Abstract: Using a natural language analysis, it is determined that a compendium requires a natural language response to a natural language query, the compendium comprising a set of stored natural language responses to natural language queries. A relevance of a portion of narrative text to the natural language query is scored according to a query relevance measure, the portion extracted from a corpus of narrative text. The compendium is enhanced according to the query relevance score with information in the portion.

Abstract: A vehicle recommendation device and a method for machine learning based recommendation of vehicle is provided. The vehicle recommendation device receives vehicle log data from a plurality of sensors associated with each of a first set of vehicles. The first set of vehicles are classified based on a set of vehicle types. The vehicle recommendation device generates vehicle trip data, associated with a plurality of trips of each of the first set of vehicles, based on the received vehicle log data. The vehicle recommendation device further determines a set of features, associated with the plurality of trips or with information about the first set of vehicles, based on the generated vehicle trip data. The vehicle recommendation device further generates a machine learning model which is trained based on the determined set of features to output a first type of vehicle from the set of vehicle types.

Abstract: Systems and methods for that allow for the creation and management of a centralized gift registry are described. Gifts are added to the centralized gift registry from retailer-specific registries. A gift registrant may select one or more retailers. The centralized gift registry system may automatically submit a request for the creation of retailer-specific registries at each of the retailers selected by the gift registrant, based in part on information provided to the centralized gift registry by the gift registrant. This creation may involve the use of APIs at the centralized gift registry or the retailer-specific registries. A gift list is created and displayed. Gift givers may select a gift from the centralized gift registry.

Abstract: A computer-implemented method optimizing genetic algorithms for finding solutions to a provided problem is described. The method implements a multi-arm bandit algorithm to determine performance scores for candidate individuals from a candidate pool in dependence on successes and failures of the one or more candidates. The method evolves the candidate individuals in the candidate pool by performing evolution steps including: determining a fitness score for each of the candidate individuals in the candidate pool in dependence on the performance scores for the candidate individuals, discarding candidate individuals from the candidate pool in dependence upon their assigned performance measure, and adding, to the candidate pool, a new candidate individual procreated from candidate individuals remaining in the candidate pool after the discarding of the candidate individuals.

Abstract: A system for assessing text content of a product. The system includes a computing device having a processor and a storage device storing computer executable code. The computer executable code, when executed at the processor, is configured to: provide text contents and confounding features of products; train a first regression model using the text content and the confounding features of the products; train the second regression model using the confounding features; operate the first regression model using the text contents and the confounding features to obtain a total loss; operate the second regression model using the confounding features of to obtain a partial loss; subtract the total loss from the partial loss to obtain a residual loss; use the residual loss to evaluate models and parameters for the regression models; and use the first regression model to obtain log odds of the words indicating importance of the words.

Abstract: Media content may be recommended based on feedback weightings. Input data describing the presentation of a media content items in association with content management accounts may be represented as data points. Each data point may identify feedback data for a media content item previously recommended for presentation in association with a content management account. The feedback data may identify a viewer reaction to the recommended media content item. A weighting factor based on the feedback data for the media content item presented in association with the content management account may be applied to produce a recommendation.

Abstract: A procedure and an installation for the control of home appliances that uses only the data that is necessary, avoiding saturation of the memories of the elements involved and/or the communications bandwidth. The procedure is simple and reliable so that it does not require the intervention of an expert user or a service person. The procedure includes the standardization of the data identifiers of each home appliance by identifying the identifier that contains a value for each data point and checking that the scale of each data point is equivalent to a predetermined scale. The installation includes a standardization system for such purpose.

