Abstract: A computer-implemented method includes extracting, by one or more processors of one or more computing devices, a product family name from each of a plurality of unstructured product titles associated with a plurality of products. The method further includes determining, by the one or more processors, a degree of similarity between model numbers of the plurality of products. The method further includes determining, by the one or more processors, that at least two of the plurality of products are variants of one another by determining that the at least two of the plurality of products have a same extracted product family name and determining that the degree of similarity between the model numbers of the plurality of products is above a predetermined threshold.

Abstract: A set of material candidates expected to yield materials with target properties can be generated. A subject matter expert's decision indicating accepted and rejected material candidates from the set of material candidates can be received. Based on the subject matter expert's input, a machine learning model can be trained to replicate the subject matter expert's decision.

Abstract: Systems and methods for deriving classification rules from documents and a database using rule-based machine learning. The method includes extracting first variables from documents corresponding to an organization. The method further includes extracting second variables from a database corresponding to the organization. The method also includes filtering the extracted second variables based on at least one of null values, repeat variables, location variables, ID variables, or data variables. The method further includes deriving first classification rules based on the first variables using a rule-based machine learning algorithm. The method also includes calculating an accuracy of the derived first classification rules. The method also includes deriving second classification rules based on the first variables and the filtered second variables. The method further includes determining a suggested additional variable based on the derived second classification rules and the calculated accuracy.

Abstract: Systems and methods for collecting, analyzing, billing and reporting data from intelligent electronic devices are provided. Also, systems and methods for managing sensor data are provided. In some embodiments, a system for managing sensor data may include intelligent electronic devices, a server, a plurality of client devices, and a network. Each of the intelligent electronic devices is configured to obtain sensor data related to power parameters distributed to a load. The server is configured to receive the sensor data from the plurality of intelligent electronic devices and store the sensor data in a database. Each client device is configured to retrieve the sensor data from the database. The network enables communication among the server, the plurality of intelligent electronic devices, and the plurality of client devices.

Abstract: In an approach to automatically downsampling DNA sequence data using variational autoencoders and preserving genomic integrity of an original file embodiments execute, by an encoder, bootstrapping on genomic sequence data to produce resamples. Furthermore, embodiments assess, by the encoder, unrepresentativeness and self-inconsistency of the resamples and selecting a representative resample according to the assessment, and build, by a modified encoder, vector representations from genotype likelihoods based on the selected representative sample. Additionally, embodiments integrate, by an analytics engine, mapping positional information and the genotype likelihoods to identify an optimum vector representation of a resample, and decode, by a modified decoder, the identified optimum vector representation of the resample to obtain a down-sampled read file that resembles and maintains the genomic integrity of the original file.

Abstract: Embodiments of the disclosure provide methods and systems for processing a neural network associated with an input matrix having a first number of elements. The method can include: dividing the input matrix into a plurality of vectors, each vector having a second number of elements; grouping the plurality of vectors into a first group of vectors and a second group of vectors; and pruning the first group of vectors and the second group of vectors.

Abstract: A non-transitory computer-readable storage medium storing a program that causes a computer to execute a process, the process includes obtaining a machine learning model having learned characteristic amounts of a plurality of training data including an objective function; calculating similarities between the characteristic amounts of the plurality of training data by inputting the plurality of training data to the obtained machine learning model; specifying a data group having a high similarity with a desired objective function from the characteristic amounts of the plurality of training data based on distances of the calculated similarities; and acquiring an optimum solution for the desired objective function by using the specified data group.

Abstract: In some implementations initially training a first neural network includes inputting the training inputs and corresponding training labels into the first neural network to produce output labels, comparing the output labels to the corresponding training labels using a second neural network that learns and applies a comparison metric, and adjusting parameters of the first neural network based on the comparing. The device then inputs additional inputs into the first neural network to produce additional output labels and corresponding confidence values from the second neural network. The device selects, based on the confidence values, an automatically-labeled training set of data including a subset of the additional inputs and a corresponding subset of the additional output labels. During a second training stage, the device trains the first neural network and the second neural network using the automatically-labeled training set of data.

Abstract: A policy improvement method of improving a policy of reinforcement learning based on a state value function is performed by a computer. The method causes a computer to execute a process including: calculating an input to a control target based on the policy and a predetermined exploration method of exploring for an input to the control target in the reinforcement learning; and updating a parameter of the policy based on a result of applying the calculated input to the control target, using the input to the control target and a generalized inverse matrix regarding a state of the control target.

Abstract: An event processing system for processing events in an event stream is disclosed. The system can launch a first CQL engine in a cluster of CQL engines using a CQL engine tracking engine. The system can schedule, using the CQL engine tracking engine, the first CQL engine to process a batch of a continuous stream of input events related to an application. The system can track, using the CQL engine tracking engine, the first CQL engine to be scheduled for execution. The system can then execute, using the CQL engine tracking engine, the first CQL engine to process the batch of the continuous stream of input events to generate a set of output events related to the application.

Abstract: A method for performing an automated medical diagnosis of a patient and for providing and utilizing a smart electronic medical record to assist in confirming diagnosis of a patent. The diagnosis will be based upon a patient's signs/symptoms/findings which are input by a user/heath care provider and then further analyzed to then output a listing of high probability differential diagnoses/diseases and a low probability differential diagnoses/diseases that the patient may have. This analysis assists health care providers in providing a fast and early indication of potential conditions based upon an analysis of a patient's signs/symptoms/findings.

