US 10,872,383 B2Granted

Using a model to estimate a payment delinquency for an invoice

AssigneeOracle International Corporation

InventorsTapomoy Dey, Aaron P. Thomas

Life of this patent

Estimating the term…

Loading patent…

Abstract

Using a model to estimate a payment delinquency for an invoice is described. Multiple delinquency estimation models are generated based on a training set of invoices. Each delinquency estimation model estimates a respective payment delinquency for a particular invoice. A refined payment delinquency is determined for the particular invoice based on respective payment delinquencies determined by the delinquency estimation models. Information associated with a set of invoices may be presented at an interface based on the respective payment delinquencies. A payment reminder application may be programmed to process the invoices according to payment reminder schedules determined based on the respective payment delinquencies. Additionally, invoices may be added to the training set of invoices to update the delinquency estimation models. Invoices that were processed according to a payment reminder schedule that was modified based on the respective payment delinquencies are not included in the training set of invoices.

Claims21

Dependent claims

A non-transitory computer readable medium comprising instructions which, when executed by one or more hardware processors, cause performance of operations comprising:

identifying a training set of invoices, for training a set of delinquency estimation models, wherein the training set of invoices is associated with a set of one or more invoice characteristics, and the training set of invoices indicate a respective first set of values for the set of invoice characteristics;
generating a first delinquency estimation model using machine learning based on the training set of invoices;
generating a second delinquency estimation model using machine learning based on the training set of invoices;
determining a set of respective estimated payment delinquencies for a target set of invoices using the first delinquency estimation model and the second delinquency estimation model;
identifying a first subset of the target set of invoices for which respective payment reminder schedules were modified based on the respective estimated payment delinquencies for the first subset of the target set of invoices;
identifying a second subset of the target set of invoices for which respective payment reminder schedules were not modified based on the respective estimated payment delinquencies for the second subset of the target set of invoices;
identifying a new training set of invoices, for training the set of delinquency estimation models, that includes the second subset of the target set of invoices but does not include the first subset of the target set of invoices, wherein the new training set of invoices is associated with the set of invoice characteristics, and the new training set of invoices indicate a respective second set of values for the set of invoice characteristics;
updating the first delinquency estimation model using machine learning based on a first subset of the new training set of invoices;
updating the second delinquency estimation model using machine learning based on a second subset of the new training set of invoices.

6 more elements

A non-transitory computer readable medium comprising instructions which, when executed by one or more hardware processors, cause performance of operations comprising:

identifying a training set of invoices, for training a delinquency estimation model ensemble, wherein the training set of invoices is associated with a set of one or more invoice characteristics;
generating the delinquency estimation model ensemble using machine learning, wherein generating the delinquency estimation model ensemble comprises:
generating a first delinquency estimation model based on a first subset of the training set of invoices, wherein the first delinquency estimation model determines a first estimated payment delinquency for a particular invoice using a first subset of the set of invoice characteristics;
generating a second delinquency estimation model based on a second subset of the training set of invoices, wherein the second delinquency estimation model determines a second estimated payment delinquency for the particular invoice using a second subset of the set of invoice characteristics, and the first subset of the set of invoice characteristics and the second subset of the set of invoice characteristics are different;
forming the delinquency estimation model ensemble using at least the first delinquency estimation model and the second delinquency estimation model;
applying the delinquency estimation model ensemble to a first target invoice, wherein applying the delinquency estimation model ensemble to the first target invoice comprises:
applying the first delinquency estimation model to values for the first subset of the set of invoice characteristics for a first target invoice to estimate a first estimated payment delinquency for the first target invoice;
applying the second delinquency estimation model to values for the second subset of the set of invoice characteristics for the first target invoice to estimate a second estimated payment delinquency for the first target invoice;
determining a first refined estimated payment delinquency for the first target invoice based at least on the first estimated payment delinquency and the second estimated payment delinquency;
applying the delinquency estimation model ensemble to a second target invoice, wherein applying the delinquency estimation model ensemble to the second target invoice comprises:
applying the first delinquency estimation model to values for the first subset of the set of invoice characteristics for a second target invoice to estimate a third estimated payment delinquency for the second target invoice;
applying the second delinquency estimation model to values for the second subset of the set of invoice characteristics for the second target invoice to estimate a fourth estimated payment delinquency for the second target invoice;
determining a second refined estimated payment delinquency for the second target invoice based at least on the third estimated payment delinquency and the fourth estimated payment delinquency;
responsive to determining that the first refined estimated payment delinquency for the first target invoice is above a threshold value:
selecting a first schedule of a plurality of schedules;
programming an application to transmit a first set of reminders for the first target invoice according to the first schedule;
responsive to determining that the second refined estimated payment delinquency for the second target invoice is below the threshold value:
selecting a second schedule of the plurality of schedules;
programming the application to transmit a second set of reminders for the second target invoice according to the second schedule.