Abstract: Embodiments include methods and systems to remotely and electronically enhance measurement accuracy and speed of electronic brassiere fitting data. A method according to an embodiment can include receiving a set of current user brassiere data from a remote user computing device, receiving a set of physiological images from the remote user computing device, and receiving brassiere fitting data from one or more data sources. The method can further include filtering the brassiere fitting data responsive to the set of current user brassiere data, and generating a first brassiere size responsive to filtering the brassiere fitting data. The method can further include modifying the first brassiere size responsive to the set of physiological images, and generating a recommended brassiere size responsive to modifying the first brassiere size.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for content recommendation using neural networks. In One aspect, a method includes: receiving context information for an action recommendation from multiple possible actions; processing the context information using a neural network that includes Bayesian neural network layers to generate, for each of the actions, one or more parameters of a distribution over possible action scores for the action, where each parameter for each Bayesian layer is associated with data representing a probability distribution over multiple possible current values for the parameter; for each parameter of each Bayesian neural network layer, selecting the current value for the parameter using data representing probability distribution over possible current values for the parameter; and selecting an action from multiple possible actions using the parameters of the distributions over the possible action scores for the action.

Abstract: Methods and devices for determining a load vector on an object are disclosed herein. An example method includes collecting location observations related to the object. The example method further includes filtering the location observations to determine an estimated model path. The example method further includes outputting a set of data from the estimated model path, wherein the set of data includes a model location, a model velocity, a model acceleration, and a model jerk. The example method further includes calculating a load vector from the set of data, scaling the load vector via a scaling index, and transmitting the scaled load vector to a remote device.

Abstract: A vehicle includes a system for recording and analyzing user preferences during a vehicle test drive. The system includes a processor and a memory that stores processor-readable instructions. When executed by the processor, the processor-readable instructions cause the system to: determine a user identity; create a user preference profile based on the user identity; communicate a prerecorded message describing one or more vehicle options to the user; record a user response of the user in response to the prerecorded message; analyze the recorded user response; and update the user preference profile based on the analysis of the recorded user response.

Abstract: An intelligent store system and an intelligent store method are provided. The method includes: sensing, by a proximity sensor, whether there is a first customer approaching the proximity sensor; sensing, by an image sensor, a movement trajectory of the first customer; and determining whether the first customer is entering or leaving a store in response to the first customer approaching the proximity sensor sensed by the proximity sensor. If the first customer is entering the store, a first virtual identity and a first virtual shopping cart corresponding to the first virtual identity are generated for the first customer, and a first product list of the first virtual shopping cart is updated according to the movement trajectory. If the first customer is leaving the store, a first operation of the first customer is received by a human machine interface, and the first product list according to the first operation is updated.

Abstract: A computer-implemented method of digital information presentation may include receiving a selection of an item reference from among digital data via a digital user interface. The method may further include receiving a request to search for information relating to the item reference. The method may include sending a request to search for the information relating to the item reference in the digital database. The method may also include receiving the information relating to the item reference. The method may further include presenting the information relating to the item reference via the digital user interface, the digital user interface including a purchase information element. The method may include facilitating a purchase of the item via the digital user interface.

Abstract: A skin care computing device that includes a user interface, a camera, a speaker, a communication interface, and a personal care assistant application. The personal care assistant application can identify personal care products being used by a user and product use event data describing the use of a personal care product by the user, such as the date and time of the use, the duration and frequency of the use, and the manner in which the personal care product is being used. Then the personal care assistant application generates user feedback information for the personal care product based on the product use event data, previous instances of product use event data, and/or user profile data for the user.

Abstract: Top attributes of a product are identified from text content of an electronic document and are presented as visual depictions of the attributes. The attributes are identified and ranked based on the text content, such as whether the text content describes the attribute in a positive or negative manner. Top attributes are selected from the ranking and a visual depiction of the top attributes is identified. The visual depiction can be a universally recognized symbol or a custom image designated to an attribute. The visual depiction is superimposed on the text content associated with the attributes when the electronic document is presented at a display. In this way, top attributes for a product can be presented in a manner that condenses the relevant information provided by the electronic document into a format better suited for providing the information on small screens of mobile devices.

Abstract: A food product recommendation system may include a camera configured to obtain an image of a food product area having available food products and a remote device associated with a given user. The system may also include a recommendation server configured to cooperate with the remote device to obtain a desired food product characteristic from the given user and determine recommended food products from the available food products based upon the desired food product characteristic. The recommendation server may also be configured to associate a corresponding food product identifier to each recommended food product and communicate the recommended food products to the remote device. The remote device may display the image of the food product area with the recommended food products augmented with the corresponding food product identifiers and obtain user selection for purchase of at least one of the recommended food products based upon the corresponding food product identifier.

