Abstract: The present invention is to provide an apparatus presenting information serving as a reason for inference as information on findings so as to be easily recognized by a doctor. A medical diagnosis support apparatus infers a diagnostic name derived from a medical image based on image feature values indicating features of a region included in the medical image, acquires information on findings representing the features of the region included in the medical image based on the image feature values, and presents the information on the findings as a reason for the inference.

Abstract: An apparatus includes a die having a receiving area for a blood sample. A chamber is mounted on the die and fluidically coupled to the receiving area to process the blood sample. The chamber includes a set of electrodes that applies a voltage to the blood sample. A hematocrit level of the blood sample is determined based on a change in impedance across the set of electrodes after the voltage is applied.

Abstract: There is disclosed in one example a computing apparatus, including: a processor and a memory; and instructions encoded within the memory to instruct the processor to provide a security agent to: identify a malicious process; construct a genealogical process tree of the malicious process, the genealogical process tree including both vertical direct inheritance and horizontal indirect inheritance relationships; and terminate the malicious process and at least some related processes in the genealogical process tree.

Abstract: The present application relates to an environment control system. The environment control system is capable of detecting variations of natural environment and artificial environment, and control the use of electronic devices automatically or semi-automatically. Based on collected information stored in a built-in storage module, the environment control system may calculate and learn the user's habit of use with respect to the electronic devices using network connection.

Abstract: Disclosed are embodiments for estimating risk associated with a user of a wireless device. In some embodiments, the risk relates to a risk of infection by a contagious disease. For example, in some embodiments, the contagious disease is Coronavirus 2019. In some embodiments, locations of multiple wireless devices are estimated based on signal strengths of signals associated with the devices. Neighboring devices are identified based on highest probability regions of the devices that are determined based on associated signals. A measure of proximity to other devices is then determined based on probabilities that each device is located in neighboring regions. The risk is then based on the measure of proximity. In some embodiments, a risk of a first user associated with a first wireless device is based, in part, on a risk of a second user within a proximity of the first user.

Abstract: An image reading apparatus has: electrodes that come into contact with a medium during transport; charge detection circuits, each of which corresponds to one of the electrodes; a reader, and a controller. The controller executes inclination inference processing for calculating the inclination of the medium with respect to a transport path as an inferred inclination value according to a signal from each charge detection circuit, inclination detection processing for deriving the inclination of the medium with respect to the transport path as a detected inclination value according to the result of reading by the reader, and correction value deriving processing for storing a correction value based on the difference between the detected inclination value and the inferred inclination value in a storage section. In inclination inference processing with the correction value stored in the storage section, the inferred inclination value is calculated in consideration of the correction value.

Abstract: A method and system for classifying a vehicle based on low frequency GPS tracks. The method and system comprise retrieving a low frequency GPS track having a sampling interval of at least 20 seconds; deriving additional data from the low frequency GPS track, the additional data including interval speed and instantaneous acceleration of the vehicle; extracting a plurality of data sets from the low frequency GPS track and the additional data; generating a plurality of features from the extracted data sets; and providing the plurality of generated features to a classifier that classifies the vehicle into a predetermined class.

Abstract: In various examples, sensor data—such as masked sensor data—may be used as input to a machine learning model to determine a confidence for object to person associations. The masked sensor data may focus the machine learning model on particular regions of the image that correspond to persons, objects, or some combination thereof. In some embodiments, coordinates corresponding to persons, objects, or combinations thereof, in addition to area ratios between various regions of the image corresponding to the persons, objects, or combinations thereof, may be used to further aid the machine learning model in focusing on important regions of the image for determining the object to person associations.

Abstract: Modification of a user skill profile based upon social presence includes receiving a document indicating skill competencies of a user, and analyzing the document to determine a first model. The first model includes skill competencies indicated within the document. A social media profile associated with the user is received. The social media profile is analyzed to determine a second model. The second model includes current skills competencies of the user indicated by the social media profile. The first model and the second model is compared to determine missing skill competencies, the missing skill competencies including current skill competencies in the second model that are missing from the document. The document is modified to include the missing skill competencies.

