Abstract: Method embodiments and/or system embodiments are provided that may be utilized to recommend online content to users based, at least in part on a prediction of diffusion of online content through a social network.

Abstract: Network flow classification can include clustering a network flow database into a number of at least one of applications and network flows. Network flow classification can include classifying the number of the at least one of applications and network flows.

Abstract: A mechanism is provided in a data processing system for article summarization. The mechanism analyzes an article to identify entities and relationships within the article. The article is an item of unstructured content. The mechanism performs information augmentation based on the identified entities and relationships using one or more cognitive services to collect augmented information from a corpus of information. The mechanism generates one or more visualization components based on the identified entities and relationships and the augmented information. The mechanism presents a summarization comprising the one or more visualization components to a user.

Abstract: In some examples, unstructured data is evaluated using a natural language processing model to output a set of subjective indicators. These subjective indicators are scored using a predictive model to determine whether a dependent user has or is likely to develop a particular condition such as a cellular abnormality.

Abstract: According to some embodiments, the present disclosure may relate to a method of neural network analysis that includes receiving a first electronic message, storing it in a storage device, and decoding it to output a first data structure. The first electronic message may reference a first dictionary entry correlating the first electronic message to the first data structure including more bits than the first message. The method may also include providing the first data structure to a processing element to perform a data structure analysis on the first data structure yielding a second data structure including more bits than the first electronic message. The method may also include matching the second data structure to a second dictionary entry correlating the second data structure to a second electronic message that includes fewer bits than the second data structure, and transmitting the second electronic message instead of the second data structure.

Abstract: In an approach for providing adaptive software inventory scan frequencies and schedules, a processor receives information from an initial scan of a set of software inventory scans, wherein the information includes at least one of: running processes, file system entries, registry entries, and software catalog evaluations. A processor analyzes the information from the initial scan. A processor predicts an outcome for future software inventory scans based on the analysis of the information, wherein the prediction includes a scanning frequency and a scanning schedule.

Abstract: A method for generating one or more push notifications to a user device is described. The method comprises: obtaining history data representing a history of online activities of a user and candidate data representing a set of candidate information; generating, based on the history data and the candidate data, user profile vectors representing a user profile associated with the user and content vectors representing a set of content profiles associated with the set of candidate information; generating, based on a machine learning model trained with a history of online activities, embedding user feature vectors and embedding content feature vectors based on the history data and the candidate data; and providing for transmission information for one or more push notifications including first candidate information of to a user device associated with the user, the first candidate information being determined from the set of candidate information based on the aforementioned vectors.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for hiding information using neural networks. One of the methods includes maintaining data mapping each of a plurality of classes to a respective piece of information that may potentially be hidden in a received data item; receiving a new data item; receiving data identifying a first piece of information to be hidden in the new data item; and modifying the new data item to generate a modified data item that, when processed by a neural network configured to classify input data items belonging to one of the plurality of classes, is classified by the neural network as belonging to a first class of the plurality of classes that is mapped to the first piece of information in the maintained data.

Abstract: An online system predicts topics for content items. The online system provides one or more topic labels for a user to apply concurrently while a user is composing a post, in response to requests periodically received from the user's device. A request includes information such as content composed by the user and contextual information. The online system employs machine learning techniques to analyze content composed by a user and contextual information thereby to predict topic labels. Different machine learning models for classifying individual topic labels, identifying relevant topic labels, and/or detecting changes in existing topic predictions are developed. Some machine learning models predict topics for full content and some predict topics for partial content. The online system trains the machine learning models to ensure accurate topic predictions are provided timely. The online system employs various machine learning model training methods such as active training and gradient training.

Abstract: Technologies for using a shifted neural network include a compute device to determine a shift-based activation function of the shifted neural network. The shift-based activation function is a piecewise linear approximation of a transcendental activation function and is defined by a plurality of line segments such that a corresponding slope of each line segment is a power of two. The compute device further trains the shifted neural network based on shift-based input weights and the determined shift-based activation function.

Abstract: The software system processes extracts reliable, significant and relevant patterns. System runs through preprocessing steps. System then generates the size 1 patterns. It then checks for both reliability and refinability of the size 1 patterns. System grows the refinable patterns by increasing the attributes and its values in the pattern by one at a time to find a size 2 pattern. The system then uses the number of pattern occurrences of size 2 pattern as a basis to find the reliable patterns. System also checks for statistical significance over the size 1 patterns and once again for the refinability of the size 2 patterns. System checks for relevance of the size 1 patterns by obtaining the disjointed record complement set. Software system readjusts the pattern statistics of size 1 and removes the non-relevant super-patterns. This process is repeated from size 2 to N.

