Affiliation:
1. Stanford Univ., Stanford, CA
Abstract
The paper describes a working analysis and generation program for natural language, which handles paragraph length input. Its core is a system of preferential choice between deep semantic patterns, based on what we call “semantic density.” The system is contrasted:
with syntax oriented linguistic approaches, and
with theorem proving approaches to the understanding problem.
Publisher
Association for Computing Machinery (ACM)
Cited by
97 articles.
订阅此论文施引文献
订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献