Semi-Structured Query Grounding for Document-Oriented Databases with Deep Retrieval and Its Application to Receipt and POI Matching

TL;DR

This paper presents a deep learning-based retrieval method for matching semi-structured queries with document-oriented databases, demonstrated on receipt and POI matching, outperforming traditional manual pattern-based approaches.

Contribution

It introduces a novel embedding-based retrieval model that eliminates manual engineering, improving accuracy and reducing development effort in real-world semi-structured data applications.

Findings

Model significantly outperforms manual pattern-based methods

Deep retrieval approach handles noisy and incomplete data effectively

Provides insights for practitioners in similar semi-structured data domains

Abstract

Semi-structured query systems for document-oriented databases have many real applications. One particular application that we are interested in is matching each financial receipt image with its corresponding place of interest (POI, e.g., restaurant) in the nationwide database. The problem is especially challenging in the real production environment where many similar or incomplete entries exist in the database and queries are noisy (e.g., errors in optical character recognition). In this work, we aim to address practical challenges when using embedding-based retrieval for the query grounding problem in semi-structured data. Leveraging recent advancements in deep language encoding for retrieval, we conduct extensive experiments to find the most effective combination of modules for the embedding and retrieval of both query and database entries without any manually engineered component. The proposed model significantly outperforms the conventional manual pattern-based model while requiring much less development and maintenance cost. We also discuss some core observations in our experiments, which could be helpful for practitioners working on a similar problem in other domains.