Abstract: A method for training a neural network of a neural network based speaker classifier for use in speaker change detection. The method comprises: a) preprocessing input speech data; b) extracting a plurality of feature frames from the preprocessed input speech data; c) normalizing the extracted feature frames of each speaker within the preprocessed input speech data with each speaker's mean and variance; d) concatenating the normalized feature frames to form overlapped longer frames having a frame length and a hop size; e) inputting the overlapped longer frames to the neural network based speaker classifier; and f) training the neural network through forward-backward propagation.

Abstract: A method for training a deep learning algorithm using N-dimensional data sets may be provided. Each data set comprises a plurality of N?1-dimensional data sets. The method comprises selecting a batch size and assembling an equally sized training batch. The samples are selected to be evenly distributed within said respective N-dimensional data sets. The method comprises also starting from a predetermined offset number, wherein the number of samples is equal to the selected batch size number, and feeding said training batches of N?1-dimensional samples into a deep learning algorithm for the training. Upon the training resulting in a learning rate that is below a predetermined level, selecting a different offset number for at least one of said N-dimensional data sets, and going back to the step of assembling. Upon the training resulting in a learning rate that is equal or higher than said predetermined level, the method stops.

Abstract: An approach is provided for optimally routing events in an IT system to solvers which provide resolutions of the events. Event streams originating from the IT system are defined. Events are classified into the event streams. An optimization problem is solved that minimizes costs incurred for using respective solvers based on constraints which include success rates of the solvers. Based on the solved optimization problem, policies are defined that associate the event streams to the solvers in a many-to-one correspondence. In real time, the defined policies are applied to the event streams. Based on the applied policies and the classified events, the events are routed to respective solvers. An indication is received that the events are resolved by the respective solvers, which reduces downtime in the IT system.

Abstract: A method for establishing semantic equivalence between a plurality of concepts including: providing an Orthogonal Semantic Equivalence Map in which first, second, and third extensional concept models are related; selecting or de-selecting a concept in the first concept model; selecting or deselecting a (relation, concept) pair representing an intensional relation from a concept in the first concept model to a concept in the second concept model over a concept in the third concept model; determining a subset of intensional relations from the selected concepts in the first concept model to concepts in the second concept model; determining a set of concepts from the first concept model that are related to concepts in the second concept model over the selected (relation, concept) pairs; and determining the narrowest common extension of the set of concepts from the first, second, or third concept models that are related over the selected intensional relations.

Abstract: A method for predicting an interface control action of a user with an in-vehicle user interface involves collecting and storing data. The data can be vehicle data about the vehicle and its environment collected from at least one sensor of the vehicle and user data about user interactions with the user interface and/or different applications inside the vehicle. Likelihoods are assigned to at least two possible interface control actions by the user based on the collected and stored data. At least one most likely interface control action is determined from the likelihoods and the user is presented with the at least one most likely interface control action so that it is selectable and performable with one single user interaction with the user interface.

Abstract: Systems and methods provide an application programming interface to offer action suggestions to third-party applications using context data associated with the third-party. An example method includes receiving content information and context information from a source mobile application, the content information representing information to be displayed on a mobile device as part of a source mobile application administered by a third party, the context information being information specific to the third party and unavailable to a screen scraper. The method also includes predicting an action based on the content information and the context information, the action representing a deep link for a target mobile application. The method further includes providing the action to the source mobile application with a title and a thumbnail, the source mobile application using the title and thumbnail to display a selectable control that, when selected, causes the mobile device to initiate the action.

Abstract: An approach is provided for a redundant feature detection engine. The approach, for instance, involves segmenting an input image into a plurality of grid cells for processing by the redundant feature detection engine. The redundant feature detection engine includes a neural network. The approach also involves, for each of the plurality of grid cells, initiating a prediction of an object code by the redundant feature detection engine. The object code is a predicted feature that uniquely identifies an object depicted in the input image. The approach further involves aggregating the plurality of grid cells into one or more clusters based on the object code predicted for said each grid cell. The approach further involves predicting one or more features of the object corresponding to a respective cluster of the one or more clusters by merging one or more feature prediction outputs of said each grid cell in the respective cluster.

