Abstract: An automated Web portal template generation method includes parsing, via a parser subsystem, a number of Webpages of a first Website from which a Web portal template to be customized is to be accessed. The method further includes producing an entity feature set for the first Website based on a result of the parsing and processing the entity feature set for the first Website via a classifier subsystem to produce a set of data that represents, for each of a plurality of entities, a respective probability of the entity belonging to a respective one of a plurality of classes. The method additionally includes performing, by a color matching subsystem, color matching on the set of data produced by the classifier subsystem to generate a number of proposed color combinations for a proposed customization of the Web portal template.

Abstract: Energy usage can be monitored within at least one building having a plurality of energy consuming components. A database can be generated that contains values for a set of data points corresponding to data received from the plurality of energy consuming components. A change in a configuration can be detected for the plurality of energy consuming components based upon a change in values received from plurality of energy consuming components relative to the database. Based upon the change, an additional data point can be added to the set of data points in the database. Based upon the values for the set of data points, a probability can be determined that a rule for the additional data point is valid. A message can then be generated that includes the determined probability.

Abstract: A method, computer program product, and computer system, for receiving a first set of ground truth instances from a first source. A second set of ground truth instances may be received from a second source. The first and second sets of ground truth instances may be weighed differently based on a level of trust associated with each of the first and second sources. The weighted first and second sets of ground truth instances may be applied in a machine learning task executed by a computer.

Abstract: The present approach relates to the training of a machine learning algorithm for image generation and use of such a trained algorithm for image generation. Training the machine learning algorithm may involve using multiple images produced from a single set of tomographic projection or image data (such as a simple reconstruction and a computationally intensive reconstruction), where one image is the target image that exhibits the desired characteristics for the final result. The trained machine learning algorithm may be used to generate a final image corresponding to a computationally intensive algorithm from an input image generated using a less computationally intensive algorithm.

Abstract: Techniques disclosed herein relate to learning and applying contextual patient similarities. Multiple template similarity functions may be provided. Each template similarity function may compare a respective subset of features of a query entity feature vector with a corresponding subset of features of a candidate entity feature vector. A composite similarity function may be provided as a weighted combination of respective outputs of the template similarity functions. A plurality of labeled entity vectors may be provided as context training data. An approximation function may be applied to approximate a first context label for each respective labeled entity vector. A first context specific composite similarity function may be trained based on the composite similarity function by learning first context weights for the template similarity functions using a first loss function based on output of application of the approximation function to the first context training data.

Abstract: A system is presented for emulating sampling of a quantum computer having a plurality of qubits arranged in a grid topology with N columns. The system includes a classical processor that is configured by operational instructions to perform operations that include producing final weights and variable assignments for the N columns based on N iterative passes through the grid topology, wherein each of the N iterative passes generates preliminary weights and variable assignments for a corresponding subset of the N columns, wherein the preliminary weights and variable assignments for a selected column of the corresponding subset based on the preliminary weights and variable assignments generated for a column adjacent to the selected column of the corresponding subset, and wherein the sampling of the quantum computer having the plurality of qubits is emulated by producing a plurality of samples from the N iterative passes based on the final weights and variable assignments for each of the N columns.

Abstract: A data processing method receives a set of time-series user data and also receives a privacy requirement of the time-series user data. Next, the time-series user data is transformed using the privacy requirement such that the transforming satisfies differential privacy.

Abstract: A mechanism is provided in a data processing system for exploring knowledge. The mechanism receives a set of known facts. The mechanism traverses paths in an ontology for a domain of knowledge from known facts in the set of known facts to one or more hypotheses. The ontology includes a plurality of entity types and a plurality of relationships between the entity types. The mechanism presents one or more hypotheses to a user.

Abstract: A disclosed method may include receiving geographic coordinates of a location at which two parties are to rendezvous, generating a human-understandable geospatial descriptor for the request location, and sending the descriptor to respective devices of the two parties for presentation to the two parties. Generating the human-understandable geospatial descriptor may include identifying a human-visible feature in the vicinity of the request location that is labeled within available map data, selecting, based on a descriptor generation model, a reference expression relative to the identified feature, and applying a grammar-based constructor to the label and the selected reference expression to form the human-understandable geospatial descriptor. The model may be tuned using machine learning. The two parties may include a ride requestor and a ride provider in a ridesharing service. The identified feature may be a point of interest, landmark, street name, intersection, marker, or structure.

Abstract: A self-learning system for analytical attribute and clustering segmentation may be provided. A text classifier may identify a log description of a log entry in response to text of the log description being associated with indicators of a word model. A datafield classifier may generate a datafield metrics including an accuracy value of the attribute identifiers representing the datafield. A metafield classifier may generate a context metrics for the context of the log entry, the context metrics including an accuracy value of the attribute identifiers representing the metafields. A combination classifier may form a weighted classification set and select an attribute identifier as being representative of the datafield based on the weighted classification set. The combination classifier may further evaluate an attribute importance value of each attribute identifier, and select an attribute identifier having a top attribute importance value.

Abstract: A computing system trains a clustering model. (A) Beta distribution parameter values are computed for each cluster using a mass parameter value and a responsibility parameter vector of each observation vector. (B) Parameter values are computed for a normal-Wishart distribution for each observation vector included in a batch of a plurality of observation vectors. (C) Each responsibility parameter vector defined for each observation vector of the batch is updated using the beta distribution parameter values, the parameter values for the normal-Wishart distribution, and a respective observation vector of the selected batch of plurality of observation vectors. (D) A convergence parameter value is computed. (E) (A) to (D) are repeated until the convergence parameter value indicates the responsibility parameter vector defined for each observation vector is converged. A cluster membership is determined for each observation vector using the responsibility parameter vector.

