Abstract: A computer implemented method for training a neural network to capture a structural feature specific to a set of chemical compounds is disclosed. In the method, the computer system reads an expression describing a structure of the chemical compound for each chemical compound in the set and enumerates one or more combinations of a position and a type of a structural element appearing in the expression for each chemical compound in the set. The computer system also generates training data based on the one or more enumerated combinations for each chemical compound in the set. The training data includes one or more values with a length, each of which indicates whether or not a corresponding type of the structural element appears at a corresponding position for each combination. Furthermore, the computer system trains the neural network based on the training data for the set of the chemical compounds.

Abstract: A system comprising: a processor; a data bus coupled to the processor; and a non-transitory, computer-readable storage medium embodying computer program code, the non-transitory, computer-readable storage medium being coupled to the data bus. The computer program code interacting with a plurality of computer operations and comprising instructions executable by the processor and configured for: receiving data from a plurality of data sources; processing the data from the plurality of data sources, the processing the data from the plurality of data sources identifying a plurality of knowledge elements; and, storing the knowledge elements within the cognitive graph as a collection of knowledge elements, the storing universally representing knowledge obtained from the data; and, generating a cognitive insight based upon the collection of knowledge elements stored within the cognitive graph, the generating the cognitive insight using an insight agent to access the collection of knowledge elements.

Abstract: A computer-implementable method for generating an insight comprising: receiving data from a plurality of data sources; processing the data from the plurality of data sources, the processing the data from the plurality of data sources identifying a plurality of knowledge elements; and, storing the knowledge elements within the cognitive graph as a collection of knowledge elements, the storing universally representing knowledge obtained from the data; and, generating a cognitive insight based upon the collection of knowledge elements stored within the cognitive graph, the generating the cognitive insight using an insight agent to access the collection of knowledge elements.

Abstract: A system comprising: a processor; a data bus coupled to the processor; and a non-transitory, computer-readable storage medium embodying computer program code, the non-transitory, computer-readable storage medium being coupled to the data bus. The computer program code interacting with a plurality of computer operations and comprising instructions executable by the processor and configured for: receiving data from a plurality of data sources; processing the data from the plurality of data sources, the processing the data from the plurality of data sources identifying a plurality of knowledge elements; and, storing the knowledge elements within the cognitive graph as a collection of knowledge elements, the storing universally representing knowledge obtained from the data, each knowledge element of the collection of knowledge elements being persisted in its original form.

Abstract: A computer-implementable method for managing a cognitive graph comprising: receiving data from a plurality of data sources; processing the data from the plurality of data sources, the processing the data from the plurality of data sources identifying a plurality of knowledge elements; and, storing the knowledge elements within the cognitive graph as a collection of knowledge elements, the storing universally representing knowledge obtained from the data, each knowledge element of the collection of knowledge elements being persisted in its original form.

Abstract: A method and associated systems for a self-learning energy switch. The switch creates an array of cognitive models for each candidate energy source. Each array returns a probability that its corresponding source is the most cost-effective and operationally suitable energy supplier at that time. Each model in an array contributes to the array's returned probability as a function of a corresponding class of decision-making factors. The system fine-tunes the models by weighting them as functions of extrinsic evidentiary information that may imply future behavior of the decision-making factors and combines each model's returned probabilities to select an optimal energy source. The system then automatically routes power from the optimal source to a consumer's energy-consuming premises. This self-learning procedure repeats indefinitely, continuously tuning the models in response to identifying additional extrinsic evidence and reasons why the system's previous energy selections were either optimal or non-optimal.

Abstract: Mechanisms are provided to implement a self-training engine of a question and answer system. The self-training engine pairs an unanswered natural language question with portions of an electronic document to generate an unlabeled data set. The self-training engine trains a model based on a labeled data set comprising a finite number of question and answer pair data structures and applies the model to the unlabeled data set to identify an answer from the portions of the electronic document to the unanswered natural language question. The self-training engine modifies the labeled data set to include the identified answer and corresponding unanswered natural language question as an additional question and answer pair data structure. The self-training engine then trains the model based on the modified labeled data set.

Abstract: A system, method and computer program product for disambiguating one or more entity mentions in one or more documents. The method facilitates the simultaneous linking entity mentions in a document based on convolution neural networks and recurrent neural networks that model both the local and global features for entity linking. The framework uses the capacity of convolution neural networks to induce the underlying representations for local contexts and the advantage of recurrent neural networks to adaptively compress variable length sequences of predictions for global constraints. The RNN functions to accumulate information about the previous entity mentions and/or target entities, and provide them as the global constraints for the linking process of a current entity mention.

Abstract: A data classification device includes an estimation unit that estimates, for each of one or more classes provided for learning data pieces in a feature-amount-data space that includes multiple learning data pieces, probability densities of learning data pieces belonging to the class and learning data pieces not belonging to the class around a judgment target data piece in the feature-amount-data space, a calculation unit that calculates, based on the probability densities, an index indicating how much the judgment target data piece is likely to belong to the class, and a judgment unit that judges which class the judgment target data piece belongs to by using the index. Based on distribution of positive data pieces belonging to the class and negative data pieces not belonging to the class around the judgment target data piece, the estimation unit determines a size of a region used for the estimation.

Abstract: A computer-implemented method that includes a computing system receiving a plurality of data items that include content data. The system then generates a first inference model based on the received content data. The system also receives metadata associated with respective data items and then generates a second inference model based on the received metadata data. The system further determines, based on the first inference model, a first weight parameter associated with the content data, and determines, based on the second inference model, a second weight parameter associated with the metadata. The system then generates a classification model for determining a classification of at least one data item of the plurality of data items. The classification model is generated based on at least one of the first weight parameter or the second weight parameter.