Abstract: A product selecting unit extracts a candidate for recommended product on the basis of browsing information in a browsing information storing unit and unpurchased product information in an unpurchased product information storing unit. Product specifications that are information specific to each product are stored in a product information storing unit. A specification correlation calculating unit extracts a product having a product specification correlated with a product specification of the candidate for recommended product on the basis of the product specifications, and extracts the product as a recommended product.

Abstract: A method, computer program product, and computing system for associating one or more fashion products on a website with a user accessing the website. One or more recommendations may be provided to the user for fashion products based upon, at least in part, one or more fashion-ability scores representative of the one or more fashion products associated with the user on the website and one or more fashion-ability scores representative of the one or more fashion products on the website.

Abstract: Computer-implemented methods of providing personalized recommendations to a user of items available in an online system, and related systems. First-level features including context features are computed based upon context data. A first-level machine learning model is then evaluated using the first-level features to generate predictions of user behavior in relation to a plurality of individual items available via the online system. A list of proposed item recommendations is constructed based upon the predictions. Second-level features are computed based upon the context data and list features based upon the list of proposed item recommendations and the corresponding predictions generated by the first-level machine learning model. A second-level machine learning model is evaluated using the second-level features to generate a prediction of user behavior in relation to the list of proposed item recommendations.

Abstract: A computer device determines requirements of a particular device/service by analyzing designated electronic communications. The computer device then monitors performance of the device/service over a period of time to determine performance of the device/service in meeting those derived requirements. An output regarding the performance of the device/service in meeting those requirements is then generated.

Abstract: Examples described herein relate to a system consistent with the disclosure. For instance, the system may comprise a data lake including information relating to an in-store activity of a customer and an online activity of the customer, a processing resource, and a non-transitory machine-readable medium storing instructions executable by the processing resource to identify the in-store activity and the online activity of the customer, aggregate and store the in-store activity and the online activity of the customer in the data lake, reduce an amount of products in a product portfolio, compare purchases including in-store purchases and online purchases, and recommend a product of the plurality of products based on the comparison.

Abstract: Embodiments of the invention make consumer application adoption more efficient by giving suppliers access to the desired target audience by displaying the applications to the right audience. Suppliers can provide criteria for the kinds of users they are looking to target using constraints. Embodiments of the invention enable viewers that match the constraints to see the service. The user gets to see an automatically generated application set, that is instantly available, with a high probability of containing the application the user is likely to seek. Identity and Payment support are built into the platform, such that the user no longer needs to register with each application or set up payment with each application. In addition, the platform reduces bandwidth consumption, storage front print, and power consumption of the user device by choosing when and which modules to download to the user device.

Abstract: The system and methods described herein provide for managing user datasets by facilitating interactions between users and their advisors following location-based notification of certain triggering events in the user dataset. The geolocation of the user is used to identify nearby advisors who can provide consultation as required by the user. Some embodiments facilitate introductions to a potential user of a set of advisors matched to the user's profile and in response to certain triggering events in the user's dataset.

Abstract: Systems, devices, and methods are described herein for distributing syndicated content or content with one or more distribution restrictions. In one aspect, a method may include assigning a unique ID to a plurality of content items, where the unique ID of at least one of the content items is associated with one or more property distribution restrictions. The method may further include streaming in a video stream the content items to at least two properties and monitoring the unique IDs of the plurality of content items streamed in the video stream to detect least one content item associated with the one or more property distribution restrictions. The distribution of the least one content item associated with the one or more property distribution restrictions may be blocked from the one or more properties concurrently with streaming the video stream, based on the one or more property distribution restrictions.

