Method of organizing data into a graphically oriented format

• US 5,819,245 A
• Filed: 09/05/1995
• Issued: 10/06/1998
• Est. Priority Date: 09/05/1995
• Status: Expired due to Fees

• Current Assignee: Motorola, Inc. (Motorola Solutions, Inc.)
• Original Assignee: Motorola, Inc. (Motorola Solutions, Inc.)
• Inventors: Leivian, Robert H., Peterson, William M.
• Primary Examiner(s): MacDonald, Allen R.
• Assistant Examiner(s): SHAH, SANJIV
• Application Number:
  US08/523,326
• Time in Patent Office: 1,127 Days
• Field of Search: 395/22, 395/24, 395/23, 395/26, 395/604, 395/2.41, 395/600, 395/161, 365/49, 364/513, 364/550, 382/159, 706/16, 706/20
• US Class Current: 706/16
• CPC Class Codes: G06N 3/04 Architecture, e.g. interconnection topology

Abstract

A neural network (10) organizes the data items into a graphically oriented format by retrieving data items from a database (68) where each data item has a plurality of attributes. The neural network is organized (102) such that data items having similar attributes are assigned to neurons located closer together. The neurons of the neural network are matched (104) with the data items from the database and stored in a cross reference table. The cross reference table is displayed (106) on a computer screen (108) in a graphical format so that user visually relates the food items and sees the similarities and differences in their attribute data by the proximity of the data items to one another. The graphic format allows easy visual interpretation of the data items. For large databases, multiple neural networks (110, 112) can be organized hierarchically.

Claims

1. A computer implemented method of organizing data into a graphically oriented format, comprising the steps of:

retrieving data items from a database where said data items each have a plurality of attributes;

training synaptic elements of a neural network with said data items so that said data items having similar attributes are assigned to a neighborhood of neurons in said neural network;

identifying a closest match between said data items from said database and said neurons of said neural network by comparing said plurality of attributes of said data items with trained values of said synaptic elements of said neurons of said neural network;

storing said closest match in a first cross reference table; and

displaying said first cross reference table in a graphical format to present similarities between said data items by positioning similar data items in proximity to one another in a display area.

2. The method of claim 1 wherein said step of training includes the steps of:

applying elements of an input signal vector representing said plurality of attributes of one of said data items to said neurons of said neural network;

identifying one of said neurons as having synapses that are most closely matched to said elements of said input signal vector; and

adjusting said synapses of said one of said neurons to more closely match said elements of said input signal vector according to a predetermined learning rule.

3. The method of claim 2 wherein said step of matching includes the steps of:

comparing said synapses of one of said neurons to said plurality of attributes of said data items; and

identifying one of said data items that most closely matches said one of said neurons.

4. The method of claim 2 wherein said step of matching includes the steps of:

applying said input signal vector representing said plurality of attributes of said one of said data items to said neurons of said neural network; and

identifying one of said neurons as having synapses that are most closely matched to said elements of said input signal vector.

5. The method of claim 1 wherein said step of displaying includes the steps of:

forming graphical icons representative of said data items; and

selecting one of said graphical icons to display attributes of said data items.

6. The method of claim 5 wherein said step of forming graphical icons includes the steps of forming said graphical icons by illustrating natural physical embodiments of said data items.

7. The method of claim 5 wherein said step of forming graphical icons includes the steps of forming said graphical icons by illustrating abstract embodiments of said data items with varying sizes and shapes for each of said data items.

8. The method of claim 1 further including the steps of:

retrieving data items from a database at a second point in time;

training said neural network with said plurality of attributes of said data items from said second point in time;

matching said neurons of said neural network with said data items from said database and storing matches in a second cross reference table; and

displaying said first and second cross reference tables in chronological sequence in a graphical format to present similarities and differences between said data items over time.

9. A system for organizing data into a graphically oriented format, comprising:

means for training memory elements of an organization map with attributes of data items so that said data items having similar attributes are assigned to a neighborhood of processing elements in said organization map;

means for identifying a closest match between said data items and said processing elements of said organization map by comparing said attributes of said data items with trained values of said memory elements of said processing elements of said projection map;

means for storing said closest match in a first cross reference table; and

means for displaying said first cross reference table in a graphical format to present similarities between said data items by positioning similar data items in proximity to one another in a display area.

