Evaluate Similarity of Requirements with Multilingual Natural Language Processing

DS 116: Proceedings of the DESIGN2022 17th International Design Conference

Year: 2022
Editor: Mario Štorga, Stanko Škec, Tomislav Martinec, Dorian Marjanović
Author: Ursina Bisang (1), J
Series: DESIGN
Institution: 1: Fraunhofer IPK, Germany; 2: CONTACT Software GmbH, Germany
Section: Artificial Intelligence and Data-Driven Design
Page(s): 1511-1520
DOI number: https://doi.org/10.1017/pds.2022.153
ISSN: 2732-527X (Online)

Abstract

Finding redundant requirements or semantically similar ones in previous projects is a very time-consuming task in engineering design, especially with multilingual data. Due to modern NLP it is possible to automate such tasks. In this paper we compared different multilingual embeddings models to see which of them is the most suitable to find similar requirements in English and German. The comparison was done for both in-domain data (requirements pairs) and out-of-domain data (general sentence pairs). The most suitable model were sentence embeddings learnt with knowledge distillation.

Keywords: artificial intelligence (AI), requirements management, information management, data-driven design, natural language processing