Abstract: A neural network structure is separated into an odd neural network including only the odd layers and an even neural network including only the even layers. In order to allow for parallel execution, for forward propagation a second input is generated from the original input, while for backward propagation a second error gradient is generated. Parallel execution may accelerate the forward and backward propagation operations without significant change in accuracy of the model. Additionally, restructuring a single neural network into two or more parallel neural networks may reduce the total time needed for training.

Abstract: An audio visual haptic signal reconstruction method includes first utilizing a large-scale audio-visual database stored in a central cloud to learn knowledge, and transferring same to an edge node; then combining, by means of the edge node, a received audio-visual signal with knowledge in the central cloud, and fully mining semantic correlation and consistency between modals; and finally fusing the semantic features of the obtained audio and video signals and inputting the semantic features to a haptic generation network, thereby realizing the reconstruction of the haptic signal. The method effectively solves the problems that the number of audio and video signals of a multi-modal dataset is insufficient, and semantic tags cannot be added to all the audio-visual signals in a training dataset by means of manual annotation. Also, the semantic association between heterogeneous data of different modals are better mined, and the heterogeneity gap between modals are eliminated.

Abstract: Described herein is a method, system, and non-transitory computer readable medium for preventing data loss from both producer and consumer systems in continuous availability event-driven applications. The process for preventing data loss may replicate events at the broker level, and selectively receive replica events at the consumer level to account for the case the originally sent event is not received, while conserving computer and network resources. Alternatively, events and replica events may be received in duplicate. In either mode of reception, machine-learning may be used for implementing algorithms which further help to conserve resources and aid in preventing further data loss.

Abstract: Systems and methods including one or more processors and one or more non-transitory storage devices storing computing instructions configured to run on the one or more processors and perform acts of receiving, from an extraction, transform, load (ETL) application, a matrix, which can comprise raw feature data; storing the matrix comprising the raw feature data in a standard format in the one or more non-transitory computer readable storage devices; receiving a configuration file over a computer network; storing the configuration file in a standard format in the one or more non-transitory computer readable storage devices; instantiating one or more unifier applications based upon the configuration file; identifying relevant feature data of the raw feature data; storing the relevant feature data in a standardized format in an output file in the one or more non-transitory computer-readable storage devices; and transmitting, over the computer network in real time to a model building system, the output file compri

Abstract: Systems and methods are provided for using an algorithm and data structure for efficient and accurate classification of data items into recovery classes. When a target recovery time (TRT) is specified for a data set, a system may obtain version metadata regarding data items in the data set. The metadata may be obtained in reverse chronological order such that the latest record representing a version or other storage operation is first, followed by the second latest record, and so on. The system may use a bidirectional doubly linked list to efficiently store version data for a particular data item in memory. As version metadata records are read and added to the data structure in reverse chronological order, classification determinations may be triggered when certain conditions are met.

Abstract: Systems and methods are described herein for detecting that a person of interest is displayed in a live video for a threshold amount of time, attempting to identify an identity of the person determining that the attempting has failed, identifying broadcast times corresponding to a period of time in which the person was displayed in the live video, identifying trending topics in a social media space that relate to the live video, ascertaining an identity of the person of interest based on the trending topics, comparing the identity of the person of interest to a database of social media profiles, identifying a social media component that corresponds to the person, determining whether the person has modified the social media component within a predefined period of time, and providing information relating to a modification of the social media component simultaneously with the video.

Abstract: Methods and systems to identify video content based on video fingerprint matching are described. In some example embodiments, the methods and systems generate a query fingerprint of a frame of video content captured at a client device, query a database of reference fingerprints, determine the query fingerprint of the frame of captured video content matches a reference fingerprint, and identify the video content based on the match of fingerprints.

Abstract: A system for extracting key phrase candidates from a corpus of documents, including a processor, a memory, and a program executing on the processor. The system is configured to run a key phrase model to extract one or more key phrase candidates from each document in the corpus and convert each extracted key phrase candidate into a feature vector. The key phrase model also filters the feature vectors to remove duplicates using a classifier that was trained on a set of key phrase pairs with manual labels indicating whether two key phrases are duplicates of each other, to produce remaining key phrase candidates. The system uses the remaining key phrase candidates in a computer-implemented application.

Abstract: Herein are techniques that train regressor(s) to predict how effective would a machine learning model (MLM) be if trained with new hyperparameters and/or dataset. In an embodiment, for each training dataset, a computer derives, from the dataset, values for dataset metafeatures. The computer performs, for each hyperparameters configuration (HC) of a MLM, including landmark HCs: configuring the MLM based on the HC, training the MLM based on the dataset, and obtaining an empirical quality score that indicates how effective was said training the MLM when configured with the HC. A performance tuple is generated that contains: the HC, the values for the dataset metafeatures, the empirical quality score and, for each landmark configuration, the empirical quality score of the landmark configuration and/or the landmark configuration itself. Based on the performance tuples, a regressor is trained to predict an estimated quality score based on a given dataset and a given HC.