Evaluate Similarity of Requirements with Multilingual Natural Language Processing

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 model...

Full description

Saved in:
Bibliographic Details
Published inProceedings of the Design Society Vol. 2; pp. 1511 - 1520
Main Authors Bisang, U., Brünnhäußer, J., Lünnemann, P., Kirsch, L., Lindow, K.
Format Journal Article Conference Proceeding
LanguageEnglish
Published Cambridge Cambridge University Press 01.05.2022
Subjects
Algorithms
Artificial intelligence
Automation
Design
Design engineering
Language
Learning
Multilingualism
Natural language processing
Neural networks
Product development
Requirements analysis
Online AccessGet full text

Cover

More Information
Summary: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.
ISSN:2732-527X
2732-527X
DOI:10.1017/pds.2022.153