Abstract: Systems, apparatus, and methods are disclosed for managing and forecasting energy usage. A distributed forecast device is located remote from a central server. The distributed forecast device receives from the central server information related to one or more accounts associated with the distributed forecast device. The distributed forecast device receives energy usage data from one or more energy meters for each of the accounts. The distributed forecast device predicts an energy usage forecast for each of the accounts based on the energy usage data and the information from the central server.

Abstract: An enclosure design facilitates heat dissipation from a space-limited computer core device. An external computer platform is provided to connect the computer core device, the external computer platform including a fan that provides an air flow to the connected computer core device. The computer core device and the computing platform may be tightly connected by connectors located on their respective enclosure walls. Both the computer core device and the external computing platform are provided air inlets and outlets on their respective enclosures. When connected, an air inlet of the computer core device faces an air outlet of the external computing platform such that a single cooling air flow flows through the external computing platform and the computer core device. The external computing platform may include a built-in fan to blow air into or draw air from the matching air inlets and outlets.

Abstract: A machining path creating device for wire electric discharge machines creates a machining path for machining a keyway in a side surface of a round hole. The machining path creating unit compares a “keyway depth” and a diameter of the round hole input by an operator. The machining path creating device determines that the input “keyway depth” is a cut depth when the input “keyway depth” is smaller than the diameter of the round hole and that a value obtained by subtracting the diameter of the round hole from the input “keyway depth” is the cut depth when the input “keyway depth” is larger than the diameter of the round hole.

Abstract: A micro grid power system includes a plurality of generator sets, a photovoltaic system, an external load, and a controller. Each generator set includes an engine and a sensor to determine a temperature of exhaust gases exiting the engine. The photovoltaic system includes a plurality of photovoltaic panels for generating power. The external load is powered by the plurality of generator sets and the photovoltaic system. The controller is configured to determine the external load, and control the plurality of generator sets to maintain the temperature above a threshold temperature. The controller is further configured to determine a remaining load which is the external load subtracted by a portion of the external load powered by the plurality of generator sets, and control the photovoltaic system to power the remaining load.

Abstract: An operation limiting device applicable to a wide variety of machine tools and which improves the freedom of setting an operation limiting area. Two or three drive axes (Xa, Za) in different directions are designated from among a plurality of drive axes in the machine tool and the designated drive axes (Xa, Za) form the two or three dimensional coordinate system. An area (Ar) for limiting the operation of the mobile body is set in the two or three dimensional coordinate system. A drive axis (Xa) which is in operation is determined, and in the two or three coordinate system including the drive axis which is in operation the operation of the mobile body is limited when the mobile body enters into the limiting area (Ar).

Abstract: A method for evaluating a three-dimensional (3D) object produced by a three-dimensional (3D) object printer includes generating with a scanner a first data file of the 3D object produced by the three-dimensional object printer. The first data file includes data corresponding to an interior surface and an exterior surface of the 3D object. The method further includes comparing the first data file with a second data file. The second data file is used to operate the three-dimensional object printer to produce the 3D object. The method further includes identifying differences between the first data file and the second data file and identifying a process to rectify a defect on the 3D object based on the identified differences to enable evaluation of the production of the 3D object by the three-dimensional object printer.

Abstract: The gap control axis following speed calculating unit of a numerical controller controlling a laser processing machine calculates a gap control axis following speed based on a displacement amount obtained from a separation distance and a reference distance between a nozzle tip and a workpiece, and a gap control gain. A first determining unit determines that the nozzle tip has entered an approach side when the separation distance becomes smaller than the reference distance. A first changing unit changes the gap control gain continuously in accordance with the displacement amount.

Abstract: A method of aggregated virtual power plant (VPP) control includes receiving control variable values. The control variables are received for control variables related to energy production and loads of devices electrically coupled to an electrical grid and communicatively coupled to the VPP controller server. The method includes inputting the control variable values into an objective algorithm. The method may include executing the objective algorithm. Executing the objective algorithm includes adjusting energy loads and energy production of prosumers, adjusting an energy amount supplied from a supply side for multiple time intervals, and adjusting curtailment of the energy loads in the prosumers based thereon. The method includes generating a VPP DR event schedule and communicating it to VPP client servers. The VPP DR event schedule includes control signals that are configured to affect an operating condition of the devices that are controlled by the VPP client servers.

Abstract: Method and computer-readable model for additively manufacturing a ducting arrangement (20) in a combustion stage are provided. The ducting arrangement may include a combustor wall (40) in a combustion stage fluidly coupled to receive a cross-flow of combustion products (21). An injector assembly (12) may be in fluid communication with cooling fluid conduits (46) in the combustor wall to receive cooling fluid that passes through the cooling fluid conduits. The injector assembly may include means for injecting (24, 25, 26) a flow of the cooling fluid (22) arranged to condition interaction of a flow of reactants (19) injected with the cross-flow of combustion products. Respective duct components or the entire ducting arrangement may be formed as a unitized structure, such as a single piece using a rapid manufacturing technology, such as 3D Printing/Additive Manufacturing (AM) technologies.

Abstract: An intelligent control method used in an intelligent control system includes an intelligent control device and an air condition device. The intelligent control device records CO2 amount value in an area continuously after the air condition device is started, and calculates a CO2 reference value based on the recorded CO2 amount value after a luminaire device in the area is turned off. Then, the intelligent control device records the CO2 amount value for a time period and calculates a CO2 compared value after the time period passes. The intelligent control device determines whether the CO2 amount value trend is varied upwardly or downwardly for choosing among keeping the current state, switching the operating mode of the air condition device or turning off the air condition device.