Data mining shape based data

• US 9,244,946 B2
• Filed: 11/26/2012
• Issued: 01/26/2016
• Est. Priority Date: 11/26/2012
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Kassab, Maroun, Pastel, Leah M., Trojanowski, Adam E.
• Primary Examiner(s): LE, THU NGUYET T
• Application Number:
  US13/685,042
• Publication Number:
  US 20140149408A1
• Time in Patent Office: 1,156 Days
• Field of Search: 707/737, 706/14, 706/46, 703/16, 716/12, 382/144, 382/145, 382/149, 382/181
• US Class Current: 1/1
• CPC Class Codes: G06F 16/2465 Query processing support for facilitating data mining operations in structured databases
• CPC Class Codes: G06F 16/5838 using colour
• CPC Class Codes: G06F 16/903 Querying for retrieval from the web G06F16/953
• CPC Class Codes: G06T 2207/30148 Semiconductor; IC; Wafer
• CPC Class Codes: G06T 7/0004 Industrial image inspection
• CPC Class Codes: G06V 10/255 Detecting or recognising potential candidate objects based on visual cues, e.g. shapes

Abstract

Embodiments of the disclosure include a method for data mining shape based data, the method includes receiving shape data for each of a plurality of data entries and creating a first abstract from the shape data for each of the plurality of data entries. The method also includes organizing the first abstracts into a plurality of groups based on a criterion and creating a second abstract for each data entry in the plurality of groups based on the criterion and information derived from the first abstract.

Claims

1. A computer system for data mining shape based data, the computer system comprising:

a data mining computer having a processor, the processor configured to perform a method comprising;

receiving a shape data for each of a plurality of data entries;

creating a first abstract from the shape data for each of the plurality of data entries, wherein the first abstract is a graphic illustration of a location of a detected error on a semiconductor chip;

organizing the first abstracts into a plurality of groups based on a first criterion, wherein the first criterion is the location of the detected error;

creating a second abstract for each data entry in the plurality of groups based on a second criterion, wherein the second criterion is the shape of the detected error, and information derived from the first abstract, wherein the second abstract is a visual representation of the detected error on the semiconductor chip that illustrates a more detailed view of a common location of detected errors; and

determining a correlation between each of the data entries in the group based on the second abstract, wherein similarities in the shapes of the detected errors in the common location indicate that the detected errors are not random defects.

2. The computer system of claim 1, wherein the first abstract includes a characteristic of the shape data.

3. The computer system of claim 2, wherein each of the plurality of groups of data entries have at least one common characteristic.

4. The computer system of claim 1, wherein the second abstract includes a second characteristic of the shape data and the correlation is based on a similarity of the second characteristic.

5. The computer system of claim 1, further comprises displaying a graphical representation of the second abstracts.

6. The computer system of claim 5, further comprising:

sorting the graphical representation of the second abstracts based on the correlation.

7. A computer program product for data mining shape based data, the computer program product comprising a computer readable storage medium having computer readable program code embodied therewith, the computer readable program code executable by a computer to perform a method comprising:

receiving a shape data for each of a plurality of data entries;

creating a first abstract from the shape data for each of the plurality of data entries, wherein the first abstract is a graphic illustration of a location of a detected error on a semiconductor chip;

organizing the first abstracts into a plurality of groups based on a fires criterion, wherein the first criterion is a location of the detected error;

creating a second abstract for each data entry in the plurality of groups based on a second criterion, wherein the second criterion is the shape of the detected error, and information derived from the first abstract, wherein the second abstract is a visual representation of the detected error on the semiconductor chip that illustrates a more detailed view of a common location of detected errors; and

determining a correlation between each of the data entries in the group based on the second abstract, wherein similarities in the shapes of the detected errors in the common location indicate that the detected errors are not random defects.

8. The computer program product of claim 7, wherein the first abstract includes a characteristic of the shape data.

9. The computer program product of claim 8, wherein each of the plurality of groups of data entries have at least one common characteristic.

10. The computer program product of claim 7, wherein the second abstract includes a second characteristic of the shape data and the correlation is based on a similarity of the second characteristic.

11. The computer program product of claim 7, further comprises displaying a graphical representation of the second abstracts.

12. The computer program product of claim 11, further comprising:

sorting the graphical representation of the second abstracts based on the correlation.