Abstract: A method (and structure) for processing an inductive learning model for a dataset of examples, includes dividing the dataset of examples into a plurality of subsets of data and generating, using a processor on a computer, a learning model using examples of a first subset of data of the plurality of subsets of data. The learning model being generated for the first subset comprises an initial stage of an evolving aggregate learning model (ensemble model) for an entirety of the dataset, the ensemble model thereby providing an evolving estimated learning model for the entirety of the dataset if all the subsets were to be processed. The generating of the learning model using data from a subset includes calculating a value for at least one parameter that provides an objective indication of an adequacy of a current stage of the ensemble model.

Abstract: Given a set of training examples—with known inputs and outputs—and a set of working examples—with known inputs but unknown outputs—train a classifier on the training examples. For each possible assignment of outputs to the working examples, determine whether assigning the outputs to the working examples results in a training and working set that are likely to have resulted from the same distribution. If so, then add the assignment to a likely set of assignments. For each assignment in the likely set, compute the error of the trained classifier on the assignment. Use the maximum of these errors as a probably approximately correct error bound for the classifier.

Abstract: Techniques for using motion recognizers are described. The motion recognizers are created or generated in advance by trained users. The motion recognizers are then loaded into a processing unit that receives motion signals from one or more motion sensitive devices being manipulated by one or more end users to control one or more objects in a virtual environment. Depending on implementation, the virtual environment may represent a remote scene or a video game, where objects in the virtual environment can be controlled by the users to perform desired actions or moves.

Abstract: A method for generating a storage policy, the method includes: receiving a storage system target function; and generating, by a machine learning entity, the storage policy in response to: (a) a set of file-related storage operation requests, (b) a state of the storage system before responding to the set of file-related storage operation requests, and (c) the storage system target function. A method for evaluating a storage policy, the method includes: simulating an application of the storage policy by the storage system during a first period, in response to a set of file-related storage operation requests that was provided to the storage system during the first period, to provide a simulation result; wherein the first period starts before the simulating.

Abstract: An event detection method is disclosed. At least one most adaptable life cycle model is generated according to at least one historical event data, at least one nutrition growing function, and at least one firing point rule. Event data is received and a strength value thereof is calculated according to a life cycle model corresponding to the event data. It is determined whether an event firing point is achieved according to the strength value variation. If the event firing point is achieved, an event corresponding to the event data is sent. The event detection method enhances the ability of event tracking and development so event firing is more accurate to fit real event occurring situations, realize event evolution, and filter false alarms.

Abstract: There is provided with a method, including: optimizing a first objective function defined by using an explanatory variable belonging an attribute in each sample, a target variable in each sample, and a first conversion parameter to find a value of the first conversion parameter; generating by using the first conversion parameter corresponding to the attribute a conversion function for converting an explanatory variable belonging to the attribute to an intermediate variable with certain range; optimizing a second objective function defined by using a plurality of the intermediate variables corresponding to the plurality of variables in each sample, the target variable in each sample, and a second conversion parameter to find a value of the second conversion parameter; and generating by using the second conversion parameter a probabilistic model for calculating from a plurality of intermediate variables a probability that a predetermined event occurs or does not occur.

Abstract: A computer processing device receives computer readable data to derive computer executable rules for mining and constructing situation categories. The received data is transformed into a predetermined standard format if the received data is not already in the predetermined standard format. The predetermined standard formatted data is parsed, and an outer, iterative loop is performed until at least one predetermined stopping criterion is met. An inner iterative loop is performed within the outer iterative loop until all desired subsets of data are processed. During the inner iterative loop, selected subsets of data are labeled with labels associated with corresponding previously labeled subsets of data. New computer executable rules are generated for mining and constructing situation categories from the labeled subsets of data. Keyword list classifiers are transformed using the stored labeled subsets of data.

Abstract: A risk assessment method and system. The method includes receiving by an inference engine, first sensor cohort data associated with a first cohort located within a pre/post security area within an airport. The inference engine receives first group technology inferences associated with the first cohort. The inference engine generates first risk cohort inferences based on the first group technology inferences and the first sensor cohort data. The inference engine receives inference data comprising inferences associated with the first cohort. The inference engine generates second inference data comprising a second plurality of inferences associated with the first cohort. The second inference data is based on the inference data and the first risk cohort inferences. The inference engine generates a first associated risk level score for the first cohort. The computing system stores the second inference data and the first associated risk level score.

Abstract: A risk assessment method and system. The method includes receiving by an inference engine, first sensor cohort data associated with a first cohort located within a security perimeter area surrounding an airport. The inference engine receives first group technology inferences associated with the first cohort. The inference engine generates first risk cohort inferences based on the first group technology inferences and the first sensor cohort data. The inference engine receives first inference data comprising a first plurality of inferences associated with the first cohort. The inference engine generates second inference data comprising a second plurality of inferences associated with the first cohort. The second inference data is based on the first inference data and the first risk cohort inferences. The inference engine generates a first associated risk level score for the first cohort. The computing system stores the second inference data and the first associated risk level score.

Abstract: A set of decision activities can be provided to a business process management and design software tool. The tool can be used to define various processes that contain one or more activities linked by transitions. During process design time, a business analyst can identify process-relevant data of the process to be presented to the end users. The relevant data can be rendered to the end users at runtime, along with a set of possible options for the activities in a process instance. Upon receiving a decision of the user, the process instance can follow the selected transition to a different activity. Over time, the system can capture the decisions taken by users and enter the data associated with these decisions into a data classifier. In doing so, the software tool can generate suggestions, provide probabilities, automate and distribute workloads, flag abnormalities and deviations and generally improve process design and management.

