Abstract: A break recommendation method, system, and non-transitory computer readable medium, include detecting a deviation between a current cognitive state of the user and a past cognitive state of the user during a predetermined amount of time for a document type based on a change in an eye gaze movement and a facial and emotional expression and recommending that the user takes a break from viewing the document for a predetermined amount of time based on the deviation being greater than a predetermined threshold value, where the deviation is related to the user viewing the document and the document type of the document.

Abstract: A method, system, and computer program product, include tracking eye-reaction of a user each of a plurality of advertisements of an advertisement group on a web page displayed on a display device, calculating an interest of the user for each advertisement of the advertisement group over time based on the tracked eye-reaction, and to assign each advertisement a user engagement value based on the interest, conducting an auction to determine winning advertisements based on advertiser parameters received from a database and the user engagement value of each of the plurality of advertisements, determining at least one time at which to cause the conducting to conduct the auction, and dynamically repopulating the advertisement group with the winning advertisements.

Abstract: A blockchain of transactions may be referenced for various purposes and may be later accessed by interested parties for ledger verification and information retrieval. One example method of operation may include identifying one or more analytic processes to process blockchain data, determining a primary type of data analytic to be performed by the one or more analytic processes, selecting a type of data store to use for performing the one or more data analytic processes based on the primary type of data analytic, accessing the blockchain data, applying the one or more analytic processes, and storing results of the applied analytic processes in a database, file or dashboard. The analytic data may be realized in any manner or preference requested.

Abstract: Techniques that facilitate machine learning using multi-dimensional time series data are provided. In one example, a system includes a snapshot component and a machine learning component. The snapshot component generates a first sequence of multi-dimensional time series data and a second sequence of multi-dimensional time series data from multi-dimensional time series data associated with at least two different data types generated by a data system over a consecutive period of time. The machine learning component that analyzes the first sequence of multi-dimensional time series data and the second sequence of multi-dimensional time series data using a convolutional neural network system to predict an event associated with the multi-dimensional time series data.

Abstract: A blockchain of transactions may be referenced for various purposes and may be later accessed by interested parties for ledger verification and information retrieval. One example method of operation may include extracting obligation data from one or more of documents and rules, identifying transaction data from one or more blockchain transactions, creating one or more transaction contexts based on the transaction data, comparing the obligation data to the transaction data to identify whether the one or more transaction contexts are subject to the obligation data, and creating a notification including a current compliance status between the transaction contexts and the obligation data.

Abstract: An advertisement remarketing method, system, and non-transitory computer readable medium, include a gaze tracking circuit configured to track a gaze of a user relative to a plurality of advertisements displayed on content-serving devices to determine a focus level of the user with respect to each of the plurality of advertisements, a list creating circuit configured to create a plurality of ranked lists ranking the focus level of the user for each of the plurality of advertisements, each of the plurality of ranked lists corresponding to a different content-serving device of the content-serving devices, and a remarketing circuit configured to remarket advertisements to the user on a user device based on the plurality of ranked lists.

Abstract: Aspects of the present disclosure describe techniques for detecting anomalous data in an encrypted data set. An example method generally includes receiving a data set of encrypted data points. A tree data structure having a number of levels is generated for the data set. Each level of the tree data structure generally corresponds to a feature of the encrypted plurality of features, and each node in the tree data structure at a given level represents a probability distribution of a likelihood that each data point is less than or greater than a split value determined for a given feature. An encrypted data point is received for analysis, and anomaly score is calculated based on a probability identified for each of the plurality of encrypted features. Based on determining that the calculated anomaly score exceeds a threshold value, the encrypted data point is identified as potentially anomalous.

Abstract: A venue occupant comfort system, comprises a processor that stores computer executable components stored in memory. A plurality of sensors sense ambient conditions associated with exterior and interior conditions of a venue. A context component infers or determines context of an occupant of the venue. A crowd estimation component infers, based at least in part on mining social networks, size of crowd expected at the venue. A comfort model component implicitly and explicitly trained on occupant comfort related data analyzes information from the plurality of sensors, the crowd estimation component and context component. A comfort controller adjusts environmental conditions of the venue based at least in part on output of the comfort model component. The adjustments to venue environment can optionally be differentiated by zone.

Abstract: Aspects of the present invention disclose a method, executed by a computer, for generating retailing analytics. The method includes receiving a spatial map corresponding to an environment. The method further includes receiving, from a purchasable item comprising one or more textile-integrated sensors, sensor information The purchasable item is located in the environment. The method further includes analyzing the sensor information to produce retail analytics corresponding to the purchasable item. The method further includes and sending the retail analytics to a client device. A computer program product and computer system corresponding to the above method are also disclosed herein.

Abstract: A flight ticket transfer system obtains an identification of a traveler and of an earlier booked flight, provides a listing of earlier departing flights, receives a selection of an earlier departing flight from among the listing of the earlier departing flights, identifies an earlier departing booked airline corresponding to the earlier departing booked flight, and sends to the earlier departing booked airline a flight luggage request, receives from a transfer feasibility module a success response to the luggage request, relative to an estimated time for transfer of the traveler's luggage from the earlier booked flight to the selected earlier departing flight, and completes a purchase of a ticket for the selected earlier departing flight.