Abstract: A method for implementing a recommendation system using an asymmetrically hierarchical network includes, for a user and an item corresponding to a user-item pair, aggregating, using asymmetrically designed sentence aggregators, respective ones of a set of item sentence embeddings and a set of user sentence embeddings to generate a set of item review embeddings based on first item attention weights and a set of user review embeddings based on first user attention weights, respectively, aggregating, using asymmetrically designed review aggregators, respective ones of the set of item review embeddings and the set of user review embeddings to generate an item embedding based on a second item attention weights and a user embedding based on second user attention weights, respectively, and predicting a rating of the user-item pair based on the item embedding and the user embedding.

Abstract: Techniques are disclosed for automatically adjusting machine learning parameters in an e-commerce system. Hyperparameters of a machine learning component are tuned using a gradient estimator and a first training set representative of an e-commerce context. The machine learning component is trained using the tuned hyperparameters and the first training set. The hyperparameters are automatically re-tuned using the gradient estimator and a second training set representative of a changed e-commerce context. The machine learning component is re-trained using the re-tuned hyperparameters and the second training set.

Abstract: Aspects of the disclosure relate to use of supervised, unsupervised, semi-supervised, or reinforcement-based machine learning algorithm(s) to perform pattern recognition and/or cluster detection on historical transactional data in order to predict events that may occur in the future and are candidates for automated delivery of items from a source to a customer in a contactless manner, via a secure delivery device, to minimize health risk(s) to the customer. The items may be securely delivered in an automated manner such as, for example, by use of a drone or autonomous vehicle, which may have advanced sensors to facilitate various aspects of the delivery. The items may be contained in a secure carrier coupled to the delivery device and may be unlocked locally or remotely using mechanical or digital means. Novel logical systems, architectures, machines, platforms, delivery devices and components thereof, and methods are disclosed.

Abstract: A message queue storage device includes: a non-volatile flash memory unit including one or more flash memory dies including one or more pages grouped into one or more flash blocks; a volatile memory; a data port; and a storage controller configured to: receive, via the data port, a message write command including a message and a queue identifier; identify a queue from one or more queues based on the queue identifier; determine that the message is a persistent message; select a write physical location in one or more pages of the flash memory dies in which to store the message; and store the message associated with the queue at the write physical location in the one or more pages of the non-volatile flash memory unit.

Abstract: The present method and system provides for interacting with a display device including accessing an external content database having images stored therein, and storing a plurality of the images to a local memory device associated with the display device. The method and system includes displaying a user interface screen on the output display of the display device and detecting, via at least one of a plurality environmental sensors, a user gesture external to the display device, the user gesture engaging in a commercial transaction relating to the display of at least one of the plurality of images on the output display. The method and system includes accessing a financial services processing system, verifying user acknowledgement of the commercial transaction, and updating the output display based on the interface command.

Abstract: Electronic content that has a tactile dimension when presented on a tactile-enabled computing device may be referred to as tactile-enabled content. A tactile-enabled device is a device that is capable of presenting tactile-enabled content in a manner that permits a user to experience tactile quality of electronic content. In one example embodiment, a system is provided for generating content that has a tactile dimension when presented on a tactile-enabled device.

Abstract: This disclosure describes systems and methods that facilitate purchase of objects from merchants. For example, a user may browse a website available from an object management service and identify objects that they desire to purchase. Rather than having to locate the seller of those objects to make a purchase, the implementations described herein facilitate a connection between the user and the merchant so that the merchant's sales are increased and the user is provided an efficient and safe shopping experience.

Abstract: The present invention elates to methods and devices for optimizing a route based on an electronic list, the method comprising the steps of: linking items in a list of items to a position in an area and providing a sorted optimized visiting order wherein the optimization is based on historical data comprising prior visited positions.

Abstract: In online apparel shopping, user finds difficulty in choosing apparel of his/her preferences. This disclosure relates to recommend an apparel specific to a user by estimated attribute values in digital environment. An information associated with the user and apparels is processed to obtain a first parameter and a second parameter and associated data independent matrix and data dependent matrix is derived. The data independent matrix and the data dependent matrix is learned to obtain trained multivariate multi structure model. A significant pattern associated with each user is determined by extracting plurality of images. The derived significant pattern is deployed on k-nearest neighbor model to obtain recommended pattern. The trained multivariate multi structure model and the recommended pattern is applied on state of the user in a real time to estimate the attribute values. The estimated attribute values is mapped with existing apparels in repository to recommend prioritized apparels.

