| CPC G06Q 10/02 (2013.01) [G06F 18/2148 (2023.01); G06N 3/08 (2013.01); G06Q 50/205 (2013.01); G09B 7/00 (2013.01); G06Q 10/0635 (2013.01)] | 19 Claims |
|
1. A system comprising:
a data store coupled to a computer network and storing:
a plurality of entity data associated with an exam registration event, and comprising:
a candidate data corresponding to a candidate;
an exam data corresponding to an exam;
a test center data corresponding to a test center, including a trend data associated with each of a plurality of testing facilities identified within the test center data; and
an exam registration event data related to an exam registration event during which the candidate registered to take the exam at the test center;
a plurality of machine learning models including a set of entity-level machine learning models and an aggregate machine learning model, wherein the set of entity-level machine learning models includes a candidate machine learning model, a test center machine learning model, an exam machine learning model, an exam registration event machine learning model, an exam delivery event machine learning model, a proctor machine learning model, and a country machine learning model;
a plurality of predefined actions;
a threshold which, when exceeded, triggers a recommendation for at least one predefined action in the plurality of predefined actions; and
a server comprising a server hardware computing device coupled to a network and including one or more electronic processors executing instructions within a memory that, when executed, cause the system to:
receive a notification indicating that the exam registration event is in a real-time exam registration state, wherein the exam registration event is in the real-time exam registration state while a candidate is actively registering for an exam;
responsive to the notification, identify a set of relevant entities for the exam registration event based on the exam registration event being in the real-time exam registration state, wherein the set of relevant entities includes the candidate, the exam, the test center, and the exam registration event;
retrieve, from the data store, at least a portion of the entity data associated with the set of relevant entities;
select, from the data store, at least one entity-level machine learning model from the set of entity-level machine learning models, wherein the at least one entity-level machine learning model is selected based on corresponding entity data types of the at least a portion of one entity data, wherein the at least one entity-level machine learning model is trained using entity data type specific training data to generate entity-level risk scores for the corresponding entity data type;
apply the at least one entity-level machine learning model to the at least a portion of the entity data to generate a set of entity-level risk scores, wherein the set of entity-level risk scores includes at least one of a candidate risk score, an exam risk score, a test center risk score, or an exam registration event risk score;
apply the aggregate machine learning model to the set of entity-level risk scores to generate an exam registration situational risk score; and
compare the exam registration situational risk score to the threshold; and
responsive to the exam registration situational risk score exceeding the threshold, update a database entry associated in the data store with the at least one entity data to prevent the candidate from completing registration for the exam as include the at least one predefined action included in the plurality of predefined actions.
|