US-12619918-B2 - Predicting targeted future engagement using trained artificial intelligence processes

US12619918B2US 12619918 B2US12619918 B2US 12619918B2US-12619918-B2

Abstract

The disclosed embodiments include computer-implemented processes that determine, in real time, a likelihood of a targeted future engagement using trained artificial intelligence processes. For example, an apparatus may generate a first input dataset based on elements of first interaction data associated with a first temporal interval, and based on an application of a trained first artificial intelligence process to the first input dataset, generate output data representative of a predicted likelihood of an occurrence of each of a plurality of target events during a second temporal interval. The second temporal interval is subsequent to the first temporal interval and is separated from the first temporal interval by a corresponding buffer interval. Further, the apparatus may transmit at least a portion of the output data to a computing system, which may generate notification data associated with the predicted likelihood, and provision the notification data to a device.

Inventors

Patrick James WHELAN
Anson Wah Chun WONG
Maksims Volkovs
Tomi Johan Poutanen

Assignees

THE TORONTO-DOMINION BANK

Dates

Publication Date: 20260505
Application Date: 20211117

Claims (20)

1 . An apparatus, comprising: a memory storing instructions; a communications interface; and at least one processor coupled to the memory and the communications interface, the at least one processor being configured to execute the instructions to: perform operations that train an artificial intelligence process using a plurality of training datasets and corresponding elements of ground-truth data, each of the training datasets being associated with a prior temporal interval, and for each of the training datasets, the elements of ground-truth data indicate an occurrence or a non-occurrence of each of a plurality of target events during a portion of the prior temporal interval; receive, via the communications interface, an identifier associated with a customer from a computing system, and based on the received identifier, obtain, from the memory, first elements of consolidated data associated with a first temporal interval and with the received identifier; generate a first input dataset based on the first elements of consolidated data; based on an application of a trained first artificial intelligence process to the first input dataset, generate output data representative of a predicted likelihood of an occurrence of each of the plurality of target events during a second temporal interval, the second temporal interval being subsequent to the first temporal interval and being separated from the first temporal interval by a corresponding buffer interval; and transmit at least a portion of the output data to the computing system via the communications interface, the computing system being configured to generate, based on the portion of the output data, notification data associated with the predicted likelihood of the occurrence of at least one of the target events and to provision the notification data to a device.
2 . The apparatus of claim 1 , wherein: the output data comprises a plurality of output data elements; and each of the output data elements comprises an identifier associated with a corresponding one of the target events and a numerical score indicative of the predicted likelihood of the occurrence of the corresponding one of the target events during the second temporal interval.
3 . The apparatus of claim 2 , wherein the at least one processor is further configured to execute the instructions to: perform operations that rank the output data elements in accordance with the numerical scores; and transmit at least a predetermined subset of the ranked output data elements to the computing system via the communications interface.
4 . The apparatus of claim 1 , wherein the trained first artificial intelligence process comprises a trained, gradient-boosted, decision-tree process.
5 . The apparatus of claim 1 , wherein the at least one processor is further configured to: obtain (i) one or more parameters that characterize the trained first artificial intelligence process and (ii) data that characterizes a composition of the first input dataset; generate the first input dataset in accordance with the data that characterizes the composition; and apply the trained first artificial intelligence process to the first input dataset in accordance with the one or more parameters.
6 . The apparatus of claim 5 , wherein the at least one processor is further configured to: based on the data that characterizes the composition, perform operations that at least one of extract a first feature value from the first elements of consolidated data or compute a second feature value based on the first feature value; and generate the first input dataset based on at least one of the extracted first feature value or the computed second feature value.
7 . The apparatus of claim 1 , wherein the at least one processor is further configured to execute the instructions to: obtain second elements of consolidated data and elements of targeting data, each of the second elements of consolidated data comprising a temporal identifier associated with a corresponding one of the prior temporal intervals, and the elements of targeting data comprising event identifiers associated with the plurality of target events; based on the temporal identifiers, determine that a first subset of the second elements of consolidated data are associated with a prior training interval, and that a second subset of the second elements of consolidated data are associated with a prior validation interval; and generate a plurality of training datasets based on the event identifiers and on corresponding portions of the first subset, and perform the operations that train the first artificial intelligence process using on the training datasets.
8 . The apparatus of claim 7 , wherein the at least one processor is further configured to execute the instructions to: generate a plurality of validation datasets based the event identifiers and on corresponding portions of the second subset; apply the trained first artificial intelligence process to the plurality of validation datasets, and generate additional elements of output data based on the application of the trained first artificial intelligence process to the plurality of validation datasets; compute one or more validation metrics based on the additional elements of output data; and based on a determined consistency between the one or more validation metrics and a threshold condition, validate the trained first artificial intelligence process.
9 . The apparatus of claim 1 , wherein the at least one processor is further configured to execute the instructions to: obtain (i) one or more parameters that characterize a trained second artificial intelligence process and (ii) data that characterizes a composition of a second input dataset associated with the trained second artificial intelligence process; based on the first elements of consolidated data, generate the second input dataset in accordance with the data that characterizes the composition; apply the trained second artificial intelligence process to the second input dataset in accordance with the one or more parameters, and based on the application of the trained second artificial intelligence process to the second input dataset, generate additional output data representative of a predicted likelihood of an occurrence of an event associated with a plurality of target event subgroups during the second temporal interval; and transmit at least a portion of the additional output data to the computing system via the communications interface.
10 . The apparatus of claim 1 , wherein: the plurality of target events comprise a plurality of target engagement events associated with the customer; and the first elements of consolidated data comprise one or more elements of activity data associated with the customer, the one or more elements of activity data characterizing an occurrence of at least one of the target engagement events during the first temporal interval.
11 . The apparatus of claim 1 , wherein the at least one processor is further configured to execute the instructions to: receive interaction data associated with the first temporal interval from a plurality of computing systems via the communications interface; generate a plurality of elements of consolidated data based on an application of one or more pre-processing operations to the interaction data; and store, within the memory, each of the plurality of elements of consolidated data elements, each of the plurality of elements of consolidated data being associated with the first temporal interval and with an identifier of a corresponding device, and the stored elements of consolidated data comprising the first elements of consolidated data.
