendation to a user to open the additional information object.

Abstract: A method for forming and using a composite production rule may include compiling, by a computer system, a decision table or a decision tree to generate a composite production rule. The method may also include generating the composite production rule and selecting, by the computer system, an algorithm for compiling the composite production rule. The method may additionally include compiling, by the computer system, the composite production rule into an executable program based on pattern matching of the selected algorithm. The method may further include executing, by the computer system, the composite production rule to provide an output based on the composite production rule.

Abstract: Embodiments of the present invention characterizing the uncertainty of the orbital state of an Earth-orbiting space object hereof using a Gauss von Mises probability density function defined on the n+1 dimensional cylindrical manifold n×. Additionally, embodiments of the present invention can include transforming a Gauss von Mises distribution under a diffeomorphism and approximating the output as a Gauss von Mises distribution. Embodiments of the present invention can also include fusing a prior state represented by a Gauss von Mises distribution with an update report, wherein the update can be either another Gauss von Mises distribution of the same dimension as the prior or an observation related to the prior by a stochastic measurement model. A Gauss von Mises distribution can be calculated from a plurality of reports, wherein the reports are either Gauss von Mises distributions or observations related to the state space by a stochastic measurement model.

Abstract: Systems and methods (e.g., utilities) for use in providing automated, lightweight collection of online, open source data which may be content-based to reduce website source bias. In one aspect, a utility is disclosed for use in extracting content of interest from at least one website or other online data source (e.g., where the extracted content can be used in a subsequent search query). In other aspects, utilities are disclosed that are operable to perform various types of analyzes on such extracted content and present graphical representations of such analyzes on a display of a client device.

Abstract: Embodiments are directed towards clustering cookies for identifying unique mobile devices for associating activities over a network with a given mobile device. The cookies are clustered based on a Bayes Factor similarity model that is trained from cookie features of known mobile devices. The clusters may be used to determine the number of unique mobile devices that access a website. The clusters may also be used to provide targeted content to each unique mobile device.

Abstract: A method for forming and using a composite production rule may include compiling, by a computer system, a decision table or a decision tree to generate a composite production rule. The method may also include generating the composite production rule and selecting, by the computer system, an algorithm for compiling the composite production rule. The method may additionally include compiling, by the computer system, the composite production rule into an executable program based on pattern matching of the selected algorithm. The method may further include executing, by the computer system, the composite production rule to provide an output based on the composite production rule.

Abstract: Response performance of a new type of transaction processing based on a new program, when the new type of transaction processing is executed by a computer of an information processing system, is predicted. The information processing system executes transaction processing requiring access to a database system in response to an external processing request. Upon prediction of response performance each value of a plurality of parameters related to operation that affects response speed of a database system assumed in the new type of transaction processing is acquired. The acquired each value of the plurality of parameters is inputted to a predetermined evaluation formula to calculate an evaluation value related to response performance of the new type of transaction processing.

Abstract: Feature and deployment recommendation systems and methods for content management systems comprises a vendor-site deployment and one or more customer-site deployments. The vendor-site deployment is configured to recommend an adaptation of any of the one or more customer-site deployments. The vendor-site deployment including a customer configuration repository adapted to store content usage metrics received from one or more customer-site deployments, an analytical engine configured to generate content usage patterns based on the stored content usage metrics, and a recommendation engine configured to recommend the adaptation.

Abstract: Embodiments of the present invention characterizing the uncertainty of the orbital state of an Earth-orbiting space object hereof using a Gauss von Mises probability density function defined on the n+1 dimensional cylindrical manifold n×. Additionally, embodiments of the present invention can include transforming a Gauss von Mises distribution under a diffeomorphism and approximating the output as a Gauss von Mises distribution. Embodiments of the present invention can also include fusing a prior state represented by a Gauss von Mises distribution with an update report, wherein the update can be either another Gauss von Mises distribution of the same dimension as the prior or an observation related to the prior by a stochastic measurement model. A Gauss von Mises distribution can be calculated from a plurality of reports, wherein the reports are either Gauss von Mises distributions or observations related to the state space by a stochastic measurement model.

