US 11,934,974 B2
Intelligent integrated remote reporting system
Gaurav Bansal, Telangana (IN); Nikhil Pathak, Uttar Pradesh (IN); and Raja Venkatesh Gottumukkala, Chennai (IN)
Assigned to Bank of America Corporation, Charlotte, NC (US)
Filed by Bank of America Corporation, Charlotte, NC (US)
Filed on Oct. 5, 2021, as Appl. No. 17/494,116.
Prior Publication US 2023/0116238 A1, Apr. 13, 2023
Int. Cl. G10L 15/18 (2013.01); G06N 20/20 (2019.01); G06Q 10/063 (2023.01); G10L 15/22 (2006.01)
CPC G06Q 10/063 (2013.01) [G06N 20/20 (2019.01); G10L 15/18 (2013.01); G10L 15/22 (2013.01); G10L 2015/223 (2013.01)] 18 Claims
OG exemplary drawing
 
1. A method for intelligent integrated remote reporting, the method comprising:
at an edge device:
receiving a natural language request from a user for remote generation of a report;
using a first machine learning model, generating a first ranking comprising a set of existing reports, the first ranking based at least in part on past database activity by the user; and
displaying the first ranking to the user and receiving an input from the user rejecting the set of existing reports; and
at an enterprise server:
in response to rejection of the set of existing reports, mapping the natural language request to one or more data sets;
using a second machine learning model, generating a second ranking comprising the mapped data sets and one or more additional data sets, the second ranking based at least in part on past database activity by the user, past database activity by a set of enterprise users, and a relationship between a mapped data set and the additional data sets, the past database activity comprising frequency of access, the relationship comprising a relationship between a column in a mapped data set and a column in one of the additional data sets;
displaying the second ranking to the user and receiving a user selection of one or more of the ranked data sets from the edge device;
based on user selection of a ranked data set, displaying a set of report schema comprising delivery method, level of aggregation, and column format, and receiving a user selection of displayed report schema from the edge device;
transmitting a request to an integrated reporting platform in JavaScript Object Notation (JSON) format, the request comprising the selected data set and report schema; and
using the integrated reporting platform:
converting the JSON request to executable instructions, the instructions formatted based at least in part on a data structure associated with the selected data set;
generating a report comprising the selected report schema populated with the selected data set; and
transmitting feedback associated with the selected data set to update the first machine learning model and the second machine learning model.