Abstract: The present disclosure generally relates to techniques for managing contextual transactions. In some examples, at a sender electronic device, a first user provides authorization to purchase an item for a second user. In some examples, at a recipient device, the second user selects one or more attributes for the item and requests to claim the item. In some examples, an electronic device displays a representation of an item and a purchase affordance for purchasing the item using a payment system that is available on the electronic device.

Abstract: An example operation may include one or more of applying a user profile to one or more transports and a transport management database, monitoring user requirements and user usage of the one or more transports over a period of time, identifying a plurality of transport profiles stored in a transport availability pool of the transport management database, identifying a proposed fractional ownership of one or more of a plurality of transports identified in the transport availability pool based on the monitored user requirements and user usage of the one or more transports, and providing the proposed fractional ownership of the one or more of a plurality of transports identified in the transport availability pool to a user device associated with the user.

Abstract: A vehicle movement tracking system that employs floor mats having ridges for generating location information in the form of modulated vibrations, detectable with an accelerometer. Two sensors are in a wheel of a vehicle. One sensor senses wheel rotation, and the other sensor senses vertical acceleration. The vehicle passes over a floor mat comprising vertically elevated ridges thereon that code the mat and thereby indicate the location at which the mat is at. When the vehicle travels over this mat the vertical acceleration sensor in the wheel detects the vertically elevated ridges and the wheel rotation sensor detects the distance between the vertically elevated ridges. In combination these two sensors are used to create a location word that denotes the mat over which the vehicle passes over. The location word is stored in non-volatile memory and later uploaded to a location collection station.

Abstract: Among other things, information is received, from two or more competitor merchants offering products to customers, about one or both of a current price and units of inventory of each of the products. The information about the current prices and units of inventory of the product offered by the two or more competitor merchants is provided through a communication network for presentation to one or more customers.

Abstract: In some embodiments, an electronic device presents a subscription management user interface. The subscription management user interface optionally includes one or more selectable options that, when selected, change the status of a subscription associated with a user account of the user. In some embodiments, the subscription management user interface is presented in response to detecting deletion of an application that utilizes an active subscription associated with the user account of the user.

Abstract: Systems and methods for computing a causal uplift in performance of an output action for one or more treatment actions in parallel are described herein. In an embodiment, a server computer receives interaction data for a particular period of time which identifies a plurality of users and a plurality of actions that were performed by each user of the plurality of users through a particular graphical user interface during the particular period of time. The server computer uses the interaction data to generate a feature matrix of actions for each user, and a set of confounding variables included to minimize spurious correlations. The feature matrix is then used to train a machine learning system, using data identifying a user's performance or non-performance of each action as inputs and data identifying performance or non-performance of a target output action as the output.

Abstract: A method for training a model are provided and an information recommendation system. The method includes the following. A reference training sample is obtained. A target training sample is obtained by inputting the reference training sample into a sample model. An updated target model is obtained by training a target model according to the target training sample. A target recommendation task is processed according to the updated target model, where the target recommendation task is a recommendation task for one or more target items to be recommended.

Abstract: A method, a device, and a non-transitory computer readable medium for item recommendation based on visual elements. The method includes: determining, by one or more processors, visual elements from an item image of an item; generating, by the one or more processors, an element descriptor for the item based on at least a part of the visual elements; and calculating, by the one or more processors, a compatibility value between the element descriptor and one or more other element descriptors for one or more other items.

Abstract: Provided is a hydraulic control device in which in the rotation control of the second pump, the required discharge pressure of the second pump can be more reliably maintained below a predetermined pressure in the control that keeps the target rotation speed constant after the feedback control ends. In the rotation control of the second pump, in the control which keeps the target rotation speed constant after the feedback control ends, the rotation of the second pump is controlled by adding a predetermined addition rotation speed to the target rotation speed corresponding to the required discharge pressure. Since the required discharge pressure of the second pump can be maintained below a predetermined pressure in the fixed mode by the addition rotation speed, it can reliably obtain the effect of reducing the work load of the first pump and contribute to the improvement of the fuel efficiency of the vehicle.

