US 12,450,622 B2
Automated customer engagement prediction and classification
Siddharth Narayanan, Cambridge, MA (US); Amin Assareh, Boston, MA (US); and Wenlu Yan, Poway, CA (US)
Assigned to FMR LLC, Boston, MA (US)
Filed by FMR LLC, Boston, MA (US)
Filed on Feb. 10, 2023, as Appl. No. 18/108,113.
Prior Publication US 2024/0273563 A1, Aug. 15, 2024
Int. Cl. G06Q 30/00 (2023.01); G06Q 10/04 (2023.01); G06Q 30/0204 (2023.01)
CPC G06Q 30/0205 (2013.01) [G06Q 10/04 (2013.01)] 18 Claims
OG exemplary drawing
 
1. A computerized method of automated customer engagement prediction and classification, the method comprising:
generating, by the server computing device, a feature vector for each user of a plurality of users comprising a multidimensional array of variables corresponding to at least a portion of historical user activity data for the user, the historical user activity data comprising transaction data, demographic data, and user contact channel response data;
encoding, by the server computing device for each feature vector, each variable in the feature vector into a corresponding weight-of-evidence value;
transforming, by the server computing device, each encoded feature vector into an embedding comprising a multidimensional array with fewer variables than the encoded feature vector and positioning the embeddings in a multidimensional vector space;
generating, by the server computing device for each user, a user engagement probability value by identifying one or more embeddings of other users in proximity to the embedding of the user using a similarity measure and determining an engagement outcome for the users associated with the identified embeddings;
assigning, by the server computing device, each user to an engagement probability cluster based upon the user engagement probability value for the user;
determining, by the server computing device, a contact channel for each user based upon the engagement probability cluster to which the user is assigned; and
transmitting, by the server computing device, data associated with the contact channel for each user to a remote computing device;
wherein the remote computing device processes the data associated with the contact channel for each user to:
automatically populate a user interface of an autodialing software application on the remote computing device with the data associated with the contact channel for each user; and
simultaneously initiate a telephone call from the remote computing device to a client computing device associated with a first user via the autodialing software application.