US 11,657,402 B2
Dynamic claims submission system
Theodore Harris, San Francisco, CA (US); Yue Li, Sunnyvale, CA (US); Craig O'Connell, San Mateo, CA (US); and Tatiana Korolevskaya, Mountain View, CA (US)
Assigned to Visa International Service Association, San Francisco, CA (US)
Appl. No. 16/610,860
Filed by Visa International Service Association, San Francisco, CA (US)
PCT Filed May 16, 2017, PCT No. PCT/US2017/032972
§ 371(c)(1), (2) Date Nov. 4, 2019,
PCT Pub. No. WO2018/212767, PCT Pub. Date Nov. 22, 2018.
Prior Publication US 2021/0142333 A1, May 13, 2021
Int. Cl. G06Q 30/016 (2023.01); G06F 16/242 (2019.01); G06N 20/00 (2019.01); G06Q 10/10 (2023.01); G06F 40/205 (2020.01)
CPC G06Q 30/016 (2013.01) [G06F 16/243 (2019.01); G06N 20/00 (2019.01); G06Q 10/10 (2013.01); G06F 40/205 (2020.01)] 17 Claims
OG exemplary drawing
 
1. A method comprising:
a) receiving, by a first computer, data relating to a claim submission from a second computer, wherein the data relating to the claim submission includes claims submission data input by a user, information relating to the user, and one or more features, wherein the claim submission is a request for something that is owed to the user, the one or more features including detected pauses between typed inputs of the user to determine if words have been cut and pasted by the user into a user interface of a user device;
b) storing, by the first computer, the data relating to the claim submission in a first database;
c) retrieving, by the first computer, from the first database, data associated with the one or more features, wherein the data associated with the one or more features is determined from an artificial intelligence model, wherein the artificial intelligence model is built by
accessing a graph database to obtain a plurality of topological graphs,
generating a plurality of edges associated with nodes in the topological graphs to build a conceptual graph, and
inputting interview outcome data, user response data, and the conceptual graph into a real time simulator to build one or more sequence graphs, the one or more sequence graphs used to generate the artificial intelligence model;
d) retrieving, by the first computer, from a second database, data associated with the information relating to the user;
e) generating, by the first computer, a first score based on the data input by the user, the data associated with the one or more features and the data associated with the information relating to the user, the first score evaluating whether the data input by the user is fraudulent;
f) determining, by the first computer, an interview script based at least upon the first score;
g) providing, by the first computer, a first question in the interview script to the second computer;
h) receiving, by the first computer, from the second computer, a response to the first question
i) generating, by the first computer, a second score based at least upon the data in the first database and the response to the first question, and then updating, by the first computer, the artificial intelligence model using at least the second score and the response to the first question;
j) updating, by the first computer, the interview script based at least upon the second score and the updated artificial intelligence model; and
k) providing, by the first computer, a second question in the updated interview script to the second computer, the second question based at least in part upon the second score,
wherein the first computer continues to provide questions to the second computer if a continually updated score remains above a predetermined value, and wherein the user is routed to a live interview with a human representative if the continually updated score drops below another predetermined value.