Abstract: Machine learning based models, for example, neural network models employ large numbers of parameters, from a few million to hundreds of millions or more. A machine learning based model is trained using fewer parameters than specified. An initial parameter vector is initialized, for example, using random number generation based on a seed. During training phase, the parameter vectors are modified in a subspace around the initial vector. The trained model can be stored or transmitted using seed values and the trained parameter vector in the subspace. The neural network model can be uncompressed using the seed values and the trained parameter vector in the subspace. The compressed representation of neural networks may be used for various applications such as generating maps, object recognition in images, processing of sensor data, natural language processing, and others.

Abstract: The disclosed embodiments illustrate a method and a system for predicting future activities of a user on a social media platform. The method includes extracting a first time series of one or more historical activities performed by the user from a social media platform server. The method further includes receiving a second time series of one or more future events from a requestor-computing device. The method further includes determining a first set of forecast values and a second set of forecast values based on the first time series and/or the second time series, wherein the first set of forecast values is determined using an ARIMA technique, and the second set of forecast values is determined using a regression modelling technique. The method further includes predicting the future activities of the user based on the first set of forecast values and the second set of forecast values.

Abstract: A safety system associated with a travel coordination system collects safety data describing safety incidents by providers and generates a plurality of safety incident prediction models using the safety data. The safety incident prediction models predict likelihoods that providers in the computerized travel coordination system will be involved in safety incidents. Two types of safety incidents predicted by the safety system include dangerous driving incidents and interpersonal conflict incidents. The safety system uses the plurality of safety incident prediction models to generate a set of predictions indicating probabilities that a given provider in the travel coordination system will be involved in a safety incident in the future. The safety system selects a safety intervention for the given provider responsive to the set of predictions and performs the selected safety intervention on the given provider.

Abstract: Embodiments of the invention relate to a multiplexed neural core circuit. One embodiment comprises a core circuit including a memory device that maintains neuronal attributes for multiple neurons. The memory device has multiple entries. Each entry maintains neuronal attributes for a corresponding neuron. The core circuit further comprises a controller for managing the memory device. In response to neuronal firing events targeting one of said neurons, the controller retrieves neuronal attributes for the target neuron from a corresponding entry of the memory device, and integrates said firing events based on the retrieved neuronal attributes to generate a firing event for the target neuron.

Abstract: A method for training a deep learning network includes defining a loss function corresponding to the network. Training samples are received and current parameter values are set to initial parameter values. Then, a computing platform is used to perform an optimization method which iteratively minimizes the loss function. Each iteration comprises the following steps. An eigCG solver is applied to determine a descent direction by minimizing a local approximated quadratic model of the loss function with respect to current parameter values and the training dataset. An approximate leftmost eigenvector and eigenvalue is determined while solving the Newton system. The approximate leftmost eigenvector is used as negative curvature direction to prevent the optimization method from converging to saddle points. Curvilinear and adaptive line-searches are used to guide the optimization method to a local minimum. At the end of the iteration, the current parameter values are updated based on the descent direction.

Abstract: The present invention provides for a computerized method for generation an action instruction based on cognitive learning. The present method and apparatus provides for accessing at least one neural network having a data set stored therein. The present method and apparatus determines at least one meaning data map of the data set. The meaning data map includes a plurality of cognitive frames that are embedded within a nine dimensional hypercube. For example, one embodiment may include a four frames making up a four dimensional cognitive dimension grammar that is embedded within the nine dimensional hypercube. The method and apparatus calculates a data meaning based on the at least one meaning data map. From this data meaning, the method and system then generates an action instruction.

Abstract: A method predicts the fall risk of a user based on a machine learning model. The model is trained using data about the user, which may be from wearable sensors and depth sensors, manually input by the user, and received from other types of sources. Data about a population of users and data from structured tests completed by the user can also be used to train the model. The model uses features and motifs discovered based on the data that correlate to fall risk events to update fall risk scores and predictions. The user is provided a recommendation describing how the user can reduce a predicted fall risk for the user.

Abstract: The present invention provides a system comprising a neurosynaptic processing device including multiple neurosynaptic core circuits for parallel processing, and a serial processing device including at least one processor core for serial processing. Each neurosynaptic core circuit comprises multiple electronic neurons interconnected with multiple electronic axons via a plurality of synapse devices. The system further comprises an interconnect circuit for coupling the neurosynaptic processing device with the serial processing device. The interconnect circuit enables the exchange of data packets between the neurosynaptic processing device and the serial processing device.

Abstract: Methods, computer program products, and systems are presented. The methods include, for instance: modeling for a subject process by machine learning with adaptive inputs. In one embodiment, the modeling may include: generating a model by use of machine learning with training data from measurements of successive components of a process to be modeled in order to predict measurements of a succeeding component within a statistically meaningful prediction range; adjusting the generated model by use of machine learning with less-deviation inducing measurements from a preceding component in case the measurement of the succeeding component is out of the prediction range; and presenting the adjusted model as a prediction model for the process.