Spelling Correction with Denoising Transformer
This improves spelling correction in products like HubSpot search and enables applications in resource-scarce languages, though it is incremental over existing transformer-based methods.
The paper tackles spelling correction for short inputs like search queries by developing a method to generate artificial typos that mimic human error patterns, training a transformer model with this data. The approach outperforms standard noise-based methods and is extended to train models for Arabic, Greek, Russian, and Setswana without labeled data.
We present a novel method of performing spelling correction on short input strings, such as search queries or individual words. At its core lies a procedure for generating artificial typos which closely follow the error patterns manifested by humans. This procedure is used to train the production spelling correction model based on a transformer architecture. This model is currently served in the HubSpot product search. We show that our approach to typo generation is superior to the widespread practice of adding noise, which ignores human patterns. We also demonstrate how our approach may be extended to resource-scarce settings and train spelling correction models for Arabic, Greek, Russian, and Setswana languages, without using any labeled data.