Enhancing Simulation Analysis Reporting Using Natural Language Feedback

这个项目是对模拟和自然语言的关系和联合使用的探索。所使用的概念结构可以表示句子的结构和意义以及基本的数学概念。句子可以转化为这些结构，通过语言生成，这些结构可以转化为数学方程。用一个纳米浮游生物呼吸模型的文字描述来说明一个例子。这位研究者以前的研究主要集中在将文本转化为适合于数值模拟的方程。目前的研究涉及相反方向的映射，生成自然语言文本，而不是从模拟运行中生成传统的数字输出。通过这种方式，分析师将能够以文本形式描述物理系统，然后获得描述系统行为的文本响应。

This project is an exploration of the relationships and the joint use of simulation and natural language. Conceptual structures are used which can represent the structure and meaning of sentences as well as basic mathematical concepts. Sentences can be transformed to these structures, and, via language generation, the structures can be transformed into mathematical equations. An example is illustrated using a text description of a nanoplankton respiration model. Previous research by this investigator has concentrated on the translation of text to equations suitable for numerical simulation. The current research concerns mapping in the opposite direction, generating natural language text instead of the traditional numerical output from a simulation run. In this manner, the analyst will be able to textually describe a physical system and then obtain textual responses that describe the system behavior.