Abstract: Dynamic materialization of a feed is described. A request for a feed is received. A state of the feed is determined. Responsive to determining that the state is the inactive state, causing materialization of the feed by populating a materialized feeds table in a second non-relational database system with feed items of the feed from the first relational database system, and setting the state of the feed to the activating state. Responsive to determining that the state is the active state, causing the request for the feed to be processed from the materialized feeds table in the second non-relational database system. Responsive to determining that the state is the disabled state, causing the request for the feed to be processed from the first relational database system, and causing an update of the materialized feeds table in the second non-relational database system to be postponed for a predetermined amount of time.

Abstract: The technology disclosed presents a novel spatial attention model that uses current hidden state information of a decoder long short-term memory (LSTM) to guide attention and to extract spatial image features for use in image captioning. The technology disclosed also presents a novel adaptive attention model for image captioning that mixes visual information from a convolutional neural network (CNN) and linguistic information from an LSTM. At each timestep, the adaptive attention model automatically decides how heavily to rely on the image, as opposed to the linguistic model, to emit the next caption word. The technology disclosed further adds a new auxiliary sentinel gate to an LSTM architecture and produces a sentinel LSTM (Sn-LSTM). The sentinel gate produces a visual sentinel at each timestep, which is an additional representation, derived from the LSTM's memory, of long and short term visual and linguistic information.

Abstract: In an example, a processor may render one or more displays of a user interaction window based on navigation through logical pages of a web-based application, the one or more displays including a plurality of controls comprising at least one navigation control to control the navigation and one or more video playback controls to initiate playback of one or more videos, respectively, in a playback window. The processor may continue a playback of one of the one or more videos using the playback window through navigation through the logical pages using the at least one navigation control and/or may launch an additional video of the one or more videos at a same playback window size/position associated with a previous playback.

Abstract: Disclosed herein are system, method, and computer program product embodiments for processing a written-language contract using an industry-specific natural language processing model to determine flows or actions to undertake in a Customer Relationship Management (CRM) solution. A CRM solution may include the ability to receive or create a binding natural-language contract. The CPQ or CRM system may use natural language processing (NLP) to determine terms and conditions included in a natural-language contract. The NLP may further use an industry-specific model that may be determined based on information in the CRM solution to more efficiently and accurately analyze the natural-language contract. The CRM solution may further receive a legal language construct and convert the terms of the legal language construct into a smart contract.

Abstract: Embodiments of the present disclosure relate to flattening hierarchal database records using inverted indexing. Other embodiments may be described and/or claimed.

Abstract: Disclosed herein are system, method, and computer program product embodiments for providing a mixed-cart solution for an in-store app. An embodiment operates by receiving a request to add an item from a physical store to an associate shopping cart associated with an account. It is determined that the account is additionally associated with an online shopping cart associated with the store, and that the requested item does not exist in the online shopping cart. It selected item is added to the associate shopping cart. Both the online shopping cart and the associate shopping cart are provided to a computing device configured to simultaneously display both the online shopping cart and the associate shopping cart.

Abstract: System receives inputs, each input associated with a label and having features, creates a rule for each feature, each rule including a feature and a label, each rule stored in a hierarchy, and distributes each rule into a partition associated with a label or another partition associated with another label. System identifies a number of inputs that include a feature for a rule in the rule partition, and identifies another number of inputs that include both the feature for the rule and another feature for another rule in the rule partition. System deletes the rule from the hierarchy if the ratio of the other number of inputs to the number of inputs satisfies a threshold and an additional number of inputs that includes the other antecedent feature is at least as much as the number. System predicts a label for an input including features by applying each remaining rule to the input.

Abstract: Systems, device and techniques are disclosed for analytics visualization. A selection of a data set may be received. A first visualization associated with the data set may be generated. A selection of a first data subset of the data set may be received. A second visualization associated with the first data subset of the data set may be generated. The second visualization may be visually linked to the first visualization with a first visual link. A selection of a second data subset of the data set may be received. A third visualization associated with the second data subset of the data set may be generated. The third visualization may be visually linked to the first visualization with a second visual link and displayed simultaneously with the second visualization.

Abstract: A database system may generate a new case associated with a customer support query. The database system may initiate a search for other cases related to the new case and identify articles linked to the related cases. The database system may identify more relevant articles by finding the related cases that use a similar vocabulary for describing similar customer problems. The database system then identifies the articles used for resolving the prior problems. Thus, the database system may bridge the gap between vocabularies used by customers to describe problems and vocabularies used in articles to describe solutions to those problems.