Abstract: A method for determining a similarity between a plurality of graphs includes inferring a low-rank representation of a first graph, inferring a low-rank representation of a second graph, wherein the low-rank representations of the first and second graphs are stored in memory, estimating a left interaction between the first and second graphs, estimating a middle interaction between the first and second graphs, estimating a right interaction between the first and second graphs, wherein the estimations are based on the low-rank representations of the first and second graphs stored in memory, and aggregating the left interaction, the middle interaction and the right interaction into a kernel, wherein the kernel is indicative of the similarity between the first and second graphs.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for parallel time series analysis. One of the methods includes receiving a plurality of data elements. The plurality of data elements are partitioned into a plurality of chunks, wherein the plurality of chunks, including a first chunk and a plurality of additional chunks, have an ordering according to the data elements included in each chunk. Each chunk is assigned to a particular segment of a plurality of segments. A first iteration of an autoregressive integrated moving average is computed for each chunk assigned to each segment. A second iteration of the autoregressive integrated moving average is computed for each chunk assigned to each segment, wherein computing uses the result data for a corresponding preceding chunk in the first iteration. One or more additional iterations of the autoregressive integrated moving average are computed until stopping criteria has been satisfied.

Abstract: Certain aspects involve optimizing neural networks or other models for assessing risks and generating explanatory data regarding predictor variables used in the model. In one example, a system identifies predictor variables compliant with certain monotonicity constraints. The system generates a neural network for determining a relationship between each predictor variable and a risk indicator. The system performs a factor analysis on the predictor variables to determine common factors. The system iteratively adjusts the neural network so that (i) a monotonic relationship exists between each common factor and the risk indicator and (ii) a respective variance inflation factor for each common factor is sufficiently low. Each variance inflation factor indicates multicollinearity among a subset of the predictor variables corresponding to a common factor.

Abstract: A method for generating a pattern matching machine for identifying matches of a symbol pattern in a sequence of in put symbols, the received symbol pattern including an indicator that one or more symbols in the symbol pattern repeat as repeated symbols, the method comprising: generating a plurality of symbol patterns for the received symbol pattern, each generated symbol pattern being equivalent to the symbol pattern for a different number of repetitions of the repeated symbols, wherein the plurality of generated symbol patterns comprise all possible symbol patterns equivalent to the received symbol pattern up to a predetermined maximum length of a generated symbol pattern.

Abstract: The method includes identifying, by one or more computer processors, a grouping of elements in a storage system. The method further includes identifying, by one or more computer processors, a first element from the identified grouping of elements. The method further includes identifying, by one or more computer processors, a root unit of the first element from the identified grouping of elements. The method further includes determining, by one or more computer processors, a past usage history of computer resources for the identified root unit of the first element. The method further includes calculating, by one or more computer processors, a future usage of computer resources for the identified root unit based upon the determined past usage history.

Abstract: An apparatus for estimating a characteristic of a posterior distribution includes a sampling section configured to generate a plurality of samples from a prior distribution, an obtaining section configured to obtain, for each sample among the plurality of samples, a likelihood of observation given the sample, a calculation section configured to calculate a parameter relating to a density at each sample in the prior distribution, and an estimation section configured to estimate, for the plurality of samples, a characteristic of the posterior distribution based on the parameter relating to the density at each sample and the likelihood of observation for each sample.

Abstract: A tunable resonator is formed by shunting a set of asymmetric DC-SQUIDs with a capacitive device. An asymmetric DC-SQUID includes a first Josephson junction and a second Josephson junction, where the critical currents of the first and second Josephson junctions are different. A coupling is formed between the tunable resonator and a qubit such that the capacitively-shunted asymmetric DC-SQUIDs can dispersively read a quantum state of the qubit. An external magnetic flux is set to a first value and applied to the tunable resonator. A first value of the external magnetic flux causes the tunable resonator to tune to a first frequency within a first frequency difference from a resonance frequency of the qubit, the tunable resonator tuning to the first frequency causes active reset of the qubit.

Abstract: Technologies are generally described for methods and systems in a machine learning environment. In some examples, a method may include retrieving training data from a memory. The training data may include training inputs and training labels. The methods may further include determining a set of datasets based on the training inputs. The methods may further include determining a set of out of sample errors based on the training inputs and based on test data. Each out of sample error may correspond to a respective dataset in the set of datasets. The methods may further include generating alternative distribution data based on the set of out of sample errors. The alternative distribution data may be used to determine weights to be applied to the training data.

