Abstract: Systems, methods, and computer storage media for handling failures in generating structured queries from natural language queries. One of the methods includes obtaining, through a natural language front end, a natural language query from a user; converting the natural language query into structured operations to be performed on structured application programming interfaces (APIs) of a knowledge base, comprising: parsing the natural language query, analyzing the parsed query to determine dependencies, performing lexical resolution, forming a concept tree based on the dependencies and lexical resolution; analyzing the concept tree to generate a hypergraph, generate virtual query based on the hypergraph, and processing the virtual query to generate one or more structured operations; performing the one or more structured operations on the structured APIs of the knowledge base; and returning search results matching the natural language query to the user.

Abstract: A server computer system receives event logs from one or more client devices for multiple consumption sessions of an online media item. The event logs are received based on random schedules. The system determines, for various times, a first number of consumption sessions during which an event log is scheduled to be sent at the corresponding times. The system determines, for the various times, a second number of consumption sessions during which an event log was sent to the server computer system at the corresponding times. The determinations being based on the event logs. The system determine, for each of the times, a probability that consumption time for the online media item is the corresponding time or less. The determination is based on the first number and the second number. The system determines consumption time metrics for the online media item based the probabilities for the plurality of times.

Abstract: Systems and methods for reducing latency times associated with sessionizing log events from streaming logs are disclosed herein. In an aspect, event data related to user consumption of media content items can be monitored. Further, respective event data based at least in part on user playback of the media content items can be analyzed. In another aspect, pending and finalized event data can be summarized into a finalized playback record. In yet another aspect, the size of a time window can be adjusted based in part on observed logsaver latency. Further, in an aspect, watch time data can be integrated into a de-spamming pipeline.

Abstract: Systems and methods for reducing latency times associated with sessionizing log events from streaming logs are disclosed herein. In an aspect, event data related to user consumption of media content items can be monitored. Further, respective event data based at least in part on user playback of the media content items can be analyzed. In another aspect, pending and finalized event data can be summarized into a finalized playback record. In yet another aspect, the size of a time window can be adjusted based in part on observed logsaver latency. Further, in an aspect, watch time data can be integrated into a de-spamming pipeline.