Abstract: In an example embodiment, a method may generate an item layout map for a shelving unit, where the shelving unit stores item types associated with a product category. The method may determine product attributes associated with the product category and divider configurations of the shelving unit, where a divider configuration divides the shelving unit into one or more candidate regions. The method may generate an ontology entry associated with a product attribute and the divider configuration based on the item layout map and the divider configuration of the shelving unit, the ontology entry indicating a regional entropy values of a product attribute for the candidate regions generated by the divider configuration. In some instances, the method may generate a guidance instruction related to the product attribute of the product category based on the ontology entry and provide the guidance instruction to a user.

Abstract: Methods and systems for generating recommendations are disclosed. In some examples, from a set of positive reviews associated with a merchant offering, at least one attribute is identified and associated with the set of positive reviews, based on reviewer profiles associated with each respective positive review. The attribute is associated with the merchant offering. A match is determined between a customer attribute in a first customer profile and the at least one attribute. A set of recommendations is generated to be presented, via a customer electronic device, to a customer associated with the first customer profile, the set of recommendations including the merchant offering associated with at least one attribute.

Abstract: Systems and methods for identifying related videos based on elements tagged in the videos are presented. In an aspect, a system includes an identification component configured to identify tagged elements in a video, a matching component configured to identify other videos that include one or more of the tagged elements, and a recommendation component configured to recommend the other videos for viewing based on a current or past request to play the video.

Abstract: Systems and methods described receiving a set of example data and a set of knowledge based data and combine the set of example data and the set of knowledge based data to generate a set of combined data. The combined set can be used to train a machine learning model based on the set of combined data. The machine learning model is applied to a new set of received data for a new subject.

Abstract: A method includes generating, as executed by a processor on a computer, a plurality of topic-specific user knowledge models for each user of a plurality of users, each topic-specific user knowledge model representing a level of knowledge possessed by a respective user on a single topic from a set of globally defined topics shared among the plurality of users, generating a plurality of topic-specific expert knowledge models, each topic-specific expert knowledge model representing an aggregate level of knowledge possessed by a plurality of expert users on a single topic from a set of globally defined topics shared among the plurality of users, comparing the topic-specific user knowledge model of the first user with the topic-specific expert knowledge model for a respective topic to determine a distance between a user knowledge level and an aggregate expert knowledge level for the topic.

Abstract: Techniques for cognitively interpreting comparative statements are provided. A corpus of documents associated with treatment of a first medical disorder is received. Relative efficacies for a plurality of therapies are determined by, for at least a first respective document in the corpus of documents, identifying and extracting, using one or more natural language processing techniques, a conclusion specified within natural language text of the respective document, where the conclusion includes a comparison between two or more therapies of the plurality of therapies, determining a sentiment of the extracted conclusion, with respect to at least one therapy of the plurality of therapies, and identifying one or more cohorts of patients that are associated with the conclusion. A knowledge graph is generated based on the relative efficacies.

Abstract: A method and a device for calculating a power generation of a wind farm is provided. The method includes: determining whether a terrain complexity of a wind farm field exceeds a predetermined complexity; determining a representativeness of anemometer tower data in the wind farm field in a case where the terrain complexity exceeds the predetermined complexity; performing a mesoscale numerical simulation of a meteorological variable in the wind farm field in a case where the anemometer tower data is unrepresentative; extracting mesoscale numerical simulation data as virtual anemometer tower data; and calculating the power generation of the wind farm by using the virtual anemometer tower data.

Abstract: Systems and methods described in the present disclosure provide a software testing framework that leverages information about actual consumer interactions with a software product to create a directed graph. The directed graph is be used to generate a large number of simulated consumer interactions to test the software product in a manner such that scenarios shown to be more likely in practice are proportionally more likely to be generated during testing. Furthermore, the directed graph can be updated dynamically so that changing trends in actual consumer interactions with the software product are reflected immediately.