Abstract: A method and apparatus for training an algorithm to recognize a walking state of a user wearing a walking assistance apparatus are provided. The method and apparatus may generate virtual sensing data associated with walking of a virtual human model through a dynamic simulation, and may train the algorithm based on the virtual sensing data.

Abstract: The present disclosure relates to the electronic document review field and, more particularly, to various apparatuses and methods of implementing batch-mode active learning for technology-assisted review (TAR) of documents (e.g., legal documents).

Abstract: The present disclosure relates to the electronic document review field and, more particularly, to various apparatuses and methods of implementing batch-mode active learning for technology-assisted review (TAR) of documents (e.g., legal documents).

Abstract: In an example embodiment, each of a plurality of members of a social networking service is mapped to a weighted skill vector, each weighted skill vector including a list of skills for the member with an associated weight indicating a strength of the skill. Members of the social networking service who belong to an industry are aggregated to obtain a weighted matrix of members and skills along with compensation vectors indicating compensation for each of the members in the matrix. The weighted matrix of members and skills and corresponding compensation vectors are used to train a machine learning skill monetary value prediction model to output a predicted monetary value for a skill contained in a candidate vector fed to the machine learning skill monetary value prediction model. A recommendation is provided to a member of one or more skills to add based on output of the model.

Abstract: One embodiment provides a method comprising answering one or more incoming phone calls received at one or more pre-specified phone numbers utilizing a bot. The bot is configured to engage in a conversation with a caller initiating an incoming phone call utilizing a voice recording that impersonates a human being. The method further comprises recording each conversation the bot engages in, and classifying each recorded conversation as one of poison data or truthful training data based on content of the recorded conversation and one or more learned detection models for detecting poisoned data.

Abstract: A method, an apparatus, and a computer program product for accessing electronic medical records are provided in which a portable computing device uniquely associated with a user authenticates an identification of the user and automatically retrieves information corresponding to the user from electronic healthcare records systems using the identification. The retrieved information may be combined with other information and electronically delivered to a healthcare provider.

Abstract: A management system for a Permitting Authority (PA) or its appointed designee to monitor and track application-based commercial ground transportation (ABCT) Provider activity through the ABCT-Driver's mobile device, the ABCT-Provider's app, the ABCT-Provider's computer systems, and the PA's electronically-enabled geo-fence, all without the need for specialized hardware such as transponders or other tracking equipment. Each mobile communication device associated with an ABCT-Vehicle continually transmits information to an ABCT-Provider Information and Communications Technology (“ICT”) System associated with an ABCT-Provider when an application on the mobile communication device is active. The information transmitted by the mobile communication device enables the ABCT-Provider ICT System to identify and monitor the ABCT-Driver's identity, the vehicle information, the geographic locus, and/or the ABCT-Vehicle's activity data.

Abstract: A sustainable business development management system and method are provided for businesses which provides these businesses with a comprehensive understanding of their sustainability goals, projects, and metrics including the environmental and societal impacts of their business with the ability to manage and report on these parameters in a way that makes sense for their business.

Abstract: An example embodiment provides a method comprising determining a launch date; determining a product launch plan by working back from the launch date, executing the product launch plan, and tracking the plurality of the product as they are transported in accordance with the product launch plan. The product launch plan comprises a configurable visibility threshold point. The method further comprises receiving a tracking request and determining whether the execution of the product launch plan has reached the configurable visibility threshold point. Responsive to determining that the execution of the product launch plan has reached the configurable visibility threshold point, providing tracking information/data. Responsive to determining that the execution of the product launch plan has not reached the configurable visibility threshold point, providing a message indicating that the expected delivery date of the product is the launch date.

Abstract: In some examples, a service provider may determine from a shift schedule that a first employee is scheduled to work a particular shift at a business within a threshold period of time. The service provider may determine, based on a location of a first mobile device of the first employee, that the first employee is unlikely to arrive at the business in time for the start of the particular shift. In response, the service provider may find a replacement employee to fill the shift of the first employee. For instance, the service provider may send the communication to a second mobile device associated with a second employee that is determined to be within a threshold distance of the business. Alternatively, if the service provider determines that the business is overstaffed on a particular day, the service provider may offer a selected employee an incentive to depart work early.