Abstract: A model induction method for explainable artificial intelligence (XAI) may be shown and described. A model of a black-box AI may be an input to the model induction method, along with a set of sample input data. A linear or non-linear predictor function may be used to predict the output of the black-box model, producing a set of data points. The data points may be partitioned by a partitioning function, and each partition may represent one or more rules. The data may also be transformed using a number of transformation functions, such as a polynomial expansion. A local model may be fitted to the transformed function or functions. A set of rules may be interpreted from the local models and may form a white-box AI model. Linear or non-linear data may be modeled by the white-box model. Further, the white-box model may be implemented on a low-power device.

Abstract: Techniques for detecting and correcting anomalies in computer-based reasoning systems are provided herein. The techniques can include obtaining current context data and determining a contextually-determined action based on the obtained context data and a reasoning model. The reasoning model may have been determined based on one or more sets of training data. The techniques may cause performance of the contextually-determined action and, potentially, receiving an indication that performing the contextually-determined action in the current context resulted in an anomaly. The techniques include determining a portion of the reasoning model that caused the determination of the contextually-determined action based on the obtained context data and causing removal of the portion of the model that caused the determination of the contextually-determined action, to produce a corrected reasoning model.

Abstract: In some aspects, a computing system can generate and optimize a neural network for risk assessment. The neural network can be trained to enforce a monotonic relationship between each of the input predictor variables and an output risk indicator. The training of the neural network can involve solving an optimization problem under a monotonic constraint. This constrained optimization problem can be converted to an unconstrained problem by introducing a Lagrangian expression and by introducing a term approximating the monotonic constraint. Additional regularization terms can also be introduced into the optimization problem. The optimized neural network can be used both for accurately determining risk indicators for target entities using predictor variables and determining explanation codes for the predictor variables. Further, the risk indicators can be utilized to control the access by a target entity to an interactive computing environment for accessing services provided by one or more institutions.

Abstract: A view generator receives support text characterizing a support requirement for available information technology (IT) support, the support text being received in sentence form via a graphical user interface (GUI). A text analyzer performs natural language processing on the support text and thereby identifies at least one sentence part and at least one named entity within the support text. A support record generator relates each of the at least one sentence part and the at least one named entity to a support record type, and generates a support data record for the support requirement, including filling individual fields of the support data record using the at least one sentence part and the at least one named entity.

Abstract: A method for training a multi-class classification model includes receiving training data corresponding to a plurality of classes. For each class in the plurality of classes, the method includes training a binary classification model configured to determine whether or not an observation of training data belongs to the class and for each observation of training data identified as belonging to the class, extracting one or more class identification features from the observation of training data based on activations of an intermediate attention layer in the binary classification model. A multi-class classification model is trained using the class identification features extracted for each of the plurality of classes.

Abstract: A system and method for an e-problem solving board is disclosed. Said e-problem solving board allows automated classification and management of one or more problems. In some embodiments, the method uses one or more machine learning algorithms for classifying problems according to their complexity. In other embodiments, the method uses collaborative filtering algorithms for classifying the complexity of the problem. In these embodiments, the method uses collaborative filtering algorithms for assigning employees to problems and providing a set of suggestions to address the one or more problems. In some embodiments, the system provides status reports regarding the one or more problems. In other embodiments, the system allows multiple teams, operating in different geographic locations, to work on a single problem. Further to these embodiments, the system allows users to track and continually update problems.

Abstract: A self-learning system for categorizing log entries may be provided. The system may display a first log entry and receive a categorical identifier for the first log entry. The system may parse the first log entry for predetermined text information and predetermined image information. The predetermined text information may be included in a datafield classifier and the predetermined image information included in a metadata classifier. The system may identify the predetermined text information in the log entry and adjust a first prioritization of respective categorical identifiers included in the datafield classifier. The system may identify the predetermined image information in the first log entry and adjust a second prioritization of the respective categorical identifiers included in the metadata classifier. The system may map a second log entry to the categorical identifier based on adjustment of the first prioritization or adjustment of the second prioritization.

Abstract: A computer implemented method of detecting complex user activities, comprising using processor(s) in each of a plurality of consecutive time intervals for: obtaining sensory data from wearable inertial sensor(s) worn by a user, computing an action score for continuous physical action(s) performed by the user, the continuous physical action(s) extending over multiple time intervals are indicated by repetitive motion pattern(s) identified by analyzing the sensory data, computing a gesture score for brief gesture(s) performed by the user, the brief gesture(s) bounded in a single basic time interval is identified by analyzing the sensory data, aggregating the action and gesture scores to produce an interval activity score of predefined activity(s) for a current time interval, adding the interval activity score to a cumulative activity score accumulated during a predefined number of preceding time intervals and identifying the predefined activity(s) when the cumulative activity score exceeds a predefined threshold

Abstract: In one embodiment, a method includes a device configured to obtain a plurality of sparse, categorical, and numerical features. The device may generate a plurality of ordered combinations of features, wherein each ordered combination of features comprises at least a first feature and a second feature. The device may identify a user account of a social networking system and generate one or more deep feature values associated with the user account for each of the plurality of ordered combination of features. The generation may comprise extracting a first feature value and a second feature value associated with the user account using a social graph of the social networking system. The first and second feature values correspond to the first and second features of the ordered combination of features, respectively. The device may then train a machine-learning model using the generated deep feature values associated with the user account.