Stopwords in Technical Language Processing

TL;DR

This paper identifies and curates a specialized stopword list for engineering texts, addressing the lack of standard stopword resources tailored for technical language processing tasks.

Contribution

It introduces a data-driven methodology to identify uninformative stopwords specific to engineering texts and provides a curated stopword list for technical NLP applications.

Findings

A new stopword list for engineering texts is created.

The list improves NLP tasks in technical domains.

Methodology can be adapted for other technical fields.

Abstract

There are increasingly applications of natural language processing techniques for information retrieval, indexing and topic modelling in the engineering contexts. A standard component of such tasks is the removal of stopwords, which are uninformative components of the data. While researchers use readily available stopword lists which are derived for general English language, the technical jargon of engineering fields contains their own highly frequent and uninformative words and there exists no standard stopword list for technical language processing applications. Here we address this gap by rigorously identifying generic, insignificant, uninformative stopwords in engineering texts beyond the stopwords in general texts, based on the synthesis of alternative data-driven approaches, and curating a stopword list ready for technical language processing applications.