Abstract: There is disclosed a computer-implemented cloud computing scoring system. In an embodiment, a parser receives unstructured sentiment data commenting on a scored service. The parser identifies in the unstructured sentiment data a service category of the scored service. The parser selects from the unstructured sentiment data text relating to the service category and matching one or more opinionative words and phrases listed in a keyword dictionary, thereby producing a structured comment associated with the service category. The structured comment is classified as positive or negative according to a list of exemplary sentiment data sets contained in a learning seed file. The exemplary sentiment data sets are manually assigned a positive or a negative polarity. The learning seed file is configured for enhancement by the ongoing addition of structured sentiment data, the structured sentiment data commenting on the scored service and having a polarity classification.

Abstract: A method of generating secondary questions in a question-answer system. Missing information is identified from a corpus of data using a computerized device. The missing information comprises any information that improves confidence scores for candidate answers to a question. The computerized device automatically generates a plurality of hypotheses concerning the missing information. The computerized device automatically generates at least one secondary question based on each of the plurality of hypotheses. The hypotheses are ranked based on relative utility to determine an order in which the computerized device outputs the at least one secondary question to external sources to obtain responses.

Abstract: A method of generating secondary questions in a question-answer system. Missing information is identified from a corpus of data using a computerized device. The missing information comprises any information that improves confidence scores for candidate answers to a question. The computerized device automatically generates a plurality of hypotheses concerning the missing information. The computerized device automatically generates at least one secondary question based on each of the plurality of hypotheses. The hypotheses are ranked based on relative utility to determine an order in which the computerized device outputs the at least one secondary question to external sources to obtain responses.

Abstract: A method, system, and computer program product for IO leveling comprising receiving an IO, determining if there is a delay for processing IO because of pending IO, based on a positive determination there is a delay for processing IO, determining a priority for the IO, and based on the priority of IO determining whether to process the IO.

Abstract: A mechanism is provided in a data processing system for conducting error analysis for a question answering system. Responsive to the question answering system generating one or more candidate answers for an input question, wherein the one or more candidate answers are determined to be incorrect, the mechanism instantiates a plurality of instances of the question answering system with a modification to each instance. The mechanism provides the input question to each of the plurality of instances of the question answering system. The mechanism analyzes results from the plurality of instances of the question answering system to identify at least one modification that led to improved results. The mechanism presents a graphical output based on the analysis.

Abstract: A method for expanding an answer key to verify a question and answer system is provided in the illustrative embodiments. A definition is constructed of an extended answer type. The extended answer type represents an answer type of an unrepresented answer. The unrepresented answer is unrepresented in the answer key as a valid response to a question in a set of valid responses to the question in the answer key. The extended answer type is created in the answer key according to the definition. The extended answer type is populated such that the unrepresented answer becomes as additional valid response to the question, the creating and the populating extending the answer key to form an extended answer key. The populated extended answer type in the extended answer key is used to verify that a generated answer from the Q and A system is correct.

Abstract: Methods, systems and computer program products are provided for cross-media recommendation by store a plurality of taste profiles corresponding to a first domain and a plurality of media item vectors corresponding to a second domain. An evaluation taste profile in the first domain is applied to a plurality of models that have been generated based on relationship among the plurality of taste profiles and the plurality of media item vectors, and obtain a plurality of resulting codes corresponding to at least one of the plurality of media item vectors in the second domain.

Abstract: In an approach to create a rule hierarchy model, a computer receives from a user a set of rules and an association between each rule in the set of rules and a concept of the plurality of concepts in a concept hierarchy. The computer receives set of rules is stored separately from the concept hierarchy. The computer determines a rule hierarchy where a rule of the set of rules is associated with one or more concepts. Furthermore, the computer creates a rule hierarchy model based on the concept hierarchy and the rule hierarchy wherein grouping related rules in a common structure provides efficient management that facilitates rule authoring, browsing, and extraction.

Abstract: Technical solutions are described to accelerate training of a multi-layer convolutional neural network. According to one aspect, a computer implemented method is described. A convolutional layer includes input maps, convolutional kernels, and output maps. The method includes a forward pass, a backward pass, and an update pass that each include convolution calculations. The described method performs the convolutional operations involved in the forward, the backward, and the update passes based on a first, a second, and a third perforation map respectively. The perforation maps are stochastically generated, and distinct from each other. The method further includes interpolating results of the selective convolution operations to obtain remaining results. The method includes iteratively repeating the forward pass, the backward pass, and the update pass until the convolutional neural network is trained. Other aspects such as a system, apparatus, and computer program product are also described.

Abstract: The method includes identifying, by one or more computer processors, a grouping of elements in a storage system. The method further includes identifying, by one or more computer processors, a first element from the identified grouping of elements. The method further includes identifying, by one or more computer processors, a root unit of the first element from the identified grouping of elements. The method further includes determining, by one or more computer processors, a past usage history of computer resources for the identified root unit of the first element. The method further includes calculating, by one or more computer processors, a future usage of computer resources for the identified root unit based upon the determined past usage history.

