Appzen, Inc.

Edris Naderan

Kunal Verma

Debashish Panigrahi

Thomas James White

Deepti Chafekar

Snigdha Purohit

Parivesh Priye

Arghyadeep Giri

Amrit Singhal

Hyram Du

Ankit Bansal

Akash Vijay Hajare

A non-transitory computer readable storage medium has instructions executed by a processor to access a purchase order data set. An invoice data set is ingested. Tuples of purchase order line-items from the purchase order data set are gathered. Tuples of invoice line-items from the invoice data set are formed. Positive training samples are created, where each positive training sample comprises a selected purchase order line-item and a corresponding invoice line-item. Negative training samples are formed, where each negative training sample comprises a fabricated purchase order line-item and a corresponding invoice line-item. A machine learning model is trained with the positive training samples and the negative training samples to form a trained purchase order machine learning model. An accounts payable invoice is received. The trained purchase order machine learning model is used to match an accounts payable invoice line-item to a purchase order line-item.

Apparatus and method for automatically matching purchase order line-items with accounts payable invoice line-items

A non-transitory computer readable storage medium has instructions executed by a processor to invoke an image processing module to ingest a digital invoice. An evaluation module derives metrics from the digital invoice. A semantic document processing module forms entity extracts from the digital invoice, where each entity extract from the digital invoice has a potential mapping to a trained machine learning model element. An entity extraction correction module overrides the potential mapping to the trained machine learning model element when user feedback from a similar entity extract from a previously processed digital invoice exists to produce a processed digital invoice with a user feedback element inconsistent with the potential mapping to the trained machine learning model element. The processed digital invoice is delivered to an accounting module for final disposition of the digital invoice.

Apparatus and method for processing digital invoices with invoice entity feedback override of trained machine learning models

A computer implemented method includes receiving a document with line item textual entries and an attachment containing images of different objects characterizing different transactions. The images of the different objects are split into individual image objects. Attributes from the individual image objects are extracted. The line item textual entries are matched with the individual image objects to form matched image objects. The matched image objects include ambiguous matches with multiple individual image objects assigned to a single line item textual entry or a single individual image object assigned to multiple line item textual entries. An assignment model is applied to resolve the ambiguous matches. The assignment model defines priority constraints, assigns pairs of line item textual entries and individual image objects that meet highest priority constraints, removes highest priority constraints when ambiguous matches remain, and repeats these operations until no ambiguous matches remain. One-to-one matches of line item textual entries and individual image objects are returned.

Apparatus and method for matching line item textual entries with individual image objects from a file with multiple image objects

Growing community of inventors

San Jose, CA, United States of America

Edris Naderan