US 11,704,307 B2
Intelligent query editor using neural network based machine learning
Arvind Kumar Maheshwari, Pleasanton, CA (US); Vamshidhar Reddy Pasham, Hyderabad (IN); and Ravi Kumar Agrawal, Noida (IN)
Assigned to Oracle International Corporation, Redwood Shores, CA (US)
Filed by Oracle International Corporation, Redwood Shores, CA (US)
Filed on Dec. 23, 2020, as Appl. No. 17/133,278.
Prior Publication US 2022/0197900 A1, Jun. 23, 2022
Int. Cl. G06F 16/00 (2019.01); G06F 16/242 (2019.01); G06N 3/08 (2023.01)
CPC G06F 16/2423 (2019.01) [G06N 3/08 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method comprising:
training a neural network using a set of queries that have been labeled as performant, wherein the queries are automatically labeled based on runtime performance metrics associated with executing the queries within a computing environment;
receiving, from a user, an incomplete query comprising a first set of one or more query tokens;
responsive to receiving the incomplete query, generating, by the neural network model based at least in part on the first set of one or more query tokens, a prediction of a second set of one or more query tokens for completing the incomplete query to form a completed query;
identifying a set of one or more queries that satisfy a similarity threshold relative to the completed query; and
presenting, to the user, the completed query and the set of one or more queries that satisfy the similarity threshold relative to the completed query.