US 11,868,398 B1
Dynamic determination of video correlation
Dmytro Popovych, Mountain View, CA (US); Michael Kamprath, Mountain View, CA (US); and Andrew Huson, Goleta, CA (US)
Assigned to Tubular Labs, Inc., Mountain View, CA (US)
Filed by Tubular Labs, Inc., Mountain View, CA (US)
Filed on Jun. 27, 2022, as Appl. No. 17/850,646.
Int. Cl. G06F 16/70 (2019.01); G06F 16/735 (2019.01); G06F 16/75 (2019.01); G06F 16/71 (2019.01)
CPC G06F 16/735 (2019.01) [G06F 16/71 (2019.01); G06F 16/75 (2019.01)] 18 Claims
OG exemplary drawing
 
1. A system, comprising:
a processor configured to:
obtain a plurality of recorded video interactions;
obtain a plurality of recorded non-video interactions;
store, at a database, event data determined based at least in part on the plurality of recorded video interactions and the plurality of recorded non-video interactions, wherein the database comprises a plurality of storage nodes; and
determine a subset of the event data that comprises correlations between a video and non-video interactions, wherein to determine the subset of the event data that comprises correlations between the video and the non-video interactions comprises to:
receive a request to determine correlation values between a given set of videos and non-video category identifiers (IDs);
search the database for a set of video event data that describes a set of panelist IDs' interactions with the given set of videos, wherein data entries pertaining to a same panelist ID are stored contiguously on a single storage node from the plurality of storage nodes;
search the database for a set of non-video event data that describes the set of panelist IDs' interactions with the non-video category IDs;
for a non-video category ID included in the set of non-video event data, determine an overlap value comprising a number of distinct panelist IDs that have interacted with the given set of videos and have performed interactions associated with the non-video category ID within an attribution window;
determine a respective correlation value between the given set of videos and the non-video category ID based at least in part on the overlap value; and
output, at a user interface, the non-video category ID in a ranked list of at least a portion of the non-video category IDs based on the respective correlation value between the given set of videos and the non-video category ID; and
a memory coupled to the processor and configured to provide the processor with instructions.