Evaluation of product-related data structures using machine-learning techniques

Patent metadata

Field	Value
Publication number	US-11615366-B2
Application number	US-202016849199-A
Country	US
Kind code	B2
Filing date	Apr 15, 2020
Priority date	Apr 15, 2020
Publication date	Mar 28, 2023
Grant date	Mar 28, 2023

Abstract

Artificial intelligence (AI)-based techniques are provided that predict a quality score for a product-related data structure associated with one or more products. One method comprises obtaining data for a given product-related data structure; evaluating a plurality of first features related to a customer account associated with the given product-related data structure using the obtained data; evaluating a plurality of second features related to the given product-related data structure using the obtained data; processing at least some of the first features and the second features using at least one model that provides a predicted quality score for the given product-related data structure; and applying one or more thresholds to the predicted quality score to determine an acceptance status related to the given product-related data structure. A weighting of the first features and the second features can be learned during a training phase.

First claim

What is claimed is:

1. A method, comprising:
obtaining data for a given product-related data structure;
evaluating a plurality of first features related to a customer account associated with the given product-related data structure using the obtained data;
evaluating a plurality of second features using the obtained data from the given product-related data structure;
training at least one machine learning model during a training phase by evaluating the plurality of first features and the plurality of second features using historical training data comprising at least one acceptance status label such that the at least one machine learning model learns (i) to predict a predicted quality score and (ii) a weighting of one or more first features and one or more second features, wherein the weighting is based at least in part on a feature importance of the one or more first features and a feature importance of the one or more second features, wherein the weighting comprises a first weighting of the one or more first features used to calculate a first score and a second weighting of the one or more second features used to calculate a second score;
implementing the at least one machine learning model using at least one processing device comprising a processor coupled to a memory;
applying the one or more first features and the one or more second features to the at least one machine learning model that predicts the predicted quality score for the given product-related data structures;
predicting, using the at least one machine learning model, the predicted quality score for the given product-related data structure, wherein the predicted quality score for the given product-related data structure comprises an aggregation based at least in part on the first score and the second score;
applying, using the at least one processing device, one or more thresholds to the predicted quality score to automatically determine an acceptance status related to the given product-related data structure; and
automatically initiating a processing of the given product-related data structure based at least in part on one or more of the acceptance status and the predicted quality score, wherein the automatically initiating the processing of the given product-related data structure comprises one or more of: (i) initiating a generation of an automated acceptance related to the given product-related data structure based at least in part on the acceptance status; (ii) initiating a generation of an automated denial related to the given product-related data structure based at least in part on the acceptance status; and (iii) initiating a prioritization of the given product-related data structure for a manual review based at least in part on the predicted quality score.

2. The method of claim 1 , wherein the acceptance status comprises one or more of an automatically accepted status, an automatically denied status and an additional review required status.

3. The method of claim 2 , wherein the acceptance status of the given product-related data structure comprises the automatically accepted status in response to the predicted quality score for the given product-related data structure exceeding a corresponding acceptance threshold, and wherein the acceptance status of the given product-related data structure comprises the automatically denied status in response to the predicted quality score for the given product-related data structure being below a corresponding denial threshold.

4. The method of claim 3 , wherein the acceptance status of the given product-related data structure comprises the additional review required status in response to the predicted quality score for the given product-related data structure being between the corresponding acceptance threshold and the corresponding denial threshold.

5. The method of claim 1 , further comprising at least one statistical model that determines a second predicted quality score for the given product-related data structure and further comprising selecting one of (i) the predicted quality score for the given product-related data structure predicted by the at least one machine learning model and (ii) the second predicted quality score for the given product-related data structure determined by the at least one statistical model.

6. The method of claim 1 , wherein the at least one machine learning model compares the given product-related data structure to similar product-related data structures based on one or more similarity criteria to evaluate a quality of the given product-related data structure.

7. The method of claim 1 , wherein the aggregation comprises an aggregation of at least two of an account score, a product-related data structure score and a product score for at least one product associated with the given product-related data structure.

8. The method of claim 1 , wherein the given product-related data structure comprises a plurality of items, and wherein a predicted quality score is separately determined for each of the plurality of items and wherein the aggregation further comprises a weighted aggregation of the predicted quality score for each of the plurality of items.

9. An apparatus comprising:
at least one processing device comprising a processor coupled to a memory;
the at least one processing device being configured to implement the following steps:
obtaining data for a given product-related data structure;
evaluating a plurality of first features related to a customer account associated with the given product-related data structure using the obtained data;
evaluating a plurality of second features using the obtained data from the given product-related data structure;
training at least one machine learning model during a training phase by evaluating the plurality of first features and the plurality of second features using historical training data comprising at least one acceptance status label such that the at least one machine learning model learns (i) to predict a predicted quality score and (ii) a weighting of one or more first features and one or more second features, wherein the weighting is based at least in part on a feature importance of the one or more first features and a feature importance of the one or more second features, wherein the weighting comprises a first weighting of the one or more first features used to calculate a first score and a second weighting of the one or more second features used to calculate a second score;
implementing the at least one machine learning model using the at least one processing device;
applying the one or more first features and the one or more second features to the at least one machine learning model that predicts the predicted quality score for the given product-related data structure;
predicting, using the at least one machine learning model, the predicted quality score for the given product-related data structure, wherein the predicted quality score for the given product-related data structure comprises an aggregation based at least in part on the first score and the second score;
applying, using the at least one processing device, one or more thresholds to the predicted quality score to automatically determine an acceptance status related to the given product-related data structure; and
automatically initiating a processing of the given product-related data structure based at least in part on one or more of the acceptance status and the predicted quality score, wherein the automatically initiating the processing of the given product-related data structure comprises one or more of: (i) initiating a generation of an automated acceptance related to the given product-related data structure based at least in part on the acceptance status; (ii) initiating a g

Assignees

• Emc Ip Holding Co Llc

Inventors

• Savir Amihai
• Wensing Arthur
• Gershon Noga
• Körner Marcel Bernard
• Mlynek Ivan
• Perez Jorge Luis
• Thatcher Michael Rupert James
• Kollannur Dhev George
• Sagi Omer

Classifications

• G06Q30/0185

  Product, service or business identity fraud · CPC title

• G06Q10/087

  Inventory or stock management, e.g. order filling, procurement or balancing against orders · CPC title

• G06Q10/10

  Office automation; Time management · CPC title

• G06N5/01

  Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound · CPC title

• G06Q10/06395 Primary

  Quality analysis or management · CPC title