# Persuasive Natural Language Generation -- A Literature Review

**Authors:** Sebastian Duerr, Peter A. Gloor  
**arXiv:** [2101.05786](https://arxiv.org/abs/2101.05786)  
**Date:** 2021-01-14  
**Summary page:** https://www.summarizepaper.com/en/paper/2101.05786v1/

## Key points

- Literature review on Persuasive Natural Language Generation (NLG)
- Explores use of NLG to detect and generate persuasive texts
- Focuses on determinants of persuasion in five business-focused categories
- Benevolence
- Linguistic appropriacy
- Logical argumentation
- Trustworthiness
- Tools and datasets
- Objective is to enhance persuasiveness of existing messages using NLG techniques
- Review highlights key aspects within each category based on previous research findings
- Analyzes seventy-seven articles to outline existing knowledge and progress in the field
- Proposes a research agenda for further investigation of persuasive NLG
- Provides valuable insights into leveraging NLG for generating persuasive texts across domains
- Contributes to understanding NLG's potential for generating persuasive content

## Summary

This literature review, titled "Persuasive Natural Language Generation - A Literature Review," conducted by Sebastian Duerr and Peter A. Gloor, explores the use of Natural Language Generation (NLG) to automatically detect and generate persuasive texts. Building upon previous research on automatic identification of persuasion in text, the authors focus on generative aspects by conceptualizing determinants of persuasion in five business-focused categories: benevolence, linguistic appropriacy, logical argumentation, trustworthiness, tools and datasets. The objective is to enhance the persuasiveness of existing messages using NLG techniques. The review highlights key aspects within each of the aforementioned categories based on previous research findings. By analyzing seventy-seven articles, the authors outline the existing body of knowledge in this field and demonstrate its steady progress. The study also proposes a research agenda to further investigate persuasive NLG. This comprehensive review provides valuable insights into how NLG can be leveraged to automatically identify and generate persuasive texts across various domains. It serves as a foundation for future studies aiming to advance this research area. Overall, this literature review contributes to our understanding of NLG's potential for generating persuasive content and paves the way for further exploration and development in this field.