Fuzzy Substring Matching: On-device Fuzzy Friend Search at Snapchat

TL;DR

This paper presents an efficient on-device fuzzy friend search method for Snapchat, enabling quick, accurate, and resource-efficient matching of user queries with friends' names despite typos.

Contribution

It introduces a novel two-step fuzzy search approach combining skip-bigram retrieval and local Levenshtein distance for resource-constrained devices.

Findings

Achieves high accuracy in fuzzy friend matching.

Reduces latency by performing search locally.

Demonstrates effectiveness on Snapchat's user data.

Abstract

About 50% of all queries on Snapchat app are targeted at finding the right friend to interact with. Since everyone has a unique list of friends and that list is not very large (maximum a few thousand), it makes sense to perform this search locally, on users' devices. In addition, the friend list is already available for other purposes, such as showing the chat feed, and the latency savings can be significant by avoiding a server round-trip call. Historically, we resorted to substring matching, ranking prefix matches at the top of the result list. Introducing the ability to perform fuzzy search on a resource-constrained device and in the environment where typo's are prevalent is both prudent and challenging. In this paper, we describe our efficient and accurate two-step approach to fuzzy search, characterized by a skip-bigram retrieval layer and a novel local Levenshtein distance computation used for final ranking.