Abstract: A model selection method is provided for choosing the number of clusters, or more generally the parameters of a clustering algorithm. The algorithm is based on comparing the similarity between pairs of clustering runs on sub-samples or other perturbations of the data. High pairwise similarities show that the clustering represents a stable pattern in the data. The method is applicable to any clustering algorithm, and can also detect lack of structure. We show results on artificial and real data using a hierarchical clustering algorithm.

Abstract: It is difficult to select parameter values for constraint programming problem solvers which will yield good performance. Automated tuning of such problem solvers on a per problem instance basis may be used and this involves learning a function for predicting the runtime of a problem solver depending on parameter values of the problem solver and features of the problem instance being solved. However, it takes time for such prediction functions to be learnt, either during operation of a problem solver or offline, using specified examples. To address this, information about such a prediction function is shared between two or more problem solvers to improve performance. A sharing system may be used to receive prediction function information and send this to problem solvers.

Abstract: A policy integrity validating unit validates whether a system model can achieve an objective set up in a policy without violating a constraint set up in the policy when a system model is operated according to a procedure set up in the policy. When the policy integrity validating unit determines that there is no integrity between a policy created by a system administrator and the system model, an alternate policy creating unit creates an alternate policy that replaces the policy created by the system administrator. Upon receipt of the alternate policy, the policy integrity validating unit validates integrity between the alternate policy and the system model. When the policy integrity validating unit determines that there is integrity between the alternate policy and the system model, a policy creating apparatus presents the alternate policy to the system administrator.

Abstract: Methods and computer apparatus are disclosed for deriving first order logic language rules for use in analyzing new data to identify anomalies in the new data. Historical data is formatted to be in a form suitable for application of a machine-learning algorithm thereto. The machine-learning algorithm is applied to the formatted historical data to generate a set of first order logic language rules that cover the formatted historical data. The set of first order logic language rules are analyzed to obtain a subset of said set of first order logic language rules which can be used as first order logic language rules in analyzing new data.

Abstract: Methods and systems capable of creating a component for determining which subset of a set of logical expressions are true with relatively few evaluations of the primitives that, together with any standard logical connectives, make up the logical expressions.

Abstract: The first fast solution to the problem of tracking wavelet representations of one-dimensional and multi-dimensional data streams based on a stream synopsis, the Group-Count Sketch (GCS) is provided. By imposing a hierarchical structure of groups over the data and applying the GCS, our algorithms can quickly recover the most important wavelet coefficients with guaranteed accuracy. A tradeoff between query time and update time is established, by varying the hierarchical structure of groups, allowing the right balance to be found for specific data streams. Experimental analysis confirmed this tradeoff, and showed that all the methods significantly outperformed previously known methods in terms of both update time and query time, while maintaining a high level of accuracy.

Abstract: Embodiments for monitoring one or more utilities consumed within a monitored area including detecting an amount of a first utility consumed by a load associated with a device and an amount of a second utility consumed by a load associated with the device. The identity of the device is determined based on the amount of the first and second utility consumed. A utility monitoring method is disclosed that includes measuring an amount of a utility consumed by a first device at a first time and an amount of a utility consumed by a second device at a second time. The identity of the first device is determined based on the measured consumption of the first and second devices at the first and second time. In certain implementations, the electricity meter uses an effective variance analysis of the conductance waveform on an electrical circuit to identify specific appliances.

Abstract: An instant messaging system for securing chat sessions containing confidential subject matter during instant messaging from external entities, the system comprising: a rules engine module configured to continuously build and update a set of rules used to determine when a chat session between a first chat party and a second chat party during an instant messaging session contains confidential or uniquely classified subject matter by loading at least one default rules or previous session rules and enabling a user to provide user-defined rules until the instant message session is ended; and an instant messaging module configured to secure the contents in the chat session based on pre-defined user settings, the contents in the chat session being secured when the contents in the chat session is determined to contain confidential subject matter derived from the set of rules, the chat session is further secured by limiting the exposure of the contents in the chat session to third parties when the chat session is dete

Abstract: Disclosed are computer systems with intelligent or autonomous decision systems which include means for determining relevancy, i.e., the threats to and opportunities of the autonomous decision system. Also disclosed are such autonomous decision systems using an efficient ontology to interact sociably with humans, including the use of natural languages and bonding. The desired “whether concrete is included in abstract” computation system is enhanced by the ontology system using categorizing of natural objects using as primitives a set of self tendencies suitable, when hierarchically assigned to objects, to do incremental simulation of “future” situations (including such objects) from a presented situation. Using such primitives and computation system, planning, learning, languaging, etc., are efficiently accomplished.

Abstract: Computer-implemented methods, systems and apparatus, including computer program apparatus, provide techniques for designing a set of experiments to be performed with a set of resources. A plurality of experimental configurations are generated based on a set of parameters describing factors to be varied in the experiments and a set of constraints representing limitations on operations that can be performed with the set of resources. A set of experiments is defined based on a selected configuration. The constraints can be represented as patterns defining an application of a parameter to a set of one or more points of an experimental lattice.