10. The system of claim 9 wherein said means for training includes:

means for applying elements of an input signal vector representing attributes of one of said data items to said processing elements of said organization map;

means for identifying one of said processing elements as having weights that are most closely matched to said elements of said input signal vector; and

means for adjusting said weights of said one of said processing elements to more closely match said elements of said input signal vector.

11. A computer implemented method of organizing data into a graphically oriented format, comprising the steps of:

training processing elements of a projection map with attributes of data items so that said data items having similar attributes are assigned to a neighborhood of said processing elements in said projection map;

identifying a closest match between said data items and said processing elements of said projection map by comparing said attributes of said data items with trained values of said processing elements of said projection map;

storing said closest match in a first cross reference table; and

displaying said first cross reference table in a graphical format to present similarities between said data items by positioning similar date items in proximity to one another in a display area.

12. The method of claim 11 wherein said step of organizing includes the steps of:

applying elements of an input signal vector representing attributes of one of said data items to said processing elements of said projection map;

identifying one of said processing elements as having weights that are most closely matched to said elements of said input signal vector; and

adjusting said weights of said one of said processing elements to more closely match said elements of said input signal vector.

13. The method of claim 12 wherein said step of matching includes the steps of:

comparing said weights of one of said processing elements to each of said attributes of said data items; and

identifying one of said data items that most closely matches said one of said processing elements.

14. The method of claim 12 wherein said step of matching includes the steps of:

applying said input signal vector representing said attributes of said one of said data items to said processing elements of said projection map; and

identifying one of said processing elements as having weights that are most closely matched to said elements of said input signal vector.

15. The method of claim 11 wherein said step of displaying includes the steps of:

forming graphical icons representative of said data items; and

selecting said graphical icons to display attributes of said data items.

16. The method of claim 15 wherein said step of forming graphical icons includes the steps of forming said graphical icons by illustrating natural physical embodiments of said data items.

17. The method of claim 15 wherein said step of forming graphical icons includes the steps of forming said graphical icons by illustrating abstract embodiments of said data items with varying sizes, shapes or coloring for each of said data items.

18. The method of claim 11 further including the steps of:

providing data items at a second point in time;

organizing said projection map with said attributes of said data items from said second point in time;

matching said processing elements of said projection map with said data items and storing matches in a second cross reference table; and

displaying said first and second cross reference tables in chronological sequence in a graphical format to present similarities between said data items over time.

19. The method of claim 11 further including the steps of:

selecting primary items from a first level module;

passing attributes of said primary items to a second level module to select secondary items in said second level module; and

passing attributes of said secondary items in said second level module back to said first level module.

20. The method of claim 19 further including the steps of:

aggregating said attributes of said secondary items for one primary item with attributes of said secondary items with a second primary item; and

providing said aggregation with said primary items.

21. The method of claim 11 wherein said step of displaying includes the steps of:

forming graphical icons representative of said data items; and

displaying said graphical icons to present the similarities between attributes of said data items.

22. The method of claim 21 wherein said graphical icons are arranged in a grid array in the display area.

23. The method of claim 11 wherein said step of displaying includes the steps of:

forming graphical icons representative of said data items; and

displaying said graphical icons of similar data items in proximity to each other and said graphical icons of dissimilar data items distant from each other.

24. A method of organizing data into a graphically oriented format, comprising the steps of:

training synaptic elements of a neural network with attributes of data items from a database so that said data items having similar attributes are assigned to a neighborhood of neurons in said neural network;

identifying a closest match between said data items and neurons of said neural network by comparing said attributes of said data items with trained values of said synaptic elements of said neurons of said neural network; and

displaying said closest match in a graphical format to present similarities between said data items by positioning similar data items in proximity to one another in a display area.

25. The method of claim 24 wherein said step of displaying includes the steps of:

forming graphical icons representative of said data items; and

displaying said graphical icons of similar data items in proximity to each other and said graphical icons of dissimilar data items distant from each other.

26. The method of claim 25 wherein said graphical icons are arranged in a grid array in the display area.