Situation dependent operation of a semantic network machine

• US 7,523,079 B2
• Filed: 04/28/2006
• Issued: 04/21/2009
• Est. Priority Date: 04/19/1999
• Status: Active Grant

• Current Assignee: Definiens AG (Astrazeneca PLC)
• Original Assignee: Definiens AG (Astrazeneca PLC)
• Inventors: Athelogou, Maria, Bobolas, Konstantinos, Binnig, Renate, Eschenbacher, Peter, Schmidt, Guenter
• Primary Examiner(s): Vincent; David R
• Assistant Examiner(s): Tran; Mai T
• Application Number:
  US11/414,000
• Publication Number:
  US 20060195407A1
• Time in Patent Office: 1,089 Days
• Field of Search: 706/12, 707/101
• US Class Current: 706/12
• CPC Class Codes: G06N 5/02 Knowledge representation; Symbolic representation
• CPC Class Codes: Y10S 707/99942 Manipulating data structure, e.g. compression, compaction, compilation

Abstract

A computer-implemented network structure includes a semantic network machine comprised of nodes containing informational contents and links containing relational contents. Relational contents describe the relationship between linked nodes. Some of the nodes are semantic Janus units. Based on time-variable states of each semantic Janus unit, the semantic Janus units perform operations on nodes and links. The operations are focused on selected portions of the semantic network machine such that each semantic Janus unit need not deal with all possible informational and relational contents within the semantic network machine. The computational resources of the computer network are thereby efficiently managed, and artificial intelligence tasks such as inferential retrieval are performed quicker. The amount of data that is processed is substantially reduced by focusing on bundled information. The semantic network is used for pattern recognition, for example, to recognize blood vessels in a medical image or streets on a digital satellite image.

Claims

1. A computer-implemented network structure, comprising:

a memory that stores instructions to create;

a first node that is a semantic Janus unit, wherein the first node possesses an existing time-variable state;

a second node containing informational contents; and

a link between the first node and the second node, wherein the link contains relational contents that describes the relationship between the first node and the second node, wherein the first node carries out operations on the second node and on the link, wherein the existing time-variable state determines which operations are carried out, and wherein a pattern in an image is recognized by carrying out the operations.

2. The computer-implemented network structure of claim 1, further comprising:

a library of elementary links, wherein the first node, the second node and the link are part of a semantic network machine within the computer-implemented network structure, wherein the computer-implemented network structure has a user, and wherein the user selects the link from the library of elementary links and adds the link to the semantic network machine.

3. The computer-implemented network structure of claim 2, further comprising:

a semantic view unit, wherein the semantic view unit generates a graphical representation of the recognized pattern and displays the graphical representation to the user.

4. The computer-implemented network structure of claim 1, wherein the operations change the informational contents of the second node.

5. The computer-implemented network structure of claim 1, further comprising:

a third node containing second informational contents, wherein the time-variable state determines whether the operations are carried out on the second node or on the third node.

6. The computer-implemented network structure of claim 5, wherein one of the operations carried out by the first node is to delete the third node.

7. The computer-implemented network structure of claim 1, wherein the first node, the second node and the link are part of a semantic network machine within the computer-implemented network structure, wherein the time-variable state of the first node reflects the situation in the semantic network machine, and wherein the operations are focused on parts of the semantic network machine based on the situation in the semantic network machine.

8. The computer-implemented network structure of claim 1, wherein The first node monitors nodes within a vicinity to be monitored, wherein the first node performs the operations on nodes within a vicinity to be shaped, and wherein the first node determines a new time-variable state based on the existing time-variable state and on the nodes within the vicinity to be monitored.

9. The computer-implemented network structure of claim 8, wherein a plurality of nodes is linked to The first node, and wherein the vicinity to be monitored encompasses a subset of the plurality of nodes.

10. The computer-implemented network structure of claim 1, wherein one of the operations carried out by the first node is to create a third node and to link the third node to the second node.