Abstract: Among other things, one or more techniques and/or systems are provided for developing a criticality profile of an industrial asset. The criticality profile describes the importance of the industrial asset to a system in terms of one or more criticality metrics and/or one or more confidence profiles. Such metrics may include an operational metric indicative of the operation impact of the industrial asset on a system if the industrial asset remains unchanged or enters a degraded or improved state, a restoration metric indicative of a complexity of restoring the industrial asset to an operational state from the unchanged or degraded state and/or complexity of improving the industrial asset to an improved state, and/or an interdependency metric indicative of a relationship between the industrial asset and one or more other industrial assets and/or between the industrial asset and one or more structures in an environment associated with the industrial asset.

Abstract: A method for executing a plurality of business rules on a plurality of incomplete data is provided. The method may include receiving the plurality of business rules, an object model, and a plurality of ruleset parameters. The method may also include generating a dependency tree based on the received plurality of business rules, the received object model, and the received plurality of ruleset parameters. The method may further include generating a plurality of ruleset source code based on a dependency tree that allows for disabling at least one of the received plurality of business rules. The method may also include determining a plurality of data corresponding to the presented set of data elements needed for evaluating the corresponding business rule is unavailable. The method may further include disabling the corresponding business rule based on determining the plurality of data is unavailable.

Abstract: Systems and methods for recommending merchant actions include a physical merchant location having a plurality of beacon devices. A system provider device receives a plurality of first location information from the plurality of beacon devices that is collected from a customer beacon communication device that is associated with a customer. The system provider device also receives a plurality of second location information over the network from the plurality of beacon devices that is collected from a merchant beacon communication device that is associated with an asset of the merchant. The system provider device analyzes the plurality of first location information and the plurality of second location information to determine at least one merchant action recommendation related to the customer and the asset of the merchant and provides the at least one merchant action recommendation over the network for display on a display device.

Abstract: Method, system, and computer program product for simulating the layout and flow of a manufacturing facility. An initial facility layout concept and a baseline input data set are received which define placement and movement of virtual objects within a virtual facility model. At least one of several analyses is performed based on the initial facility layout concept and the baseline input data set. An event flow analysis that outputs timing for movement of the objects from the initial positions may be performed. A geometric flow analysis that outputs interactions between the objects resulting from the movements from the initial positions may be performed. An ergonomic analysis that outputs human postures and motions resulting from interactions with objects may be performed. A simulated flow model for the manufacturing facility is generated. A three-dimensional virtual environment of the simulated flow model is generated and output for display.

Abstract: A command center provides ad-hoc search and navigation through different dimensional topic levels that allow users to better analyze and correlate data. The command center not only generates and displays content and metrics associated with an event, but can also generate and display real-time content and metrics for other topics, products, and/or services related to a currently displayed topic. The command center enables a user to navigate through the different dimensional levels simply by selecting items displayed by a remote control application. The command center also automatically synchronizes the content and metrics for any selected dimension to a same corresponding time period.

Abstract: A method of automated remote terminal benchmarking includes providing a computing system including a processor having an associated memory which implements a benchmarking algorithm. The benchmarking algorithm implements receiving raw data associated with a plurality of Key Performance Indicator (KPIs) including real-time data from different bulk liquid terminals spanning sites across a plurality of continents, and calculating a global target benchmark value or global target benchmark range for at least a first of the plurality of KPIs from the raw data. Responsive to a user' request at a selected first of the different bulk liquid terminal (first terminal), a benchmark report is generated which benchmarks a KPI performance of the first terminal including for the first KPI by a comparison to the global target benchmark value or global target benchmark range.

Abstract: The present disclosure relates to systems and methods for assigning node profiles to record objects. The method includes maintaining a plurality of node profiles. The method includes identifying a record object to which to assign a node profile. The method includes selecting a subset of node profiles that satisfy a node profile selection policy. The method includes generating, for each of the subset of node profiles, a performance profile using electronic activities of the node profiles and one or more object field-value pairs of the record object. The method includes determining, for a first node profile, that a match score between the first node profile and the record object based on the performance profile and one or more object field-value pairs of the record object satisfies a threshold. The method includes storing an association between the first node profile and the record object based on the match score.

