In some examples, a prototype model that includes a representative subset of data points (e.g., inputs and output classifications) of a machine learning model is analyzed to efficiently interpret the machine learning model's behavior. Performance metrics such as a critic fraction, local explanation scores, and global explanation scores are determined. A local explanation score capture an importance of a feature of a test point to the machine learning model determining a particular class for the test point and is computed by comparing a value of a feature of a test point to values for prototypes of the prototype model. Using a similar approach, global explanation scores may be computed for features by combining local explanation scores for data points. A critic fraction may be computed to quantify a misclassification rate of the prototype model, indicating the interpretability of the model.

Automated event monitoring is performed utilizing a Knowledge Graph (KG) constructed by grouping and consolidation of a variety of log Entry Types. A log entry is received by a knowledge graph parser (Kg parser). That parser finds contiguous sub-strings in a log entry that have a parameterized basic-format. The parser figures out which basic-formats are present, where, and with which parameters. Given a sub-string, its basic-format and its parameters, the parser can correctly parse the sub-string to components (e.g., keys and values if a key-value format; fields if a structured format). A result of the parsing is an entity type tree structure. Next, a grouping and consolidation capability functions to modify the KG to incorporate an incoming new entry type structure. The KG may be consumed by a user (e.g., visualization; querying), and may provide an artifact to an event monitoring system to automatically trigger certain actions (e.g., alerts).

A fault recovery system including a fault controller is disclosed. The fault controller is coupled between a processor and an interconnect, and configured to receive a time-out signal that is indicative of a failure of the processor to execute a transaction after a fault is detected in the processor. The failure in the execution of the transaction results in queuing of the interconnect. Based on the time-out signal, the fault controller is further configured to generate and transmit a control signal to the processor to disconnect the processor from the interconnect. Further, the fault controller is configured to execute the transaction, and in turn, dequeue the interconnect. When the transaction is successfully executed, the fault controller is further configured to generate a status signal to reset the processor, thereby managing a fault recovery of the processor.

Techniques described herein relate to a method for reporting database backup information. The method may include starting a monitor process; generating a flag file for the database backup comprising a parent process identifier (PID) of a parent process of a first child process; associating a second child process with the monitor process by identifying the flag file using the parent PID; receiving channel close alerts from the child processes; making a first determination that backup jobs performed in a first channel were successful; making a second determination that backup jobs performed in the second channel were successful; and transmitting, based on the first determination and the second determination, a backup success report to a backup coordinator.

A first and a second control modules of a control apparatus of the present disclosure mutually monitor a state of the other end and send a reset request signal to a monitoring module when the other end should be reset. The monitoring module sends a reset signal to one of the first and second control modules when the monitoring module receives the reset request signal indicating that the one of the first and second control modules should be reset from the other and the monitoring module does not send the reset signal to the other. The monitoring module prohibits a reset of one of the first and second control modules when the monitoring module receives the reset request signal indicating that the one of the first and second control modules should be reset from the other and the monitoring module sends the reset signal to the other.

A system can determine performance data for updates created by update creators that execute on devices during a time period, the performance data indicating performance of a group of metrics for a respective update of the updates. The system can determine a first ranking of the update creators for the time period based on the performance data. The system can determine a second ranking of the update creators based on applying a pairwise comparison count of respective rankings of update creators of the first ranking and rankings for time periods prior to the time period. The system can determine whether to publish a first update from a first update creator of the update creators based on a position of the first update creator among the update creators in the second ranking.

Provided herein are systems and methods for creating, testing and providing digital guidance for use in an underlying computer application. In one exemplary implementation, a method includes using a browser extension to record steps of digital guidance content as the steps are created by a content author, automatically testing the content with the same browser and making the tested content available for playback. The testing may include receiving an execute testing command, fetching test case data from a server, executing a test case in the browser using the test case data, running tests on the test case in a background tab of the browser, recording passes and or failures of the tests, sending reports of the passes and failures to the server, sending a notification to the content author that the testing results are ready to view, and displaying the testing results in the dashboard.

A collection of multiple software applications may be tested when a patch is issued. However, an emergency patch for a time-sensitive incident may not allow for full regression or functional testing. Provided herein are techniques for performing a multiple application smoke test. Access information, login information, and a success indicator are obtained for each of a plurality of software applications. A test plan including two or more test packages is determined. Each test package indicates a subset of applications and includes access information, login information, and success indicators corresponding to the subset. The test packages are executed in parallel, including authenticating, loading, and validating. Logs are generated and a user interface is provided to present the logs and whether validation of the application interfaces passed or failed, and a failure reason for failed tests.

Techniques are disclosed relating to reporting for network events within a computer network. A computer system may access a set of data corresponding to a particular network event within a computer network, where the set of data includes captured attributes of the particular network event. The computer system may then calculate, using the set of data, a security score indicative of suspiciousness of the event and an actionability score that is based on an extent to which of a particular group of attributes are missing from the set of data. The computer system may determine, based on the two scores, a combined score for the event. The computer system may then report a notification for the event, based on the combined score. Such techniques may decrease a number of reported events for a network, which may advantageously allow resources to be focused on a smaller set of events.

A system can display content generated from one type of machine-generated data to a user via a graphical user interface. Based on an interaction with the machine-generated data, the system can determine an entity identifier associated with the machine-generated data and determine an entity type for the entity identifier. The system can map the entity type to one or more content generators associated with the entity type and communicate the entity identifier to the identified content generators. The content generators can determine if they have content associated with the machine-generated data. The system can generate and display a link to the related content via a graphical user interface.