[Paper Review] Natural Language Processing in-and-for Design Research

L Siddharth, Luciënne Blessing+1 more|arXiv (Cornell University)|Nov 27, 2021

Design Education and Practice4 citations

TL;DR

This paper reviews 223 NLP studies in design research from 1991 to present, categorizing natural language sources such as design reports, concepts, and consumer opinions. It maps NLP applications to a design innovation framework, identifies research gaps, and proposes methodological and theoretical directions for future work in NLP-in-design research.

ABSTRACT

We review the scholarly contributions that utilise Natural Language Processing (NLP) techniques to support the design process. Using a heuristic approach, we gathered 223 articles that are published in 32 journals within the period 1991-present. We present state-of-the-art NLP in-and-for design research by reviewing these articles according to the type of natural language text sources: internal reports, design concepts, discourse transcripts, technical publications, consumer opinions, and others. Upon summarizing and identifying the gaps in these contributions, we utilise an existing design innovation framework to identify the applications that are currently being supported by NLP. We then propose a few methodological and theoretical directions for future NLP in-and-for design research.

Motivation & Objective

• To examine the current state of Natural Language Processing (NLP) applications in design research across diverse text sources.
• To identify methodological and theoretical gaps in existing NLP-in-design research through a systematic review of 223 scholarly articles.
• To map NLP applications to a design innovation framework to clarify current capabilities and limitations.
• To propose actionable research directions for advancing NLP-in-design research in both methodology and theory.
• To support the integration of NLP techniques into design processes by highlighting underexplored text sources and application areas.

Proposed method

• Conducted a heuristic review of 223 articles published in 32 journals between 1991 and 2022.
• Categorized the natural language text sources used in the studies into six types: internal reports, design concepts, discourse transcripts, technical publications, consumer opinions, and others.
• Mapped NLP techniques and applications to a pre-existing design innovation framework to assess their alignment with design process stages.
• Identified recurring methodological patterns and theoretical assumptions in NLP applications within design research.
• Synthesized findings to highlight underutilized text sources and underdeveloped NLP applications in design contexts.
• Proposed future research directions based on identified gaps in both methodological rigor and theoretical grounding.

Experimental results

Research questions

• RQ1 Which types of natural language text sources are most commonly used in NLP for design research?
• RQ2 How are NLP techniques currently applied across different stages of the design innovation process?
• RQ3 What are the key methodological and theoretical limitations in existing NLP-in-design research?
• RQ4 Which text sources and design process stages remain underexplored in current NLP applications?
• RQ5 What future methodological and theoretical directions can strengthen the integration of NLP into design research?

Key findings

• The majority of NLP applications in design research focus on technical publications and consumer opinions, with limited use of internal design reports and discourse transcripts.
• NLP techniques are predominantly used for sentiment analysis, topic modeling, and keyword extraction, with few studies employing advanced NLP methods like transformer-based models.
• Significant gaps exist in the use of NLP for analyzing design concepts and internal design team discussions, which are critical for understanding ideation processes.
• Few studies integrate NLP with formal design theory or process models, limiting theoretical grounding and reproducibility.
• There is a lack of standardized evaluation metrics and benchmark datasets for NLP in design contexts, hindering methodological comparability.
• The study identifies a strong need for more interdisciplinary research that bridges NLP methodologies with design science frameworks to enhance both practical and theoretical impact.