Abstract: Techniques are disclosed relating to grouping user interface elements into components. The disclosed techniques may include, for example, determining, based on hierarchical user interface code, a plurality of visible elements of a graphical user interface. In addition, the disclosed techniques include determining coordinates for bounding boxes for ones of the plurality of visible elements and determining depths within the hierarchical user interface code for the plurality of visible elements. The disclosed techniques include generating groupings for the plurality of visible elements using unsupervised clustering and the determined coordinates and depths and storing information specifying the generated groupings.

Abstract: A computer-implemented method for enabling a mobile enterprise application of a database system to use a machine learning (ML) service comprises modifying, by a provider of the database system, a mobile enterprise application made available to one or more tenants of the database system to include a native code layer capable of downloading and invoking a trained ML model. One or more servers of the database system receives an upload of the trained ML model and configuration data from a tenant user, wherein the configuration data defines a format of input data and a format of output data of the trained ML model. The one or more servers download the trained ML model to the mobile enterprise application running on a mobile device in response to a request by the native code layer of the mobile enterprise application.

Abstract: Embodiments of the present disclosure relate to software application performance regression analysis. Other embodiments may be described and/or claimed.

Abstract: Digital data processing systems of the type in which a server digital data device (“server”) is coupled to a client digital data device (“client”) over a network, e.g., the Internet, include web server software executing within an application layer on the server that responds to a request from the client by (i) validating a key received from the client with that request, (ii) generating a result code indicative of a success of that validation, (iii) initiating processing of the request, including invoking server resource software executing outside the application layer. The server resource software, which checks the result code upon invocation and before performing a protected operation required for processing the request, responds to a result code indicating that the result did not validate by exiting before executing the protected operation.

Abstract: Methods and systems for query optimization for a multi-tenant database system are provided. Some embodiments comprise receiving, at a network interface of a server computer in a multi-tenant database system, an original query transmitted to the multi-tenant database system by a user associated with a tenant, wherein the original query is associated with data accessible by the tenant, and wherein the multi-tenant database system includes a first two-column custom index. Metadata associated with the data accessible by the tenant is retrieved, using a processor of the server computer, wherein at least a portion of the data accessible by the tenant is stored in a common table within the multi-tenant database system. A tenant-selective query syntax is determined, using the processor, wherein determining comprises analyzing at least one of metadata generated from information about the tenant or metadata generated from the data accessible by the tenant.

Abstract: The system and methods of the disclosed subject matter provide a hybrid machine learning approach for recommending items that a consumer should be shown as a next best offer. The recommendation may be based on the consumer's previous behavior, other consumers' previous behavior, and the consumer's profile. The system and methods may cluster an input dataset using an unsupervised clustering engine. The dataset output from the unsupervised clustering engine may be subsequently provided to the input of a supervised machine learning engine to generate a rules-based model. The system and methods may use the rules-based model to subsequently cluster new user data and generate recommendations based on the user's assigned cluster.

Abstract: System and methods are described for a monitoring system to monitor operations of a distributed computing system and provide feedback to subscribers. The monitoring system includes a data collection engine to receive data items comprising events, metrics, or one or more configurations about a distributed computing system being monitored, and to store selected data items into a message queue. The monitoring system also includes a data processing engine to receive the selected data items from the queue, transform the selected data items into transformed data items based at least in part on one or more settings for a subscriber and the one or more configurations of the distributed computing system, and store the transformed data items into one or more of a database and a cache.

Abstract: Embodiments of the present disclosure relate to software application optimization. Other embodiments may be described and/or claimed.

Abstract: Techniques are disclosed relating to handling queues. A server-based platform, in some embodiments, accesses queue information that includes performance attributes for a plurality of queues storing one or more messages corresponding to one or more applications. In some embodiments, the platform assigns, based on the performance attributes, a corresponding set of the plurality of queues to each of a plurality of processing nodes of the platform. In some embodiments, the assigning of a corresponding set of queues to a given one of the plurality of processing nodes causes instantiation of: a first set of one or more dequeuing threads and a second set of one or more processing threads. The dequeuing threads may be executable to dequeue one or more messages stored in the corresponding set of queues. The processing threads may be executable to perform one or more tasks specified in the dequeued one or more messages.

Abstract: Computer implemented methods and systems are provided for generating podcast files. In one embodiment, a method includes storing, in a template database, a podcast template, wherein the podcast template includes a sentence definition including one or more data tags and a sentence order for a plurality of sentences; storing, in a user database, user information for a plurality of users; generating, by a processor, a podcast text file by selectively populating the podcast template with user information associated with a first user of the plurality of users based on the data tags; converting, by the processor, the podcast text file to a podcast audio file; and storing, by the processor, the podcast text file and the podcast audio file in a podcast database for podcast playback by the first user.

