Duplicate Invoice Detection via Attribute Segmentation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Detecting and managing duplicate invoices in spend management systems is a complex and time-consuming task due to variations in invoice structure, format, and content, which consumes significant CPU cycles, storage, and memory, especially in environments with large-scale datasets.

Innovation Solution

A duplicate invoice detection subsystem identifies candidate invoices with similar attributes, creates nodes to represent them, generates edges to indicate potential duplicates, and applies filters to confirm duplicates, sending notifications and executing mitigation strategies to delete or block duplicate invoices.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If software inspection of individual invoice documents is performed to detect duplicates, then duplicate invoice detection accuracy is improved, but CPU cycles, storage, memory, and network bandwidth consumption increase significantly

Engineering Contradiction:

Improveduplicate invoice detection accuracyVSAvoidCPU cycles and computational resources

Core Design Contradiction:

Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent segments the invoice inspection process into multiple stages: first extracting key attributes (invoice number, amount, date, supplier) without full document analysis, then performing duplicate detection on these extracted attributes, and only conducting full software inspection on candidate duplicates. This segmentation reduces computational resources while maintaining detection accuracy.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent performs preliminary attribute extraction and comparison before conducting full software inspection. By pre-processing invoices to extract key identifying attributes and performing initial duplicate candidate identification, the system avoids expensive full document inspections for all invoices, thereby reducing CPU and memory usage while preserving detection capability.

Inventive Principle:

Principle #10Preliminary action

2Reliability

If full software inspection of all invoices is performed to ensure accurate duplicate detection, then detection reliability is improved, but system response time deteriorates

Engineering Contradiction:

Improveduplicate invoice detection reliabilityVSAvoidsystem response time

Core Design Contradiction:

ReliabilityVSSpeed

Solution Approach 1:

The inspection process is divided into rapid attribute extraction phase and detailed verification phase. The majority of invoices are processed through the fast attribute extraction and comparison phase, while only potential duplicate candidates proceed to the more time-consuming verification phase, thus improving overall system response time while maintaining reliability.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent applies partial inspection to most invoices (extracting only key attributes) and excessive/detailed inspection only to candidate duplicates. This selective approach ensures high reliability for actual duplicate detection while keeping the average processing time low by not performing full inspection on all invoices.

Inventive Principle:

Principle #16Partial or excessive action

3Adaptability or versatility

If comprehensive attribute comparison is performed on all invoices to handle format variations, then adaptability to different invoice formats is improved, but device complexity increases

Engineering Contradiction:

Improvehandling of different invoice formatsVSAvoidsoftware complexity for format processing

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent implements a universal attribute extraction framework that handles multiple invoice formats through a common interface. By defining a standardized set of key attributes (invoice number, amount, date, supplier) that must be extracted regardless of format, the system achieves adaptability to different formats without proportionally increasing complexity. The extraction module adapts to various formats while the core duplicate detection logic remains unchanged.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS12106384B2Duplicate invoice detection and management

Publication Date: 2024.10.01 COUPA SOFTWARE INC

AI summary

A computer-implemented method for detecting and managing duplicate invoices is provided. In one embodiment, the method includes accessing, in a digital storage device, invoices associated with a buyer computer. Candidate invoices having similar attributes are identified from the invoices and nodes representing the candidate invoices are created and stored in computer memory. At least two of the candidate invoices are determined likely to be duplicate invoices and an edge is generated between nodes representing the at least two candidate invoices, indicating that the at least two candidate invoices are likely to be duplicate invoices. A set of filters is programmatically applied using the at least two candidate invoices as inputs. The at least two candidate invoices are determined to be duplicate invoices based on an output of the set of filters and a notification is sent to the buyer computer indicating that the at least two candidate invoices are duplicate invoices.