Controlled Language and Baby Turing Test for General Conversational Intelligence

TL;DR

This paper proposes extending the Turing Test with a Baby Turing Test approach and using controlled language with semantic graph models to evaluate and develop general conversational intelligence.

Contribution

It introduces a novel combination of Baby Turing Test and controlled language with semantic graphs for building versatile conversational systems.

Findings

Semantic graph models enable domain extensibility

Baby Turing Test offers a new evaluation method

Framework supports intelligent assistants for media and social data

Abstract

General conversational intelligence appears to be an important part of artificial general intelligence. Respectively, it requires accessible measures of the intelligence quality and controllable ways of its achievement, ideally - having the linguistic and semantic models represented in a reasonable way. Our work is suggesting to use Baby Turing Test approach to extend the classic Turing Test for conversational intelligence and controlled language based on semantic graph representation extensible for arbitrary subject domain. We describe how the two can be used together to build a general-purpose conversational system such as an intelligent assistant for online media and social network data processing.