US 12,131,233 B1
Systems and methods for auto-deployment of a machine learning component within a pipelined search query
Chinmay Madhav Kulkarni, Sunnyvale, CA (US); Lin Ma, Burnaby (CA); Amir Malekpour, Vancouver (CA); Mohan Rajagopalan, Mountain View, CA (US); John C. Reed, Saratoga, CA (US); and Ram Sriharsha, Oakland, CA (US)
Assigned to Splunk Inc., San Francisco, CA (US)
Filed by SPLUNK Inc., San Francisco, CA (US)
Filed on Oct. 19, 2020, as Appl. No. 17/074,407.
Int. Cl. G06N 20/00 (2019.01); G06F 16/245 (2019.01); G06F 18/214 (2023.01); G06N 3/08 (2023.01); G06N 5/025 (2023.01); G06N 5/04 (2023.01)
CPC G06N 20/00 (2019.01) [G06F 16/245 (2019.01); G06F 18/2148 (2023.01); G06N 3/08 (2013.01); G06N 5/025 (2013.01); G06N 5/04 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method, comprising:
identifying a sequence of operators forming at least a portion of a pipelined search query, wherein the sequence of operators includes a source operator, a first operator, a second operator, and a destination operator, wherein a set of data or results produced through execution of the first operator is provided as input to the second operator, wherein the source operator is associated with a collection of data for the query and the destination operator is associated with providing resultant data generated by the pipelined search query to a destination, and wherein the second operator is configured to perform operations, in accordance with a machine learning component, on data received as the input based on execution of the first operator;
identifying a pipeline fragment including the first operator and the second operator;
retrieving an input schema and an output schema associated with the machine learning component from metadata stored in non-transitory storage medium;
determining compatibility between an output schema associated with the first operator and the input schema associated with the machine learning component; and
storing a representation of the pipeline fragment in a data store for subsequent use.