11. The computer-implemented network structure of claim 1, wherein the second node represents a graphic object, and wherein the informational contents contained in the second node are digital pixels of a portion of the image.

12. The computer-implemented network structure of claim 11, wherein the first node, the second node and the link are part of a semantic network machine within the computer-implemented network structure, wherein the graphic object has a shape and a color, and wherein the semantic network machine changes the shape and the color based on the existing time-variable state of the first node.

13. The computer-implemented network structure of claim 1, wherein the computer-implemented network structure has a user, and wherein the user interactively changes the time-variable state of the first node.

14. The computer-implemented network structure of claim 1, wherein the recognized pattern is taken from the group consisting of:

a road, a river, a railroad track, an electricity line, a blood vessel, a ligament, and an urban area.

15. A method comprising:

linking a first node that is a semantic Janus unit of a computer-implemented network structure to a second node of the computer-implemented network structure using a first link, wherein the first node, the second node and the first link are parts of a semantic network machine, wherein the first node possesses a state that varies with time, and wherein the second node contains informational contents;

generating changed informational contents by carrying out computational operations on the informational contents based on the state of the first node;

storing the changed informational contents in the second node;

adding a third node to the semantic network machine;

changing the first link to link the first node to the third node, wherein the changed informational contents and the changing the first link to link the first node to the third node are used to recognize a pattern; and

generating an image that includes the recognized pattern.

16. The method of claim 15, further comprising:

deleting the second node.

17. The method of claim 15, wherein the image depicts a traffic network, and wherein the informational contents are digital pixels of a portion of the image, and wherein the recognized pattern is a portion of the traffic network.

18. The method of claim 17, wherein the semantic network machine generates a transportation infrastructure map using the recognized pattern of the traffic network.

19. The method of claim 15, further comprising:

searching for unusual changes in the informational contents.

20. A computer memory comprising program instructions for performing the steps of:

linking a first node that is a semantic Janus unit of a computer-implemented network structure to a second node of the computer-implemented network structure using a first link, wherein the first node, the second node and the first link are parts of a semantic network machine, wherein the first node possesses a state that varies with time, and wherein data is stored in the second node;

generating changed data by carrying out computational operations on the data based on the state of the first node;

storing the changed data in the second node;

adding a third node to the semantic network machine;

changing the first link to link the first node to the third node, wherein the changed data and the changing the first link to link the first node to the third node are used to recognize a pattern; and

generating an image that includes the recognized pattern.

21. The computer memory of claim 20, wherein the recognized pattern is taken from the group consisting of:

a road, a river, a railroad track, an electricity line, a blood vessel, a ligament, and an urban area.

22. The computer memory of claim 20, further comprising program instructions for performing the step of:

changing an attribute of the image that includes the recognized pattern, wherein the attribute is taken from the group consisting of;

a shape, a texture, a color and a contrast.

23. A semantic network machine, comprising:

a plurality of nodes in a computer-implemented network structure;

a link that links a first one that is a semantic Janus unit of the plurality of nodes to a second one of the plurality of nodes, wherein the second one of the plurality of nodes contains informational contents, and wherein the informational contents includes pixels of a digital image; and

means for recognizing a pattern in the digital image by performing an operation only on a subset of the plurality of nodes based on a state of the semantic network machine.

24. The computer-implemented network structure of claim 23, wherein the means possesses a state, and wherein the state of the semantic network machine is dependent on the state of the means.

25. The computer-implemented network structure of claim 24, wherein the means monitors nodes within a vicinity of the means, and wherein the state of The means is dependent on the nodes within the vicinity.

26. The computer-implemented network structure of claim 23, wherein the operation is performed on the informational contents.

27. The computer-implemented network structure of claim 23, wherein each of the plurality of nodes contains informational contents, and wherein the operation is changing the informational contents of each of the subset of the plurality of nodes.