Learning Alternative Name Spellings
This work addresses a common bottleneck in entity resolution systems for applications like genealogy or record linkage, offering a practical improvement over existing methods.
The paper tackled the problem of matching alternative spellings of names in entity resolution by framing it as a character-level machine translation task, using genealogy records and search logs to build models that substantially outperformed standard phonetic and string similarity methods in precision and recall.
Name matching is a key component of systems for entity resolution or record linkage. Alternative spellings of the same names are a com- mon occurrence in many applications. We use the largest collection of genealogy person records in the world together with user search query logs to build name matching models. The procedure for building a crowd-sourced training set is outlined together with the presentation of our method. We cast the problem of learning alternative spellings as a machine translation problem at the character level. We use in- formation retrieval evaluation methodology to show that this method substantially outperforms on our data a number of standard well known phonetic and string similarity methods in terms of precision and re- call. Additionally, we rigorously compare the performance of standard methods when compared with each other. Our result can lead to a significant practical impact in entity resolution applications.