Translational NLP: A New Paradigm and General Principles for Natural Language Processing Research

Area of Science:

• Computer Science
• Artificial Intelligence
• Computational Linguistics

Background:

• Natural Language Processing (NLP) research traditionally separates basic science from applied science.
• The exchange between basic and applied NLP is often assumed, leading to underutilized innovations and unaddressed research questions.

Purpose of the Study:

• To introduce a new research paradigm: Translational NLP.
• To structure and facilitate the bidirectional exchange between basic and applied NLP research.
• To focus on how application needs can drive innovation in basic NLP science and technology.

Main Methods:

• Describing a new paradigm of Translational NLP.
• Presenting a conceptual framework for translational research in NLP.
• Outlining stakeholders and key questions within translational NLP.

Main Results:

• Many significant NLP advances arise from the intersection of basic principles and application needs.
• A conceptual framework is provided to guide Translational NLP development.
• General translational principles are identified to enhance research exchange.

Conclusions:

• Translational NLP offers a structured approach to connect fundamental research with practical applications.
• This paradigm shift can accelerate innovation and ensure the relevance of NLP research.
• The proposed framework serves as a roadmap for establishing Translational NLP as a distinct research area.