CloudScan - A configuration-free invoice analysis system using recurrent neural networks

TL;DR

CloudScan is an invoice analysis system that uses recurrent neural networks to accurately extract information from invoices without requiring templates or manual annotations, generalizing well to unseen layouts.

Contribution

It introduces a template-free, globally trained neural network model for invoice data extraction that learns from automatically generated feedback, eliminating manual annotation.

Findings

Recurrent neural network outperforms baseline in unseen invoice layouts.

Achieves high F1 scores on a large dataset of 326,471 invoices.

Generalizes well to new invoice formats without prior templates.

Abstract

We present CloudScan; an invoice analysis system that requires zero configuration or upfront annotation. In contrast to previous work, CloudScan does not rely on templates of invoice layout, instead it learns a single global model of invoices that naturally generalizes to unseen invoice layouts. The model is trained using data automatically extracted from end-user provided feedback. This automatic training data extraction removes the requirement for users to annotate the data precisely. We describe a recurrent neural network model that can capture long range context and compare it to a baseline logistic regression model corresponding to the current CloudScan production system. We train and evaluate the system on 8 important fields using a dataset of 326,471 invoices. The recurrent neural network and baseline model achieve 0.891 and 0.887 average F1 scores respectively on seen invoice layouts. For the harder task of unseen invoice layouts, the recurrent neural network model outperforms the baseline with 0.840 average F1 compared to 0.788.