Abstract: A method of training a predictive model to predict a likely field value for one or more user selected fields within an application. The method comprises providing a user interface for user selection of the one or more user selected fields within the application; analyzing a pre-existing, user provided data set of objects; training, based on the analysis, the predictive model; determining, for each user selected field based on the analysis, a confidence function for the predictive model that identifies the percentage of cases predicted correctly at different applied confidence levels, the percentage of cases predicted incorrectly at different applied confidence levels, and the percentage of cases in which the prediction model could not provide a prediction at different applied confidence levels; and providing a user interface for user review of the confidence functions for user selection of confidence threshold levels to be used with the predictive model.

Abstract: Disclosed are examples of systems, apparatus, methods and computer program products for automatically matching start criteria and completion criteria of a walkthrough for an application or an online service. A database storing data objects identifying walkthroughs and walkthrough stages can be maintained. An indication can be received that walkthrough stages of a first walkthrough in the database have been added, removed, or modified. It can be determined that a first completion criterion associated with a first walkthrough stage of the first walkthrough and a first start criterion associated with a second walkthrough stage of the first walkthrough do not match. The first start criterion and the first completion criterion can be processed to generate a matched start criterion and a matched completion criterion. The first start criterion and the first completion criterion can be replaced with the matched start criterion and the matched completion criterion to generate a modified walkthrough.

Abstract: Disclosed are examples of systems, apparatus, methods, and computer program products for processing offline updates to records of a database system. In some implementations, a first update to a record is processed, and the record is modified according to a first action and first data of the first update. A second update to the record from a device with an offline status is processed. It can be determined that the second update and the first update are associated with a same portion of the record. It can be determined that the second timestamp is prior to the first timestamp. A notification can be provided to the device. The record can be modified according to a second action and second data of the second update.

Abstract: A request to update an original data value in a first row in a database table in a database system. An updated data value is written to a second row in a staging table in the database system. The updated data value corresponds with the original data value. The first row includes a database table key, which is also included in the second row. The original data value in the database table is replaced with a corresponding replacement value, which is determined based on a value replacement update function that takes as input the updated data value. The staging table maintains a record value for reversing the update to the database table.

Abstract: A system tokenizes raw values and corresponding standardized values into raw token sequences and corresponding standardized token sequences. A machine-learning model learns standardization from token insertions and token substitutions that modify the raw token sequences to match the corresponding standardized token sequences. The system tokenizes an input value into an input token sequence. The machine-learning model determines a probability of inserting an insertion token after an insertion markable token in the input token sequence. If the probability of inserting the insertion token satisfies a threshold, the system inserts the insertion token after the insertion markable token in the input token sequence. The machine-learning model determines a probability of substituting a substitution token for a substitutable token in the input token sequence.

Abstract: A plurality of database access requests are received at a database system within a designated database access scope. The plurality of database access requests may include a first request to read from a first database table and a second request to write to a second database table. The second request may occur during or after the receipt of the first request. A database access dependency rule linking the first database table and the second database table may be created. The database access dependency rule may impose a protected database access framework when executing a subsequent write request to the second database table when the first database table is updated. The protected database access framework may store one or more values associated with the subsequent write request in a staging table that are capable of being used to undo the subsequent write request.

Abstract: A training set is created via creating adjacent classified substrings by using character classes to replace corresponding characters in adjacent substrings in each training character string, and associating each pair of adjacent classified substrings and each pair of adjacent substrings with corresponding labels indicating whether corresponding pairs include any token boundary. The system splits input character string into beginning and ending parts and creates classified beginning part by replacing beginning part character with corresponding class and classified ending part by replacing ending part character with corresponding class. The machine-learning model determines probability of token identification, based on training set to determine count of instances that classified beginning part is paired with classified ending part and count of corresponding labels that indicate inclusion of any token boundary.

Abstract: A request for a transaction between a client system and a server system may be processed. The transaction may be associated with transmission of data between the client system and the server system. The data may be encrypted using a transient encryption key to form encrypted data. The transient encryption key may be a synced-clock random number configured to automatically change when a designated time interval elapses. The encrypted data may be transmitted between the client system and the server system.

Abstract: Data field lifecycle management mechanisms. Custom metadata about a database object is generated. The custom metadata includes at least a specified name for the custom metadata type, a specified one or more fields to contain configuration information for the custom metadata type, and specification of at least one relationship field having fields as its domain relating the custom metadata type to a column or field of the database object stored in a database. One or more objects of the custom metadata type are maintained with referential integrity to provide metadata corresponding to the database object. The custom metadata objects maintain tenant-specific information. The one or more objects of the custom metadata type are updated automatically. A notification to one or more users is generated for one or more objects of the custom metadata type that cannot be updated automatically, if any.

Abstract: In various embodiments, a method of transferring data between datacenters may be performed. The method may include running a first plurality of host programs and a first plurality of encryption units at a first datacenter. The method may further include establishing, between the first datacenter and a second datacenter, secure communication connections between each of the first plurality of encryption units and a corresponding one of a second plurality of encryption units running at the second datacenter. The method may further include transferring, by the first datacenter, data from the first plurality of host programs to a second plurality of host programs running at the second datacenter.