Self-Comprehension for More Coherent Language Generation

George A. Wright, Matthew Purver

Abstract: Linguoplotter is a distributed and chaotic architecture where an entanglement of different processes interact to generate a text describing a raw data input. This paper describes recent additions to the architecture whereby a greater degree of language comprehension is used to improve the coherence of generated text. Some examples of the architecture operating are considered, including where it performs well and generates a good quality text; and instances where it gets trapped in loops that either prevent an output from being generated or cause a lower quality output to be produced before there is a chance to find a better alternative. Finally, ideas from the program Metacat are considered which could allow the program to observe its own processes and become a more human-like intelligence.