Abstract: An urban design pipeline automatically generates design options for an urban design project. The urban design pipeline includes a geometry engine and an evaluation engine. The geometry engine analyzes design criteria, design objectives, and design heuristics associated with the urban design project and then generates numerous candidate designs. The design criteria specify a property boundary associated with a region of land to be developed. The design objectives indicate a specific type of topology that is derived from an existing urban layout. The design heuristics include different sets of construction rules for generating designs with specific types of topologies. The geometry engine generates candidate designs that conform to the property boundary and have topological characteristics in common with the existing urban layout.

Abstract: Methods, caches, database systems and computer program products for processing database requests and retrieving results from the cache of a database system. The cache stores a representation of a data space of one or more databases of the database system, including one or more results for any database request for which the data space is able to output a result. In response to receiving a database request from a client, the cache determines by utilizing the stored representation whether the data space is able to output one or more results for the database request. If affirmative, one or more results are returned to the client from the representation of the cache. Otherwise, the cache returns an invalidity notification.

Abstract: Using a personal entity analyzer, a personal entity difference between a natural language response to a natural language query and an edited version of the natural language response is scored, the natural language response selected from a set of natural language response recommendations. Using a product entity analyzer, a product entity difference between the natural language response and the edited version of the natural language response is scored. Using a sentence similarity analyzer, a sentence similarity between the natural language response and the edited version of the natural language response is scored. Based on the personal entity difference score, the product entity difference, score and the sentence similarity score, the set of natural language responses is updated. In a natural language interaction, a selected natural language response from the set of natural language responses is outputted.

Abstract: A motion processing analysis method, system, and computer program product include selecting a first spatial graticule size, determining that a physical entity is associated with a first spatial graticule during a first timeframe and that the physical entity is associated with a second spatial graticule during a second timeframe, as a result of determining that the physical entity is associated with the first spatial graticule during the first timeframe and that the physical entity is associated with the second spatial graticule during the second timeframe, adding to a sum, determining that the sum is beyond an acceptable range, and as a result of determining that the sum beyond the acceptable range, selecting a second spatial graticule size.

Abstract: Techniques are provided for system operational analytics using additional features over time-series counters for health score computation. An exemplary method comprises: obtaining log data from data sources of a monitored system; applying a counting function to the log data to obtain time-series counters for a plurality of distinct features within the log data; applying an additional function to the time-series counters for the plurality of distinct features; and processing an output of the additional function using a machine learning model to obtain a health score for the monitored system based on the output of the additional function.

Abstract: In one example of the technology, device information associated with a device upgrade and a plurality of devices includes risk parameters including values associated with a minimum health value that is associated with a minimum acceptable number of healthy devices among the plurality of devices and a confidence value associated with a minimum acceptable probability that the number of healthy devices among the plurality of devices is at least as great as the minimum health value; and, for each device a success probability value that is associated with a probability that the device will be healthy after the device upgrade is performed on the device. A Poisson binomial distribution is iteratively used to determine a set of devices among the plurality of device for which the largest possible number of devices are included in the set of devices while meeting the risk parameters. The set of devices is then upgraded.

Abstract: In some embodiments, a method receives a set of interactions regarding interactions with an application and generates a plurality of sequences from the set of interactions. Each sequence includes multiple interactions. A plurality of results are generated for the plurality of sequences where a result in the plurality of results is a result for a sequence in the plurality of sequences. The method then trains a prediction network using the plurality of sequences and the plurality of results. The prediction network is usable to predict a next user interaction on the application when being used on a user interface.

Abstract: A method for retrieving information includes detecting a first neural activity of a user, wherein the first neural activity corresponds to a key thought of the user; in response to detecting the first neural activity, generating a search query based on speech occurring prior to detecting the first neural activity; retrieving information based on the search query; and transmitting the information to one or more output devices.

Abstract: Various aspects of the subject technology relate to systems, methods, and machine-readable media for user drawing based image search. These aspects include an image retrieval system using a convolutional neural network trained to identify how users draw semantic concepts and using an image search engine to search against images having a similar concept. The aspects include mapping between concepts of the user drawing space and concepts of the image space such that images associated with the same concept are identified. For each input user drawing, the drawing is first processed through a concept classifier to identify a corresponding concept, and then through a feature extractor to form a corresponding feature vector. The results from the concept classifier and the feature extractor may be combined to search against a collection of images having a similar concept to determine a listing of images ranked by visual and semantic similarity to the input.