12 . The apparatus of claim 1 , wherein the at least one processor is further configured to execute the instructions to: receive, via the communications interface, a plurality of additional identifiers from the computing system, each of the additional identifiers being associated with a corresponding one of a plurality of customers; obtain, from the memory, subsets of the elements of consolidated data associated with the first temporal interval and with corresponding ones of the additional identifiers; and generate an additional input dataset based on each of the subsets, and based on an application of the trained first artificial intelligence process to the each of additional input datasets, generate, for corresponding ones of the additional identifiers, corresponding elements of additional output data representative of a predicted likelihood of an occurrence of each of the plurality of target events during the second temporal interval.
13 . The apparatus of claim 1 , wherein: the trained artificial intelligence process is characterized by a value of one or more parameters; and the at least one processor is further configured to execute the instructions to: perform operations that apply the trained artificial intelligence process to a corresponding plurality of validation datasets in accordance with the one or more parameter values, and that generate elements of validation output data based on the application of the trained artificial intelligence process to the plurality of validation datasets; compute a value of a validation metric based on the elements of validation output data, and based on a determined inconsistency between the validation metric value and a threshold condition, update at least of the parameter values that characterize the trained artificial intelligence process; and apply the trained first artificial intelligence process to the first input dataset in accordance with the at least one of the updated parameter values.
14 . A computer-implemented method, comprising: performing operations, using at least one processor, that train an artificial intelligence process using a plurality of training datasets and corresponding elements of ground-truth data, each of the training datasets being associated with a prior temporal interval, and for each of the training datasets, the elements of ground-truth data indicate an occurrence or a non-occurrence of each of a plurality of target events during a portion of the prior temporal interval; receiving an identifier associated with a customer from a computing system using the at least one processor, and based on the received identifier, obtaining, using the at least one processor, and from a data repository, first elements of consolidated data associated with the received identifier and with a first temporal interval; generating, using the at least one processor, a first input dataset based on the first elements of consolidated data; using the at least one processor, and based on an application of a trained first artificial intelligence process to the first input dataset, generating output data representative of a predicted likelihood of an occurrence of each of the plurality of target events during a second temporal interval, the second temporal interval being subsequent to the first temporal interval and being separated from the first temporal interval by a corresponding buffer interval; and transmitting, using the at least one processor, at least a portion of the output data to the computing system, the computing system being configured to generate, based on the portion of the output data, notification data associated with the predicted likelihood of the occurrence of at least one of the target events and to provision the notification data to a device.
15 . The computer-implemented method of claim 14 , wherein: the output data comprises a plurality of output data elements; and each of the output data elements comprises an identifier associated with a corresponding one of the target events and a numerical score indicative of the predicted likelihood of the occurrence of the corresponding one of the target events during the second temporal interval.
16 . The computer-implemented method of claim 15 , wherein: the computer-implemented method further comprises performing, using the at least one processor, operations that rank the output data elements in accordance with the numerical scores; and the transmitting comprises transmitting at least a predetermined subset of the ranked output data elements to the computing system.
17 . The computer-implemented method of claim 14 , wherein: the trained first artificial intelligence process comprises a trained, gradient-boosted, decision-tree process; and the computer-implemented method further comprises: using the at least one processor, obtaining (i) one or more parameters that characterize the trained first artificial intelligence process and (ii) data that characterizes a composition of the first input dataset; based on the data that characterizes the composition, performing operations, using the at least one processor, that at least one of extract a first feature value from the first elements of consolidated data or compute a second feature value based on the first feature value; and generating, using the at least one processor, the first input dataset based on at least one of the extracted first feature value or the computed second feature value; and applying, using the at least one processor, the trained first artificial intelligence process to the first input dataset in accordance with the one or more parameters.
18 . The computer-implemented method of claim 14 , further comprising: obtaining, using the at least one processor, second elements of consolidated data and elements of targeting data, each of the second elements of consolidated data comprising a temporal identifier associated with a corresponding one of the prior temporal intervals, and the elements of targeting data comprising event identifiers associated with the plurality of target events; based on the temporal identifiers, determining, using the at least one processor, that a first subset of the second elements of consolidated data are associated with a prior training interval, and that a second subset of the second elements of consolidated data are associated with a prior validation interval; and generating, using the at least one processor, a plurality of training datasets based on the event identifiers and on corresponding portions of the first subset, and perform the operations that train the first artificial intelligence process using on the training datasets.
19 . The computer-implemented method of claim 18 , further comprising: generating, using the at least one processor, a plurality of validation datasets based the event identifiers and on corresponding portions of the second subset; using the at least one processor, applying the trained first artificial intelligence process to the plurality of validation datasets, and generating additional elements of output data based on the application of the trained first artificial intelligence process to the plurality of validation datasets; computing, using the at least one processor, one or more validation metrics based on the additional elements of output data; and based on a determined consistency between the one or more validation metrics and a threshold condition, validating the trained first artificial intelligence process using the at least one processor.
20 . The computer-implemented method of claim 14 , further comprising: using the at least one processor, obtain (i) one or more parameters that characterize a trained second artificial intelligence process and (ii) data that characterizes a composition of a second input dataset associated with the trained second artificial intelligence process; based on the first elements of consolidated data, generating, using the at least one processor, the second input dataset in accordance with the data that characterizes the composition; using the at least one processor, applying the trained second artificial intelligence process to the second input dataset in accordance with the one or more parameters, and based on the application of the trained second artificial intelligence process to the second input dataset, generating additional output data representative of a predicted likelihood of an occurrence of an event associated with a plurality of target event subgroups during the second temporal interval; and transmitting, using the at least one processor, at least a portion of the additional output data to the computing system.

