Improving Address Matching using Siamese Transformer Networks

TL;DR

This paper presents a deep learning model using Siamese Transformer Networks to improve address matching accuracy and speed for Portuguese addresses, significantly reducing errors and increasing efficiency in delivery processes.

Contribution

The paper introduces a novel bi-encoder and cross-encoder architecture specifically designed for Portuguese address matching, achieving high accuracy and faster inference compared to traditional methods.

Findings

Address matching accuracy exceeds 95% at the door level

Inference speed is 4.5 times faster than BM25 with GPU

Model effectively retrieves and reranks address candidates

Abstract

Matching addresses is a critical task for companies and post offices involved in the processing and delivery of packages. The ramifications of incorrectly delivering a package to the wrong recipient are numerous, ranging from harm to the company's reputation to economic and environmental costs. This research introduces a deep learning-based model designed to increase the efficiency of address matching for Portuguese addresses. The model comprises two parts: (i) a bi-encoder, which is fine-tuned to create meaningful embeddings of Portuguese postal addresses, utilized to retrieve the top 10 likely matches of the un-normalized target address from a normalized database, and (ii) a cross-encoder, which is fine-tuned to accurately rerank the 10 addresses obtained by the bi-encoder. The model has been tested on a real-case scenario of Portuguese addresses and exhibits a high degree of accuracy, exceeding 95% at the door level. When utilized with GPU computations, the inference speed is about 4.5 times quicker than other traditional approaches such as BM25. An implementation of this system in a real-world scenario would substantially increase the effectiveness of the distribution process. Such an implementation is currently under investigation.