Abstract: Process model merging receives data representative of a first model that models a first process and a second model that models a second process, and determines, based on the data and a predetermined plurality of basic model manipulation operations, atomic differences in the second model relative to the first model. Compound differences are determined in the second model relative to the first model. A user is informed of the atomic differences and the compound differences. The user is allowed to select a set of the plurality of basic model manipulation operations. A modified first model is obtained by applying the selected set of basic model manipulation operations to the first model.

Abstract: A system, computer-implemented method, and software for automatically planning and scheduling ocean-going vessels for oil distribution is provided. The scheduling of the vessels is based on a filtered beam search and greedy heuristic. A server can be used for receiving a schedule request and one or more constraints for scheduling one or more vessels from one or more users. An optimization engine can be used for generating a schedule based at least in part on the one or more constraints using a beam search algorithm.

Abstract: Disclosed herein is a system for managing item delivery. The system includes an order management system, comprising an order terminal for receiving an order, a transmitter for receiving the order from the order terminal and sending the order to one or more product providers, a verification terminal by which a recipient can verify completion of delivery of items associated with the order, and a scheduler. The scheduler schedules a robot to be available at an interchange to receive the one or more items, and the order management system provides a delivery destination to the robot by which, or so that, the robot can autonomously navigate to the destination.

Abstract: In some embodiments, methods and systems are provided that provided for delivering products ordered by a customer of a retailer to a delivery destination designated by the ordering customer by way of autonomous transport vehicles configured to identify products to be dropped off at their next delivery destinations and to prepare such products for deployment while the ATVs are still en-route to their next delivery destinations, and to automatically deploy such products upon arrival at such delivery destinations.

Abstract: In some embodiments, methods and systems are provided that provided for delivering products ordered by a customer of a retailer to a delivery destination designated by the ordering customer by way of having another customer pick up the ordered products at the store location of the retailer and then hand off the picked up products to an autonomous transport vehicle that delivers the handed off products to a delivery destination designated by the ordering customer.

Abstract: In some embodiments, systems, apparatuses and methods are provided herein useful to delivery packages using unmanned delivery aircrafts. Some embodiments include product delivery systems, comprising: a transceiver; a control circuit; a memory coupled to the control circuit and storing computer instructions that when executed by the control circuit cause the control circuit to perform the steps of: receive, from a customer, an authorization to deliver a product by an unmanned delivery aircraft; receive, from a portable user interface unit associated with the customer, global location information of a current location of the user interface unit and that designates a delivery location where the customer would like the product delivered; and initiate a delivery, by an unmanned delivery aircraft, of the product to the delivery location defined by the global location information received from the user interface unit.

Abstract: Determining a remaining amount of material in a material package may be provided. First, a weight of the material package may be received in response to the ingress of the material package back into an operations area. Then, based upon a beacon code received in response to the ingress of the material package back into the operations area, a material database may be queried for a material package record. Next, a remaining amount of material corresponding to the material package may be calculated based upon the received weight and data from the material package record found from the material database query. The material package record may then be updated in the material database with the calculated remaining amount of material corresponding to the material package.

Abstract: In some embodiments, methods and systems for determining stocking locations of products having more than one stocking location on a sales floor of a retail facility include an electronic inventory management database that stores product identification data indicating an identify of one or more products at the retail facility and product location data indicating a stocking location of the products on the sales floor of the retail facility. An electronic inventory management device including a processor-based control circuit is configured to correlate the product identification data and the product location data stored in the electronic inventory management database, determine that one or more products are stocked in first and second stocking locations on the sales floor of the retail facility, track worker task events associated with the products, and determine whether the product is to be stocked or was stocked at the first or second stocking location.

Abstract: Some embodiments provide systems, apparatuses, and methods to direct the consolidation of products in fulfilling a product order for a retail shopping facility. In some embodiments, a system is configured to identify a shipment order for products to be delivered to a shopping facility and that includes a request for a full pull pallet of a first product and additional products; define the full pull pallet as an enhanced case intended to be treated by an orderfiller worker as less than a full pull pallet; generate a pallet assembly list identifying the enhanced case and the items of a plurality of the additional products; and communicate the pallet assembly list to a user interface unit to be accessible by an orderfiller worker and directing the orderfiller worker to retrieve and cooperate the enhanced case and the additional products as an assembled pallet.