Description

CROSS-REFERENCE TO RELATED APPLICATION This application claims the benefit of priority under 35 U.S.C. § 119(e) to prior U.S. Provisional Application No. 63/132,981, filed Dec. 31, 2020, the disclosure of which is incorporated by reference herein to its entirety. TECHNICAL FIELD The disclosed embodiments generally relate to computer-implemented systems and processes that facilitate a prediction of targeted future engagement using trained artificial intelligence processes. BACKGROUND Financial institutions offer a variety of financial products or financial services to their customers, both through in-person branch banking and through various digital channels, and offer a variety of access products that enable these customers to access the financial products or financial services via the various digital channels. Computing systems associated with these financial institution maintain often elements of data that characterize their customers' engagement with these financial products, financial services, or access products, and the elements of maintained, customer-specific data may characterize a time-evolving relationship between the customers and the financial, institutions. SUMMARY In some examples, an apparatus includes a memory storing instructions, a communications interface, and at least one processor coupled to the memory and the communications interface. The at least one processor is configured to execute the instructions to generate a first input dataset based on elements of first interaction data associated with a first temporal interval. Based on an application of a trained first artificial intelligence process to the first input dataset, the at least one processor is further configured to execute the instructions to generate output data representative of a predicted likelihood of an occurrence of each of a plurality of target events during a second temporal interval. The second temporal interval is subsequent to the first temporal interval and is separated from the first temporal interval by a corresponding buffer interval. The at least one processor is further configured to execute the instructions to transmit at least a portion of the output data to a computing system via the communications interface. The computing system is configured to generate, based on the portion of the output data, notification data associated with the predicted likelihood of the occurrence of at least one of the target events and to provision the notification data to a device. In other examples, a computer-implemented method includes generating, using at least one processor, a first input dataset based on elements of first interaction data associated with a first temporal interval. The computer-implemented method also includes, using the at least one processor, and based on an application of a trained first artificial intelligence process to the first input dataset, generating output data representative of a predicted likelihood of an occurrence of each of a plurality of target events during a second temporal interval. The second temporal interval is subsequent to the first temporal interval and is separated from the first temporal interval by a corresponding buffer interval. The computer-implemented method also includes transmitting, using the at least one processor, at least a portion of the output data to a computing system. The computing system is configured to generate, based on the portion of the output data, notification data associated with the predicted likelihood of the occurrence of at least one of the target events and to provision the notification data to a device. Further, in some examples, an apparatus includes a memory storing instructions, a communications interface, and at least one processor coupled to the memory and the communications interface. The at least one processor is configured to execute the instructions to receive, via the communications interface, output data associated with a plurality of target events from a computing system. The output data is generated based on an application of a trained artificial intelligence process to elements of interaction data associated with a first temporal interval, and the output data is representative of a predicted likelihood of an occurrence of each of the target events during a second temporal interval. The second temporal interval is subsequent to the first temporal interval and is separated from the first temporal interval by a corresponding buffer interval. Based on the output data, the at least one processor is further configured to execute the instructions to generate elements of notification data associated with the predicted likelihood of the occurrence of at least one of the target events during the second temporal interval. The at least one processor is further configured to execute the instructions to transmit the elements of notification data to a device via the communications interface. The elements of notification data cause an application prog