Abstract: A method and system for optimizing insulin dosages for diabetic subjects which includes a processor for calculating basal and bolus dosages to be recommended for meal types including breakfast, lunch, dinner, snack, or at miscellaneous times. The bolus calculations are specifically directed to time periods which are taken from either pre-meal, post-meal, bedtime, mid-sleep or miscellaneous times. The processor calculates an optimized bolus for a specific time period and meal type based upon prior basal dosages, prior blood glucose doses, hypoglycemia thresholds, mid-point of target ranges, and subject insulin sensitivity factors. A display is provided to the subject for sensing the optimized insulin dosage recommended at a specific time period and for a specific meal type.

Abstract: Software that performs the following steps: (i) receiving data from a first database and data from a second database, (ii) identifying a training subset and a test subset from the received data; (iii) generating a first graphical model using data from the training subset; (iv) generating a second graphical model using data from the training subset; (v) determining respective weights for the first graphical model and the second graphical model by using an expectation maximization method on data from the test subset; (vi) generating a third graphical model by interpolating at least the first graphical model and the second graphical model using their respectively determined weights; and (vii) defining one or more links between the data from the first database and the data from the second database using the third graphical model.

Abstract: A communication system including a receiver to receive training data. An input interface to receive input data coupled to a hardware processor and a memory. The hardware processor is configured to initialize the privacy module using the training data. Generate a trained privacy module, by iteratively optimizing an objective function. Wherein for each iteration the objective function is computed by a combination of a distortion of the useful attributes in the transformed data and of a mutual information between the sensitive attributes and the transformed data. Such that the mutual information is estimated by the auxiliary module that maximizes a conditional likelihood of the sensitive attributes given the transformed data. Receive the input data via the input interface. Apply the trained privacy module on the input data to produce an application specific transformed data. A transmitter to transmit the application specific transformed data over a communication channel.

Abstract: Methods and systems for processing events are disclosed. According to one embodiment, input source data that includes event data of one or more events is received. One or more event definitions that match the event data are selected. For each matching event definition, the event definition is inputted into a template to generate a set of events. The template includes a number of assertions and has the event definition as one of the assertions, where each assertion includes a constraint. The constraints of the assertions are progressively processed to produce one or more solutions that are subsets of the set of events. For each constraint and each solution, a set of target events that is viable for the solution is identified, and a new solution is produced based on the solution and the identified set of target events, whereby a set of new solutions is produced.

Abstract: This disclosure relates generally to method and system for low sampling rate electrical load disaggregation. At low sampling rates, disaggregation of energy load is challenging due to unavailability of events and signatures of the constituent loads. The disclosed energy disaggregation technique receives aggregated load data from a utility meter and sequentially obtains training data for determining disaggregated energy load at low sampling rate. Dictionaries are used to characterize the different loads in terms of power values and time of operation. The obtained dictionary coefficients are treated as graph signals and graph smoothness is used for propagating the coefficients from the training phase to the test phase by formulating an optimization model. The derivation of the optimization model identifies the load of interest and estimate their power consumption based on optimization model constraints. This method achieves accuracy greater than 70% for the loads of interest at low sampling rates.

Abstract: A method of and system for converting domain names to a vector space embedding may include receiving data relating to internet traffic over a network, organizing the data into one or more documents that make up a training dataset, each of the one or more documents including one or more Domain Name Server (DNS) queries, providing the training dataset to a deep neural network to generate at least one vector space embedding for one or more domain names, and providing the vector space embedding as an output.

Abstract: The first apparatus (2100) computes an determination accuracy representing a probability that it is correct to use, as a determination result, the first result. The first apparatus (2100) outputs the first result as the determination result in a case where the determination accuracy is greater than a first threshold value, and outputs a second result as the determination result in a case where the determination accuracy is smaller than a second threshold value. In a case where the determination accuracy is equal to or greater than the second threshold value and is equal to or smaller than the first threshold value, the first apparatus (2100) requests the second apparatus (2200) to perform the predefined determination on the determination target. The first apparatus (2100) updates at least one of the first threshold value and the second threshold value based on the determination result output by the second apparatus (2200).

