Entity Matching using Language Models

Linking records that belong to the same entity, also known as entity matching is an important part of data integration and management. Many existing solutions for entity matching requires the user to have a deep understanding of the data schema and available techniques in order to match correctly. But this is not the case in real-world use cases where data has lots of variety, "dirty" and the schema of the databases are not the same. This thesis presents a framework for entity matching using fine-tuned language models, where it does not need an aligned schema to perform matching.