US 11,818,105 B2
Systems and methods for encoded communications
Michael Mossoba, Arlington, VA (US); Joshua Edwards, Philadelphia, PA (US); and Salik Shah, Washington, DC (US)
Assigned to Capital One Services, LLC, McLean, VA (US)
Filed by Capital One Services, LLC, McLean, VA (US)
Filed on May 21, 2021, as Appl. No. 17/326,765.
Application 17/326,765 is a continuation of application No. 16/280,606, filed on Feb. 20, 2019, granted, now 11,019,039.
Application 16/280,606 is a continuation of application No. 16/058,470, filed on Aug. 8, 2018, granted, now 10,250,574, issued on Apr. 2, 2019.
Prior Publication US 2021/0281549 A1, Sep. 9, 2021
Int. Cl. H04L 29/06 (2006.01); H04L 9/40 (2022.01); G06N 5/02 (2023.01); G06Q 10/067 (2023.01); G06F 21/62 (2013.01); G06F 16/248 (2019.01); G06F 16/9535 (2019.01)
CPC H04L 63/0428 (2013.01) [G06F 16/248 (2019.01); G06F 16/9535 (2019.01); G06F 21/6245 (2013.01); G06N 5/02 (2013.01); G06Q 10/067 (2013.01); H04L 63/083 (2013.01); H04L 63/0876 (2013.01)] 18 Claims
OG exemplary drawing
 
1. A method for training an encoded communications module comprising:
providing an encoded communications module of at least one server;
receiving training data comprising a financial query and a plurality of associated inputs;
determining the semantic meaning of the received training data by generating a vector representation for each financial query among the plurality of financial queries, wherein the vector representation comprises a node for each word in the financial query, and wherein the plurality of associated inputs are within a threshold of semantic meaning and the financial query is outside the threshold of semantic meaning; and
training the encoded communications module to encode communications based on the received training data and the determined semantic meaning.