Abstract: A conversation analysis system has a local system configured to capture conversation data associated with a conversation between an individual and a user and a backend system electronically connected to the local system. The backend system has a suggestion system configured to receive the conversation data from the local system and extract, by an analysis unit, information from the conversation data indicative of a mood of the individual. The suggestion system is further configured to classify, by a classifier unit, the individual based on the extracted information, including determining a current classification and determine, by a suggestion unit, one or more rules based at least on the current classification and stored information in a knowledge base. The suggestion system is further configured to provide, by the suggestion unit, at least one of an alert or suggestion to the user based on the one or more rules.

Abstract: Techniques that facilitate machine learning using multi-dimensional time series data are provided. In one example, a system includes a snapshot component and a machine learning component. The snapshot component generates a first sequence of multi-dimensional time series data and a second sequence of multi-dimensional time series data from multi-dimensional time series data associated with at least two different data types generated by a data system over a consecutive period of time. The machine learning component that analyzes the first sequence of multi-dimensional time series data and the second sequence of multi-dimensional time series data using a convolutional neural network system to predict an event associated with the multi-dimensional time series data.

Abstract: Techniques that facilitate machine learning using multi-dimensional time series data are provided. In one example, a system includes a snapshot component and a machine learning component. The snapshot component generates a first sequence of multi-dimensional time series data and a second sequence of multi-dimensional time series data from multi-dimensional time series data associated with at least two different data types generated by a data system over a consecutive period of time. The machine learning component that analyzes the first sequence of multi-dimensional time series data and the second sequence of multi-dimensional time series data using a convolutional neural network system to predict an event associated with the multi-dimensional time series data.

Abstract: An example operation may include one or more of generating, by a plurality of training participant clients, gradient calculations for machine learning model training, each of the training participant clients comprising a training dataset, converting, by a training aggregator coupled to the plurality of training participant clients, the gradient calculations to a plurality of transaction proposals, receiving, by one or more endorser nodes or peers of a blockchain network, the plurality of transaction proposals, executing, by each of the endorser nodes or peers, a verify gradient smart contract, and providing endorsements corresponding to the plurality of transaction proposals to the training aggregator.

Abstract: An example operation may include one or more of generating, by a training participant client comprising a training dataset, a plurality of transaction proposals that each correspond to a training iteration for machine learning model training related to stochastic gradient descent, the machine learning model training comprising a plurality of training iterations, the transaction proposals comprising a gradient calculation performed by the training participant client, a batch from the private dataset, a loss function, and an original model parameter, receiving, by one or more endorser nodes of peers of a blockchain network, the plurality of transaction proposals, and evaluating each transaction proposal.

Abstract: An example operation may include one or more of generating, by a training participant client, a plurality of transaction proposals, each of the plurality of transaction proposals corresponding to a training iteration for machine learning model training related to stochastic gradient descent, the machine learning model training comprising a plurality of training iterations, the transaction proposals comprising a gradient calculation performed by the training participant client, transferring the plurality of transaction proposals to one or more endorser nodes or peers each comprising a verify gradient smart contract, executing, by each of the endorser nodes or peers, the verify gradient smart contract; and providing endorsements corresponding to the plurality of transaction proposals to the training participation client.

Abstract: A gaze tracking system, method, and computer product for tracking an eye gaze on a screen of a device including a single monocular camera, the system including measuring a rotation of a hinged plane of a display screen with respect to the eye gaze, combining the rotation with a three-dimensional movement of the camera, a position of the camera being constant with respect to the display screen, and estimating a point of gaze localization on the display screen using the single monocular camera as the input, in absence of a sensor, and without performing a display screen calibration.

Abstract: A location-based advertisement method, system, and non-transitory computer readable medium, include a location-based advertisement system, including a gaze tracking circuit configured to track a gaze of a user relative to a plurality of advertisements displayed on content-serving devices to determine a focus level of the user with respect to each of the plurality of advertisements, a location detecting circuit configured to detect when a user is within a predetermined proximity of a location of a business having products related to the plurality of advertisements displayed on the content-serving devices, and an advertisement sending circuit configured to send a notification to the user based on the user being within the predetermined proximity of the location of the business and the plurality of ranked lists ranking the focus level of the user.

Abstract: A system and method utilizing a mass user gaze fixation behavior with respect to products and product racks (using a camera-based inputs and gaze-tracking algorithms) and a financial profitability of each product/product type; generates gaze-based heat maps for indicating blind spots and eye gaze points at different positions of the racks/shelves and indicating product placement positions on the heat map; and advises product placement in the retail store for optimizing financial profitability by performing an iterative refinement of product placements over a period for the product/similar products. The system translates mass eye gaze of retail store customers into financial profitability via product placement optimization, and iteratively runs a method to optimize product placements, and generate gaze-based heat maps and perform blind spot analysis for avoiding product placements at certain locations and placing products at other locations, e.g., by running a decision module on consumer gaze pattern data.

Abstract: A humanness validation method, system, and computer program product, include generating a sequence of actions including gaze actions and emotional state actions, instructing an access requestor to perform the actions in the sequence one-by-one, and granting access to the access requestor if a total error rate of each of the performed one-by-one actions in the sequence of actions is less than a predefined threshold value.