16 more elements

A non-transitory computer readable medium comprising instructions which, when executed by one or more hardware processors, cause performance of operations comprising:

identifying a training set of invoices, for training a delinquency estimation model ensemble, wherein the training set of invoices is associated with a set of one or more invoice characteristics;
generating the delinquency estimation model ensemble using machine learning, wherein generating the delinquency estimation model ensemble comprises:
generating a first delinquency estimation model based on a first subset of the training set of invoices, wherein the first delinquency estimation model determines a first estimated payment delinquency for a particular invoice using a first subset of the set of invoice characteristics;
generating a second delinquency estimation model based on the second subset of the training set of invoices, wherein the second delinquency estimation model determines a second estimated payment delinquency for the particular invoice using a second subset of the set of invoice characteristics, and the first subset of the set of invoice characteristics and the second subset of the set of invoice characteristics are different;
forming the delinquency estimation model ensemble using at least the first delinquency estimation model and the second delinquency estimation model;
applying the delinquency estimation model ensemble to a target set of invoices, wherein applying the delinquency estimation model ensemble to the target set of invoices comprises;
applying the first delinquency estimation model to values for the first subset of the set of invoice characteristics for the target set of invoices to estimate a first set of respective estimated payment delinquencies for the target set of invoices;
applying the second delinquency estimation model to values for the second subset of the set of invoice characteristics for the target set of invoices to estimate a second set of respective estimated payment delinquencies for the target set of invoices;
determining a set of respective refined estimated payment delinquencies for the target set invoices based at least on the first set of respective estimated payment delinquencies and the second set of respective estimated payment delinquencies;
identifying a subset of the target set of invoices that is associated with respective refined estimated payment delinquencies that are above a threshold value;
presenting, at a user interface, information associated with the subset of the target set of invoices.

8 more elements

Dependencies are inferred from the claim text.

Facts

Application: 15/893,446Filed 9 Feb 2018 · us provisional application
Priority: 14 Feb 2017
Published: US 2018/0232814 A116 Aug 2018
Granted: 22 Dec 2020
Term ends: Not estimatedThe source records don’t support an estimate.

Classification4 codes

G06Q 40/12
Finance; Insurance; Tax strategies; Processing of corporate or income taxes: accounting
G06Q 10/067
Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling: enterprise or…
G06Q 40/10
Finance; Insurance; Tax strategies; Processing of corporate or income taxes: tax strategies
G06Q
Information and communication technology [ict] specially adapted for administrative, commercial, financial, managerial

Family

Application62/458,638Us provisional application · filed 14 Feb 2017
Application62/468,199Us provisional application · filed 7 Mar 2017
US 2018/0232814 A1Pre-grant publication · published 16 Aug 2018
US 10,872,383 B2This patent

From the US filing history. Foreign family members aren’t listed.

Timeline

Citations

Open elsewhere

Building an agent or pulling many patents? Use the API or MCP server instead of scraping pages. Get an API key