Abstract: Some embodiments provide systems to identify products comprising: product vector database; a plurality of portable computing devices comprising a camera and a control circuit configured to: access an image captured by the camera; perform an optical character recognition on the image; apply a vector modeling rule to key text, generate a first query product vector and wirelessly communicate the first query product vector to the product recommendation system; the product recommendation system is configured to apply a vector evaluation rule to the first query product vector to identify a first product; and wirelessly communicate to the portable computing device the first product identifier; wherein the control circuit receives the first product identifier, accesses a product information, causes product information to be displayed; and causes the first product to be virtually added to a virtual cart.

Abstract: A digital facial image from a user is collected. Digital facial images from a plurality of people are collected from a database. A neural network is used to compare a plurality of pixels of the digital facial image of the user to a plurality of pixels of the digital facial images from a plurality of people to determine a facial shape type of the user. A neural network is used to compare the plurality of pixels from the digital facial image of the user to the digital facial images from a plurality of people to determine an ability of the user to grow hair in an area of the user's face. A facial hair style is recommended to the user based on the facial shape type of the user and the ability of the user to grow hair in an area of the user's face.

Abstract: A system for targeting advertising comprises a processor configured to compute a similarity score between a plurality of customers in a seed set and each of a pool of in-store purchasers having no history of online purchases from the online store of an entity. Each user in the seed set has a history of online purchases from the online store of the entity and a history of in-store purchases from stores of the entity. The processor is configured for selecting a subset of the pool of in-store purchasers having similarity scores above a predetermined threshold. The processor is configured for initiating rendering of advertisements for the online store to the subset of the pool of in-store purchasers.

Abstract: Systems and methods are provided for determining pairwise scores for items in an electronic catalog that is stored in a storage device communicatively coupled to a server, and determining when a pair of items in the electronic catalog is valid. A sum of the pairwise scores by activity type for a plurality of activity types may be determined, and the pairwise scores for the plurality of activity types may be summed. A sorted list of items that compliment an anchor item of the electronic catalog may be generated. Items from the generated sorted list may be filtered, and an item ranking in the filtered list may adjusted. Consistency between recommendations of the items in the adjusted ranking list may be determined, final recommendations of the items based on the determined consistency between recommendations may be selected and transmitted for display.

Abstract: Methods and systems for determining information of applications are described herein. A computing device may receive, from one or more client devices, data indicating application usage events. The computing device may determine, based on the data, popular applications that a target user regularly uses. The computing device may determine, based on the data, recommended applications that users similar to the target user regularly use. The computing device may determine, based on the popular applications and/or the recommended applications, a collection of applications for the target user. The collection of applications may be displayed on a user interface of a client device associated with the target user for expedited access of interested applications.

Abstract: An ordering system described herein can include a rapid reorder of items without needing to re-add the individual items into a shopping cart. The rapid reorder may be triggered via a 3D touch or a long press on a mobile device. The ordering system can also provide an architecture that can allow a user to modify an order, such as, e.g., the pickup time or store location, after the order has been submitted.

Abstract: A machine may be configured to perform image evaluation of images depicting items for sale and to provide recommendations for improving the images depicting the items to increase the sales of the items depicted in the images. For example, the machine accesses a result of a user behavior analysis. The machine receives an image of an item from a user device. The machine performs an image evaluation of the received image based on an analysis of the received image and the result of the user behavior analysis. The performing of the image evaluation may include determining a likelihood of a user engaging in a desired user behavior in relation to the received image. Then, the machine generates, based on the evaluation of the received image, an output that references the received image and indicates the likelihood of a user engaging in the desired behavior.

Abstract: A system, computer-readable storage medium storing at least one program, and computer-implemented method for providing recommendations based on social network sharing activity. Sharing activity relating to the sharing of the content item on a social network by a first user is accessed. Consumption information related to the consumption of the content item. A correlation between the sharing activity and the consumption information is determined. A recommendation is then generated based on the correlation.