Abstract: A composition of autonomous self-learning skill software components interact with one another to solve a real time complex task. Each software component includes a knowledge base and an inference algorithm that uses the knowledge base to solve a given data-processing task involving input data. Each software component may also include a machine learning algorithm for training the knowledge base with new data in real-time as the new data is received.

Abstract: One embodiment of an accelerator includes a computing unit; a first memory bank for storing input activations and a second memory bank for storing parameters used in performing computations, the second memory bank configured to store a sufficient amount of the neural network parameters on the computing unit to allow for latency below a specified level with throughput above a specified level. The computing unit includes at least one cell comprising at least one multiply accumulate (“MAC”) operator that receives parameters from the second memory bank and performs computations. The computing unit further includes a first traversal unit that provides a control signal to the first memory bank to cause an input activation to be provided to a data bus accessible by the MAC operator. The computing unit performs computations associated with at least one element of a data array, the one or more computations performed by the MAC operator.

Abstract: A content engagement system includes game logic configured to help users and their social contacts to find interesting content. The outputs can be used for optimizing ads, content, search results, etc., on mobile devices, social networking sites and similar domains.

Abstract: Systems and methods may include receiving, by an expert system, performance data for a monitored system. The systems and methods may include generating a prediction for the monitored system in response to determining that the performance data satisfies a condition. The prediction may identify an anomaly that is predicted to occur. The systems and methods may include receiving, by a filter system, the prediction, information identifying the condition, and user information. The user information may include user preference information and user feedback information. The systems and methods may include determining a filter criteria based on the user information. The filter criteria may be based on the preferences for predictions to be provided to the user and on the historical user feedback regarding the historical predictions. The systems and methods may include providing the prediction to the user in response to determining that the particular prediction satisfies the filter criteria.

Abstract: A system is provided for event-based monitoring of a subject's well-being within an unattended setting. A plurality of sensors are disposed within the setting for sensing disparate events, and an analytics processing portion is coupled to the sensors to collectively acquire sensing data therefrom, and map a plurality of sensed data points for a selected combination of disparate events to a conduct adaptively characterized for the subject. The mapping occurs according to a set of pre-established reference event patterns, relative to which each characterized conduct is screened for excessive aberration. The analytics processing portion actuates generation of a graphic user interface displaying at least one reporting page. The reporting page contains for each characterized conduct certain graphic indicia determined responsive to the screening thereof.

Abstract: Methods, systems, apparatuses, and computer program products are provided for altering the behavior of an electronic personal assistant based on a situation associated with a mobile device. A situation is sensed with a plurality of sensors to generate sensor data. A situation score is calculated based on the sensor data. Behavior of an electronic personal assistant is altered based on the calculated situation score. In one aspect, the situation is a driving situation in which a driver drives a vehicle on a roadway. In such case, a driving situation score is calculated based on the sensor data, and behavior of the electronic personal assistant is altered based on the calculated driving situation score, such as suspending interactions by the electronic personal assistant with the driver to avoid the driver being distracted.

Abstract: Techniques herein train a multilayer perceptron, sparsify edges of a graph such as the perceptron, and store edges and vertices of the graph. Each edge has weight. A computer sparsifies perceptron edges. The computer performs a forward-backward pass on the perceptron to calculate a sparse Hessian matrix. Based on that Hessian, the computer performs quasi-Newton perceptron optimization. The computer repeats this until convergence. The computer stores edges in an array and vertices in another array. Each edge has weight and input and output indices. Each vertex has input and output indices. The computer inserts each edge into an input linked list based on its weight. Each link of the input linked list has the next input index of an edge. The computer inserts each edge into an output linked list based on its weight. Each link of the output linked list comprises the next output index of an edge.

Abstract: An automatic scaling system and method for reducing state space in reinforced learning for automatic scaling of a multi-tier application uses a state decision tree that is updated with new states of the multi-tier application. When a new state of the multi-tier application is received, the new state is placed in an existing node of the state decision tree only if a first attribute of the new state is same as a first attribute of any state contained in the existing node and a second attribute of the new state is sufficiently similar to a second attribute of each existing state contained in the existing node based on a similarity measurement of the second attribute of each state contained in the existing node with the second attribute of the new state.

Abstract: An apparatus and method of training a convolutional neural network (CNN) are provided. A method of training a CNN including a plurality of convolution layers stored in a memory involves approximating, using a processor, a convolution layer among the plurality of convolution layers using a low-rank approximation; reducing the number of output reconstruction filters of the approximated convolution layer; and modifying a structure of the CNN based on an approximation result and the reduced number of output reconstruction filters.

Abstract: A method for applying personalized machine learning models is provided. The method includes producing one or more feature vectors that represents features of one of a plurality of files of a file system and selecting, from a plurality of personalized machine learning models that model user accesses to the files of the file system a subset of the personalized machine learning models each of which has a plurality of non-zero weights corresponding to non-zero features of the one or more feature vectors. The method includes determining from the subset of personalized machine learning models which users of a plurality of users of the file system are likely to access the one of the plurality of files.