A neural mechanism for human language processing

Authors:
Alexander Borzenko
Affiliations:
Proto-Mind Machines Inc., Concord, Canada
Venue:
Neurocomputing
Year:
2010

Citing 5
Cited 0

Memory-Based Language Processing (Studies in Natural Language Processing)

Memory-Based Language Processing (Studies in Natural Language Processing)
Associative memory with a controlled chaotic neural network

Neurocomputing
2009 Special Issue: Neural associative memories for the integration of language, vision and action in an autonomous agent

Neural Networks
2009 Special Issue: Discrete combinatorial circuits emerging in neural networks: A mechanism for rules of grammar in the human brain?

Neural Networks
Language Processing in Human Brain

Proceedings of the 2008 conference on Artificial General Intelligence 2008: Proceedings of the First AGI Conference

Quantified Score

Hi-index	0.01

Visualization

Abstract

We describe a computational model implementing a novel theory regarding how the human brain may process English and other human languages. This theory treats the human language as a multi-modal symbolic system independent of the symbols' origin. The main objective of our research is to understand how tiny single neurons working together can produce intellectual behavior that exhibits itself in proper speech comprehension and generation in accordance with the current context. Based on our theory in this regard we describe the virtual neural computer, which we call GSL. It encapsulates our key theoretical results and allows us to run linguistic experiments. We have built and tested a (software) intelligent assistant aimed at military applications, based on the GSL algorithms. This military assistant speaks English in real time and supplies its user with up-to-date information.