Abstract: Embodiments of the present invention are directed to reducing the computational cost of propagating the uncertainty in the state of an object, which requires the numerical solution of an initial value problem (IVP) ensemble, by means of a computer. The present invention uses an implicit-Runge-Kutta-based (IRK) method that exploits the proximity of the initial conditions within the IVP ensemble. More specifically, one of the IVPs can be solved over a given time span consisting of one or more time steps with an IRK method, and the resulting trajectory can be recorded. The system of equations that arises on each time step of the IRK method can be solved using various iterative methods. The remaining IVPs can be solved over the same timespan using the same IRK method and time steps, together with the recorded trajectory from the first propagation.

Abstract: Aspects of the disclosure provide for detection of spam business listings. Aspects operate to identify business listing characteristics in trusted sources and untrusted sources. As untrusted sources are likely to contain more spam, characteristics that are present in untrusted sources but not present in trusted sources are typically indicative of spam listings, and vice versa. Thus, statistical analysis of the frequency of characteristics within each source may be used to identify common characteristics of spam listings. These characteristics may further be analyzed in specific listing contexts, as different listing contexts (e.g., different types of businesses) typically use different terms and vocabularies, such that terms that are indicative of spam in one context may not be indicative of spam in another. Various methods for leveraging this context-specific statistical information to improve spam detection operations are disclosed.

Abstract: A novel domain adaption/transfer learning method applied to the problem of classifying abbreviated documents, e.g., short text messages, instant messages, tweets. The method uses a large number of multi-labeled examples (source domain) to improve the learning on the partial observations (target domain). Specifically, a hidden, higher-level abstraction space is learned that is meaningful for the multi-labeled examples in the source domain. This is done by simultaneously minimizing the document reconstruction error and the error in a classification model learned in the hidden space using known labels from the source domain. The partial observations in the target space are then mapped to the same hidden space, and classified into the label space determined by the source domain.

Abstract: A novel domain adaption/transfer learning method applied to the problem of classifying abbreviated documents, e.g., short text messages, instant messages, tweets. The method uses a large number of multi-labeled examples (source domain) to improve the learning on the partial observations (target domain). Specifically, a hidden, higher-level abstraction space is learned that is meaningful for the multi-labeled examples in the source domain. This is done by simultaneously minimizing the document reconstruction error and the error in a classification model learned in the hidden space using known labels from the source domain. The partial observations in the target space are then mapped to the same hidden space, and classified into the label space determined by the source domain.

Abstract: Certain aspects of the present disclosure present a technique for primary visual cortex (V1) cell training and operation. The present disclosure proposes a model structure of V1 cells and retinal ganglion cells (RGCs), and an efficient method of training connectivity between these two layers of cells such that the proposed method leads to an autonomous formation of feature detectors within the V1 layer. The proposed approach enables a hardware-efficient and biological-plausible implementation of image recognition and motion detection systems.

Abstract: A personal taste assessment system receives a person's ratings for various rated items, such as wines or other beverages, and it develops a preference model for the person based on the received ratings. When developing the preference model, the system accesses a database of candidate items, wherein each candidate item is associated with at least one trait value. The system retrieves from the database a value for at least one trait for each of the rated items, identifies at least one pattern of dependency between at least one of the retrieved trait values and at least one of the received ratings, and determines a preference model based on the at least one pattern of dependency. The system may use the preference model to develop a preference profile for the person, make recommendations, and/or make predictions as to items that the person and/or a group that includes the person may enjoy.

Abstract: Aspects of the invention collect data or information from a dynamic and adaptive target group selection. Sets of experts are selected. Responses to a survey are used to expand the set of experts by adding experts noted in answer referrals. The survey is updated by removing satisfied questions, and expert selections, answers analysis and survey updating and resending steps are iteratively repeated until each question is satisfied, wherein the expert sets may be dynamically revised in each iteration.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for determining query suggestions from multiple correlation measures. In one aspect, a method includes receiving a first query and second queries, each of the first and second queries including one or more terms; for each second query and a linear model, receiving correlation scores measuring the correlation between the first query and the respective second query, each correlation score received from a respective correlation process, and each respective correlation process being different from the other respective correlation processes, and applying the linear model to the plurality of correlation scores to determine a combined correlation score that quantifies a combined correlation between the first query and the respective second query based on the plurality of correlation scores. The second queries are ranked in an order according to their respective combined correlations scores.