Cory Massaro is a literature and language expert with experience in big data, software engineering, and natural language processing. He is the developer of a module that automatically segments transcriptions of two-person dialogues and identifies the broad topic of each segment, and of a pipeline that processes doctor’s notes and identifies the relationships among persons, ailments, and medications therein. The module was based on his MA in Computational Linguistics, during which he explored automated approaches to questions of narrative texture.

Massaro worked as a software engineer at Google, where he applied computational linguistic techniques over massive data sets. He improved the quality of speech recognition and keyboard language models in hundreds of languages. Using models derived from big data sets he performed uniquely accurate morphological analysis on languages with high degrees of compounding and agglutination in an effort to improve speech recognition quality. He also worked on transliteration models to improve data coverage for languages with multiple orthographies.

Massaro has already partnered with Dr Christophe Fricker to apply insights from theoretical and computational linguistics to questions of political discourse, with a focus on the way notions of citizenship are discussed in UK and German media. Results are available open-access in a co-authored research article and a blog contribution for the Royal Society of Arts.