Abstract: Techniques for data sharing between a data miner and a data provider are provided. A set of public parameters is downloaded from the data miner. The public parameters are data miner parameters associated with a feature set of training sample data. A set of private parameters in the data provider can be replaced with the set of public parameters. The private parameters are data provider parameters associated with the feature set of training sample data. The private parameters are updated to provide a set of update results. The private parameters are updated based on a model parameter update algorithm associated with the data provider. The update results is uploaded to the data miner.

Abstract: An alarm management system includes a plurality of data sources configured to provide data relating to an event at a building and a processing circuit. The processing circuit is configured to receive the data from the data sources, apply a Bayesian network model to the data to generate a probability score for each of a plurality of possible causes of the event, identify the most probable cause of the event as a first cause of the plurality of possible causes having a highest probability score, execute an action with respect to the identified most probable cause of the event.

Abstract: Systems, methods and apparatuses are disclosed to efficiently and accurately scan a plurality of documents located in any number of unstructured data sources. Preprocessed metadata is generated for each document and metadata features are determined based on the preprocessed metadata. A trained machine learning system may utilize the metadata features to predict whether each of the documents contains personal information, without requiring any information relating to the content of such documents.

Abstract: Methods and systems are described for controlling parameters in a building. According to at least one embodiment, a method for controlling a building system includes using at least one sensor to detect occupancy in a building over time, determining a predictive schedule based on the occupancy detected with the at least one sensor, and associate real time events that occur simultaneously with an occupancy state of the predictive schedule.

Abstract: The technology disclosed determines which field values in a set of unique field values for a particular field in a fielded dataset are anomalous using six similarity measures. A factor vector is generated per similarity measure and combined to form an input matrix. A convolutional neural network processes the input matrix to generate evaluation vectors. A fully-connected network evaluates the evaluation vectors to generate an anomaly scalar for a particular unique field value. Thresholding is applied to anomaly scalar to determine whether the particular unique field value is anomalous.

Abstract: A method and apparatus for generating information. A specific embodiment of the method includes: acquiring a set of geographic information point sequences and a set of identifiers comprising an identifier of each geographic information point sequence in the set of the geographic information point sequences; for the each geographic information point sequence in the set of the geographic information point sequences, clustering geographic information points in the each geographic information point sequence and generating an element sequence corresponding to the each geographic information point sequence; learning, by utilizing a machine learning method based on the set of identifiers and the generated element sequences, to obtain a matrix for the identifiers in the set of identifiers; and generating, for each identifier in the set of identifiers, based on the identifier and the matrix, information of a user to which the geographic information point sequence indicated by the identifier belongs.

Abstract: The technology disclosed determines one or more field values in a set of field values for a particular field in a fielded dataset that are similar to an input value using six similarity measures. A factor vector is generated per similarity measure and combined to form an input matrix. A convolutional neural network processes the input matrix to generate evaluation vectors. A fully-connected network evaluates the evaluation vectors to generate suggestion scalars for similarity to a particular input value. Thresholding is applied to suggestions scalars to determine one or more suggestion candidates for the particular input value.

Abstract: Detecting patterns and sequences associated with an anomaly in predictions made a predictive system. The predictive system makes predictions by learning spatial patterns and temporal sequences in an input data that change over time. As the input data is received, the predictive system generates a series of predictions based on the input data. Each prediction is compared with corresponding actual value or state. If the prediction does not match or deviates significantly from the actual value or state, an anomaly is identified for further analysis. A corresponding state or a series of states of the predictive system before or at the time of prediction are associated with the anomaly and stored. The anomaly can be detected by monitoring whether the predictive system is placed in the state or states that is the same or similar to the stored state or states.

