ITRI- UniversityofBrighton

Title: ITRI- UniversityofBrighton

Authors:

Main Research Question: How can scripted dialogue control global properties to generate more natural and effective communication?

Methodology: The study explores the use of revision strategies in scripted dialogue generation. It discusses how these strategies can be applied at the level of a dialogue plan and how the generator can decide when to apply a given revision operator.

Results: The research suggests that by applying revision operators at the level of a dialogue plan, the generator can control global properties of the text generation, leading to more natural and effective communication.

Implications: The findings of this study have significant implications for Natural Language Generation (NLG) systems. By understanding how to apply revision strategies at the level of a dialogue plan, these systems can generate more natural and effective dialogue, improving overall communication. This can lead to improved user experiences in various applications, such as customer service, healthcare, and education.

Link to Article: https://arxiv.org/abs/0312052v1 Authors: arXiv ID: 0312052v1