Abstract: Disclosed herein is a mechanism, which may be in a community-driven system, to assist a user, or member of the community, to articulate a request, such as a question, that is likely to receive a response, e.g., an answer to a question posed, from the user community. A machine learning approach may be used to generate a model, which model may be trained using previously-posted requests, e.g., questions, and response, e.g., answers. The generated model may be used to make a prediction, e.g., to predict a potential number of answers for a given question, which prediction may also have an accompanying confidence score. A prediction generated by the model may be presented to the user as the user is providing the input, e.g., typing a question. Such prediction may be generated for a partial question, e.g., as the question is being typed, as well as a completed question.

Abstract: Usages of language are analyzed in ways that are at least partially language independent. In preferred embodiments, portions of a document are hashed, and the resulting hash values are compared with each other and with those of other documents in real-time. Analyses can be used to gauge conformity of a document to one or more standards utilizing a hash triplet consisting of a hash for each document word and two involving the word and its preceding and following words, to provide suggestions to the author, and to filter email.

Abstract: A calendar view smoothly and progressively transitions between different calendar views. Instead of snapping to a week view in response to changing from a month view to the week view, the month view progressively transitions to the week view while receiving an interaction. For example, the week view may smoothly expand while the view of the other weeks in the month smoothly decrease in size while the interaction is being received. A user may navigate between different calendar views such as year to month, month to week, week to multiple days, days to single days, and the like. When navigating month to month, week to week, day to day, and the like, the transitions between each time element are progressively and smoothly transitioned. Task information or other information that is displayed in the calendar progressively transitions to show more or less information depending on the interaction being received.

Abstract: A technique for conducting a transaction via a network is described. In this technique, in response to receiving transaction information associated with the transaction via the network from a network browser in a computer, a computer system provides an encrypted version of the received transaction information via the network to an application that executes in the computer, which facilitates a comparison of the received transaction information with original transaction information provided by a user to the computer. Note that this application is separate from the network browser, and the communication with the application via the network is independent of the network browser. Moreover, if the received transaction information matches the original transaction information, the computer system receives approval for the transaction from the computer via the network.

Abstract: A system and method for performing customized money transfer send transactions uses a computer system connectable to receive data representing a send transaction identifier of at least one token. The computer system presents to the sender a completion interface that includes fields for receiving sender identification information, recipient identification information and a notification method, and message options related to the transaction. In response to receipt of sufficient information entered at the completion interface, an active, funded send transaction record is stored in the computer system including the message options for recipient notification.

Abstract: A system and method are disclosed for providing integrated circuit (IC) chip modules for inclusion in payment objects. IC chip modules may be received from a supplier of IC chip modules, wherein the IC chip modules are supportably interconnected to a flexible substrate and transport key data is encoded in the IC chip modules. Each IC chip module may be encoded with personalization data and issuer key data. Encoding of IC chip modules may be completed utilizing either a contact IC chip interface device or a non-contact IC chip interface device. Encoding the IC chip modules may be completed with the IC chip modules supportably interconnected to the flexible substrate as supplied by the supplier of the IC chip modules, free from fixed interconnection of the IC chip modules and carrier to another support structure.

Abstract: Methods and systems for completing financial operations via a contactless automated teller machine are disclosed. A financial institution computing system includes a network interface circuit exchanging information over a network, a customer database retrievably storing financial information relating to a plurality of customers, and a data exchange circuit. The data exchange circuit receives a financial operation request. The financial operation request is generated by an automated teller machine in response to a payment token received from a mobile wallet circuit on a mobile device. The data exchange circuit authorizes the financial operation request based on information in the customer database.

Abstract: Arrangements relate to processing financial transactions on an automated teller machine (ATM). An ATM identifier can be presented on a display of the ATM. An ATM identifier confirmation can be received from a mobile computing device. Responsive to determining that the ATM identifier confirmation matches the identifier presented on the display of the ATM, a customer authentication or a transaction request can be received from the mobile computing device. Responsive to verifying the customer authentication or the transaction request, the transaction request can be processed.

Abstract: A method and a system conducts payment transactions via instant messaging. The system includes a messaging interface to receive a payment transaction request at a payment application from a network-based device via a network. The messaging interface facilitates substantially instant communication over the network. The payment application, responsive to receiving the payment transaction request, transfers a payment from a user associated with the network-based device to a third party.