Neural Graph Matching for Modification Similarity Applied to Electronic Document Comparison

TL;DR

This paper introduces a neural graph matching method incorporating textual features and layout analysis to accurately compare electronic documents, especially for legal and scientific texts, addressing layout inconsistencies and modifications.

Contribution

It presents a novel neural graph matching approach that combines layout analysis, textual features, and Sinkhorn divergence to improve document comparison accuracy in complex cases.

Findings

Effective in legal and scientific document comparison

Handles layout inconsistencies with many-to-many block matching

Outperforms traditional methods in accuracy

Abstract

In this paper, we present a novel neural graph matching approach applied to document comparison. Document comparison is a common task in the legal and financial industries. In some cases, the most important differences may be the addition or omission of words, sentences, clauses, or paragraphs. However, it is a challenging task without recording or tracing whole edited process. Under many temporal uncertainties, we explore the potentiality of our approach to proximate the accurate comparison to make sure which element blocks have a relation of edition with others. In beginning, we apply a document layout analysis that combining traditional and modern technics to segment layout in blocks of various types appropriately. Then we transform this issue to a problem of layout graph matching with textual awareness. About graph matching, it is a long-studied problem with a broad range of applications. However, different from previous works focusing on visual images or structural layout, we also bring textual features into our model for adapting this domain. Specifically, based on the electronic document, we introduce an encoder to deal with the visual presentation decoding from PDF. Additionally, because the modifications can cause the inconsistency of document layout analysis between modified documents and the blocks can be merged and split, Sinkhorn divergence is adopted in our graph neural approach, which tries to overcome both these issues with many-to-many block matching. We demonstrate this on two categories of layouts, as follows., legal agreement and scientific articles, collected from our real-case datasets.