METHOD AND APPARATUS FOR DEFINING AN ARTIFICIAL BRAIN VIA A PLURALITY OF CONCEPT NODES CONNECTED TOGETHER THROUGH PREDETERMINED RELATIONSHIPS

• US 20080243741A1
• Filed: 06/10/2008
• Published: 10/02/2008
• Est. Priority Date: 01/06/2004
• Status: Abandoned Application

• Current Assignee: Neuric Technologies, LLC
• Original Assignee: Neuric Technologies, LLC
• Inventors: VISEL, THOMAS A., HAMILTON, GENE P., WOMACK, LUKAS K.
• Application Number:
  US12/136,670
• Publication Number:
  US 20080243741A1
• Time in Patent Office: Days
• US Class Current: 706/41
• CPC Class Codes: G06N 3/02 Neural networks

Abstract

A method for defining a network of nodes is provided, each representing a unique concept, and making connections between individual concepts through unique relationships to other concepts. Each of the nodes is operable to store a unique identifier in the network and information regarding the concept in addition to the unique relationships.

Claims

1. A method for emulating human behavior, the method comprising

providing a plurality of neurons, wherein one neuron represents one concept,

creating relational connection between at least one of the plurality of neurons and a second of the plurality of neurons;

wherein the meaning of the neuron is determined by a connection between the at least one of the plurality of neurons and the second of the plurality of neurons established through the relational connection.

2. The human emulation method of claim 1, wherein the relational connection uses an AND/OR predicate logic.

3. The human emulation method of claim 2, wherein the AND/OR predicate logic uses a sequence in a list to express a logic.

4. The human emulation method of claim 2, wherein the AND/OR predicate logic uses conventional logic gates.

5. The human emulation method of claim 1, wherein the relational connection further comprises the step of using a percentage weight to establish relative strength between the at least one of the plurality of neurons and the second of the plurality of neurons.

6. The human emulation method of claim 1, wherein the relational connection further comprises the step of using an encoded or enumerated connection type.

7. The human emulation method of claim 1, wherein the relational connection further comprises the step of using an encoded or enumerated connection type, and a percentage weight to establish relative strength.

8. The human emulation method of claim 6, wherein the relational connection further comprises the step of pointing to a different class of neurons, wherein a class implied by the specific code used.

9. The human emulation method of claim 6, wherein the step of using an encoded or enumerated connection type, further comprises a first set of types serving the role of left-brained factual connections, and wherein a second set of types serving for the factual connections.

10. The human emulation method of claim 6, wherein the step of using an encoded or enumerated connection type, further comprises a first set of types serving the role of left-brained factual connections, and wherein a second set of types serving for right-brained connections of experiential or emotional roles.

11. The human emulation method of claim 6, further comprising the step of separating the plurality neurons into general classifications, such that separate numbering spaces may be used for their serial or neuron-ID numbers, and wherein the numbering spaces are implied by the encoded type of the relational connection.

12. The human emulation method of claim 1, further comprising the step of creating a translating between temporary and permanent memory spaces by means of a simple translation table, and which table may contain ‘

hit’

entries to evaluate the relative merit of a permanent connection.

13. The human emulation method of claim 1, further comprising the step of creating a neuron relational connection via the conceptualization of sentence intent and content.

14. The human emulation method of claim 1, creating a neuron relational connection via interactive experience with objects and concepts in the external world around the brain.

15. A method for storing a collection of concepts in a nodal network, comprising the steps of:

defining a plurality of concepts, each associated with a node;

uniquely defining for a given node a unique representation of the associated concept in the collection of concepts;

each node including;

an associated identifier for identifying the given node;

a plurality of relationship indicators, each defining a unique relationship from the associated concept to one or more of the other concepts in the collection of concepts;

wherein the associated concept can bear a unique relationship to one or more of a plurality of the other of the concepts from that one or more of the plurality of the other concepts to the associated concept.

16. The method of claim 15, wherein select ones of the relationship indicators are hierarchical in nature.

17. The method of claim 15, wherein a plurality of the nodes are grouped in a semantic grouping defined by the relationship each of the nodes therein and their associated concepts have to a semantic flow of concepts.

18. The method of claim 17, wherein the semantic flow comprises a sentence.

19. The method of claim 17, wherein the semantic flow comprises one of the concept nodes having a relationship of action in the semantic flow, one of the concept nodes having a relationship of subject in the semantic flow and one of the concept nodes having a relationship of object in the semantic flow.