Abstract: Recommendation information is associated with an information classification label by a server of an online interaction platform, and a correspondence between the recommendation information and the information classification label is stored in a feature label database. The recommendation information associated with the information classification label is delivered to a user. Upon triggering of the recommendation information by a user, the information classification label associated with the recommendation information and an unique identifier associated with the user are obtained. By using the unique identifier, a user feature database is queried to determine whether a correspondence between the information classification label and the unique identifier has been established. If it is determining that the correspondence is not established, the correspondence between the information classification label and the unique identifier is established in the user feature database.

Abstract: Free-form discovery of differential equations for modeling a physical system or process under investigation that includes defining a formal language sentence grammar that describes admissible relationships between quantities, and incorporating the use of such grammar for free-(functional) form, automatic discovery of differential equations. The method requires minimum knowledge of the desired differential equation and is universally applicable to ordinary and partial differential equations.

Abstract: Image correct data is received from a plurality of terminals that belong to a plurality of organizations, a learning-terminated discriminator that is a learning discriminator that has performed learning using the image correct solution data is obtained, distribution of resources capable of being used by each discriminator is determined in accordance with the number of pieces of the received image correct solution data or the degree of performance improvement of the learning-terminated discriminator, and a discrimination result “Output” obtained by performing discrimination of a discrimination target image “Input” received from the terminal using the determined resources of the distribution is transmitted to the terminal that is a transmission source.

Abstract: A method for processing fusion data and an information recommendation system are provided. The method includes the follows. The information recommendation system obtains a plurality of heterogeneous data sources. The information recommendation system obtains a unified evaluation index of each heterogeneous data source, by processing the plurality of heterogeneous data sources through the value-model set. The information recommendation system obtains a loss quantitative evaluation index of the fusion-strategy model. The information recommendation system obtains a fusion sorting recommendation result of the plurality of heterogeneous data sources, by inputting the unified evaluation index of each heterogeneous data source and the loss quantitative evaluation index to the fusion-strategy model.

Abstract: Techniques are disclosed for detecting knowledge based on brainwave response to external stimulation. A subject can be exposed to stimuli and brainwave responses indicating a p-300 signal can be detected. Further stimuli or sequences of stimuli can be selected be presented to the subject based on the category correlated with the stimuli that indicate a p-300 response.

Abstract: Methods, computer program products, and systems are presented. The method computer program products, and systems can include, for instance: obtaining during an interactive user support session user transmitted data including conversation content from a client computing device associated to the user; subjecting the user transmitted data to processing to provide feedback data associated to the conversation content obtained from the client computer device, wherein the feedback data indicates a level of satisfaction of the user, and wherein the processing includes subjecting content of the conversation content to natural language processing to return sentiment data of the user; examining data of the feedback data; performing selecting of a conversation strategy in dependence on the examining; and sending for presentment to the user returned conversation content in accordance with the certain conversation strategy.

Abstract: An Internet-of-things device can detect, using one or more sensors, a condition exceeding a predefined condition threshold. If the condition is occurring at a location outside of an area monitored by the sensors of the Internet-of-things device, a communication device can transmit an investigation request to an unmanned aircraft. The investigation request can include a request to monitor an environment at the location. The Internet-of-things device can then receive an event report identifying whether a class of event is or was occurring at the location from the unmanned aircraft. The event report can include one or more event labels. The Internet-of-things device can transform, with an artificial intelligence engine, the one or more event labels into machine learned knowledge in an Internet-of-things knowledge domain.

Abstract: Embodiments for feature engineering by one or more processors are described. A plurality of transformations are applied to a set of features in each of a plurality of datasets. An output of each of the plurality of transformations is a score. For each of the sets of features, selecting those of the plurality of transformations for which said score is above a predetermined threshold. A signal representative of said selection is generated.

Abstract: Aspects of the present disclosure describe systems, methods, and devices for vehicle mode detection based on acceleration data and location data during a trip detected by mobile devices in a vehicle. According to some examples, location data will be compared to known routeway data to detect a vehicle mode. In other examples acceleration data will be used to calculate the consistency of vehicular velocity during a trip to detect a vehicle mode. In some additional aspects, multiple analyses will be performed on the same data sets to check for false positives with vehicle mode detection.