Abstract: Disclosed are systems, apparatus, methods, and computer readable media for interfacing with a phone system in an on-demand service environment. In one embodiment, a client machine is in communication with a data provider and the phone system. The phone system is operated by a phone provider separate from the data provider. A message is received at the client machine from the phone system. The message includes information regarding a phone event in relation to a phone call. The phone event information is provided in a first part of a user interface in a web browser on the client machine. A second part of the user interface is capable of displaying information received at the client machine from the data provider. The first part of the user interface and the second part of the user interface are adapted to communicate with one another. For example, the first part of the user interface is adapted to send a message including the phone event information to the second part of the user interface.

Abstract: A multi-tenant system stores a hierarchy of machine-learned models, wherein each machine-learned model is configured to receive as input a set of search results and generate as output scores for ranking the set of search results. Each machine-learned model is associated with a set of dimensions. The system evaluates search query performance. Performance below a threshold causes a new model to be generated and added to the hierarchy of models. Upon execution of a new search query associated with the same set of dimensions as the newly created model, the new model is used to rank that search query's search results.

Abstract: A data driven approach to emulating application performance is presented. By retrieving historical network traffic data, probabilistic models are generated to simulate wireless networks. Optimal distribution families for network values are determined. Performance data is captured from applications operating on simulated user devices operating on a virtual machine with a network simulator running sampled tuple values.

Abstract: Disclosed are examples of systems, apparatus, methods and computer program products for combining updates of a social network feed. In some implementations, a feed of a social networking system can be configured to share feed items associated with an enterprise record. The feed items can include a first update associated with a first user. A first request to share a comment on the first update can be processed. A second request to share a first post in a feed can be processed. The first post can be related to the first update based on a first attribute conforming to a second attribute. A combined update can be generated based on the first post being related to the first update. The combined update can be displayed in a feed.

Abstract: In an example, a database system may be configured to validate relational database instructions using a plurality of validators. In some examples, validation may be pre-production for relational database instructions based on simulated user inputs and/or file, such as plan files (e.g., PL/SQL (procedural language/structured query language) files). In some examples, validation may be in production for relational database instructions based on system views of a database engine.

Abstract: A computer system generates augmented training datasets to train neural network models. The computer system receives an initial training dataset comprising images for training a neural network model, and generates an augmented training dataset by modifying images from the first training dataset. The computer system identifies a representation of a target object against a background from the initial training dataset and extracts a portion of the image displaying the target object. The computer system generates samples for including in the augmented training dataset based on the image. For example, new images may be obtained by performing transformations on the portion of the image displaying the target object and/or by overlaying the transformed portion of the image over a different background. The modified images are included in the augmented training dataset used for training the neural network model to recognize the target object.

Abstract: In accordance with disclosed embodiments, there are provided systems, methods, and apparatuses for implementing a scheduler with preemptive termination of existing workloads to free resources for high priority items.

Abstract: Based on a test step execution order for a test case such as one against a database system, a specific test step to be executed next is identified. In response to identifying the specific test step, a test step message is published to indicate that the specific test step is to be executed next, which causes a subscriber of the test step message to execute the specific test step. In response to determining that the specific test step has ended, a dependent test step message is published accordingly to cause a subscriber of the dependent test step message to perform: determining whether a next test step should be executed following the specific test step in the test step execution order; in response to determining that a next test step should be executed, the foregoing may be repeated by using the next test step in place of the specific test step.

Abstract: A platform is provided to allow the developer to log into a subscriber's computer system by using the credential of the developer or the subscriber to view and to debug at least a portion of executable code that is used by the subscriber. The executable code used by the subscriber may be part of managed package code, at least portions of which are customized for various subscribers. Cooperative debugging may be provided between developer and subscriber with lifting of code encapsulation for the developer in the subscriber implementation. The developer and the subscriber may be granted debugging licenses each authorizing a limited number of debugging sessions, and each debugging session run by the developer may be counted against the debugging license of either the developer or the subscriber. Debugging licenses may be transferred between the developer and the subscriber.

Abstract: In accordance with disclosed embodiments, there are provided systems, methods, and apparatuses for implementing a stateless, deterministic scheduler and work discovery system with interruption recovery.

Abstract: In accordance with disclosed embodiments, there are provided systems, methods, and apparatuses for implementing a stateless, deterministic scheduler and work discovery system with interruption recovery.