US 12,468,742 B2
Text-triggered database and API actions
Qian Rui Chow, Vancouver (CA); Donald Creig Humes, Yorktown, VA (US); Kaarthik Balasubrahmanian, Belmont, CA (US); Sridhar Tadepalli, Bangalore (IN); Saravanan Anandan, Fremont, CA (US); and Kartik Raghavan, Pleasanton, CA (US)
Assigned to Oracle International Corporation, Redwood Shores, CA (US)
Filed by Oracle International Corporation, Redwood Shores, CA (US)
Filed on Feb. 13, 2023, as Appl. No. 18/168,358.
Claims priority of provisional application 63/358,789, filed on Jul. 6, 2022.
Prior Publication US 2024/0012837 A1, Jan. 11, 2024
Int. Cl. G06F 16/31 (2019.01); G06F 16/383 (2019.01); G06F 40/30 (2020.01); G06N 20/00 (2019.01)
CPC G06F 16/313 (2019.01) [G06F 16/383 (2019.01); G06F 40/30 (2020.01); G06N 20/00 (2019.01)] 19 Claims
OG exemplary drawing
 
1. One or more non-transitory computer readable media comprising instructions which, when executed by one or more hardware processors, causes performance of operations comprising:
training a machine learning model to generate a recommendation to update a target contact list based on contact information comprised in a set of human-understandable text, the training comprising:
obtaining training data sets, each training data set of historical data comprising:
historical contact information;
historical metadata associated with the historical contact information; and
a historical modification to a historical contact list based on the historical contact information and/or the historical metadata; and
training the machine learning model based on the training data sets;
obtaining the set of human-understandable text;
applying a semantic analysis engine to the human-understandable text to identify the contact information comprised in the set of human-understandable text;
at runtime: selecting the target contact list, of a plurality of contact lists, to be updated based on the contact information in the set of human-understandable text, the selecting operation comprising:
analyzing metadata corresponding to the human-understandable text to determine that the human-understandable text is mapped to a first data object of a first type;
identifying a relationship between the first data object and the target contact list; and
selecting the target contact list based on the relationship between the first data object and the target contact list;
identifying a database schema associated with the target contact list;
applying the machine learning model to the contact information to generate a particular recommendation for modifying at least the target contact list;
based at least in part on the particular recommendation, generating a first operation to update the target contact list based on the contact information comprised in the set of human-understandable text and the database schema associated with the target contact list; and
executing the first operation to update the target contact list.