Developing a Natural Language Understanding System for Dealing with the Sequencing Problem in Simulating Brain Damage

Abstract

This paper is an attempt to show how a Human-Robot Interface (HRI) system in the Greek language can help people with brain damage in speech and its related perception issues. This proposal is not the product of research conducted on how to treat brain injuries. It is a conclusion stemming from research on intelligent Human-Robot interfaces, as a part of Artificial Intelligence and Natural Language Processing, which approaches the processing and understanding of natural language with specific methods. For the same reason, experiments on real patients have not been conducted. Thus, this paper does not propose a competing method, but a method for further study. Since it is referring to a very general and quite complex issue, an approach is presented here for the Sequencing problem. A person with such a problem cannot hierarchically organize the tasks needed to be performed. This Hierarchy has to do with both time and practicality. The particular problem here, as much as the innovation of our approach, lies not when there are explicit temporally defined instructions, but in the ability to derive these temporal values through the person’s perception from more vague temporal references. The present approach is developed based on our related previous works for deploying a robotic system that relies on Hole Semantics and the OMAS-III computational model as a grammatical formalism for its communication with humans.