Daniel Parmenter

Background

About Daniel Parmenter

I am not a software engineer. If you are looking for someone to build an NLP or machine learning system from the ground up, I am probably not the person you need. But if you're looking for a committed and dedicated linguistic analyst who is adaptable to a wide variety of technologies and approaches, we should talk. I have many years of experience with NLP, information extraction and text analytics technology. I have a real talent for identifying and finding linguistic patterns in unstructured data. I have written dictionaries and grammars for an old-school chart parser system as well as grammars for finite state transducer cascade systems; I am also very adept at Boolean query-writing for social listening platforms such as Brandwatch, Crimson Hexagon, Linkfluence Radarly, Pulsar. My recent work has involved building machine learning models for linguistic data, including defining relevancy criteria, building datasets, coding, and testing these models for precision/recall and other metrics. I am not a professional programmer, but I have some programming background and often write small, ad hoc programs (usually in Emacs Lisp). I also have a lot of experience working with programmers/developers and one of my genuine, if obscure, talents is for facilitating discussions between programmers who don't have a background in linguistics and linguists who don't have a background in programming.