Systems and methods for determining relationships between defects

Patent metadata

Field	Value
Publication number	US-10620618-B2
Application number	US-201615385664-A
Country	US
Kind code	B2
Filing date	Dec 20, 2016
Priority date	Dec 20, 2016
Publication date	Apr 14, 2020
Grant date	Apr 14, 2020

Abstract

Systems and methods are provided for identifying relationships between defects. The system may obtain defect items and associated information. Defect items may be compared to one another based on their attributes to determine how related they are. According to the comparisons, defect items may be grouped together into issue items for further analysis by a user. The system may further update a defect comparison model according to user interaction with defect items.

First claim

The invention claimed is:

1. A system comprising:
one or more processors; and
memory storing instructions that, when executed by the one or more processors, cause the system to:
obtain a first defect data object stored in a database, the first defect data object including a first unstructured data field and a first structured data field;
determine a set of respective pairwise distances between the first defect data object and at least a portion of a set of second defect data objects stored in the database, each of the second defect data objects of the at least a portion of the set of second defect data objects including a respective second unstructured data field and a respective second structured data field, a first term of the respective pairwise distances being based on a first weighted comparison between the first unstructured data field and the respective second unstructured data field, and a second term of the respective pairwise distances being based on a second weighted comparison between the first structured data field and the respective second structured data field;
store the set of respective pairwise distances in the database;
compare one or more respective pairwise distances of the set of respective pairwise distances stored in the database to a threshold pairwise distance;
store one or more results of the comparison in the first defect data object stored in the database;
identify, based on the one or more results of the comparison stored in the first defect data object stored in the database, one or more of the second defect data objects as being related to the first defect data object;
store in the database an issue item comprising the first defect data object and the one or more related second defect data objects;
receive a request to present the first defect data object;
retrieve, based on the request to present the first defect data object, the issue item from the database; and

present at least a portion of the one or more related second defect data objects in response to retrieving the issue item from the database.

2. The system of claim 1 , wherein the second term of the respective pairwise distances is based on a summation of the second weighted comparison and one or more other weighted comparisons.

3. The system of claim 2 , wherein one or more of the first and second weighted comparisons are weighted according to a set of model weight parameters.

4. The system of claim 3 , wherein the system is further caused to:
update the set of model weight parameters according to the modification of the issue item.

5. The system of claim 1 , wherein the issue item is stored as an issue data object, and wherein to store the issue item the issue data object, the system is further caused to:
compute an issue quality score of the issue according to the respective pairwise distances between the first defect data object and the related second defect data objects.

6. The system of claim 1 , wherein the system is further caused to:
receive, from the user, a modification of the issue item, wherein the modification includes an additional defect data object or the removal of a defect data object.

7. The system of claim 6 , wherein to identify the one or more second defect items the system is further caused to:
obtain a stored issue item comprising the related one or more of the second defect data objects; and
to store the issue item the system is further caused to add the first defect data object to the stored issue item to generate the issue item.

8. The system of claim 1 , wherein the system is further caused to:
obtain a first issue item, the first issue item including the first defect item and a plurality of additional defect data objects;
determine one or more additional sets of respective pairwise distances between the additional defects and a least a portion of the set of second defect data objects;
compare the additional sets of respective pairwise distances to the threshold pairwise distance;
identify, based on the comparison, one or more of the second defect data objects as related to the first issue item;
store the issue item comprising the first defect data object, the additional defect data object, and the related second defect data objects.

9. A computer implemented method for rules-based identification of relationships between defects, the method being performed on a computer system having one or more physical processors programmed with computer program instructions that, when executed by the one or more physical processors, cause the computer system to perform the method, the method comprising:
obtaining, by the computer system, a first defect data object stored in a database, the first defect data object including a first unstructured data field and a first structured data field;
determining, by the computer system, a set of respective pairwise distances between the first defect data object and at least a portion of a set of second defect data objects stored in the database, each of the second defect data objects of the at least a portion of the set of second defect data objects including a respective second unstructured data field and a respective second structured data field, a first term of the respective pairwise distances being based on a first weighted comparison between the first unstructured data field and the respective second unstructured data field, and a second term of the respective pairwise distances being based on a second weighted comparison between the first structured data field and the respective second structured data field;
storing, by the computer system, the set of respective pairwise distances in the database;
comparing, by the computer system, one or more respective pairwise distances of the set of respective pairwise distances stored in the database to a threshold pairwise distance;
storing, by the computer system, one or more results of the comparison in the first defect data object stored in the database;
identifying, by the computer system, based on the one or more results of the comparison stored in the first defect data object stored in the database, one or more of the second defect data objects as being related to the first defect data object;
storing, by the computer system, an issue item in the database, the issue item comprising the first defect data object and the one or more related second defect data objects;
receiving, by the computer system, a request to present the first defect data object;
retrieving, by the computer system, based on the request to present the first defect data object, the issue item from the database; and
presenting, by the computer system, at least a portion of the one or more related second defect data objects in response to retrieving the issue item from the database.

10. The method of claim 9 , wherein the second term of the respective pairwise distances is based on a summation of the second weighted comparison and one or more other weighted comparisons.

11. The method of claim 10 , wherein one or more of the first and second weighted comparisons are weighted according to a set of model weight parameters.

12. The method of claim 11 , further comprising:
updating, by the computer system, the set of model weight parameters according to the modification of the issue item.

13. The method of claim 9 , wherein the issue item is stored as an issue data object, and wherein storing the issue item further includes:
computing, by the computer system, an issue quality score of the issue according to the respective pairwise distances between the first defect data object and the related second defect data objects.

14. The metho

Assignees

• Palantir Technologies Inc

Inventors

• Poh Andrew
• Menck Andre Frederico Cavalheiro
• Sprague Arion
• Grabham Benjamin
• Lee Benjamin
• Rahill-Marier Bianca
• Omont Gregoire
• Inoue Jim
• Scheinerman Jonah
• Albin Maciej
• Scolnick Myles
• Gribelyuk Paul
• Fackler Steven
• Nguyen Tam-Sanh
• Powell Thomas
• Seaton William

Classifications

• H10P74/23

  characterised by multiple measurements, corrections, marking or sorting processes · CPC title

• G06Q10/06395

  Quality analysis or management · CPC title

• G06N20/00 Primary

  Machine learning · CPC title

• G05B23/0235 Primary

  based on a comparison with predetermined threshold or range, e.g. "classical methods", carried out during normal operation; threshold adaptation or choice; when or how to compare with the threshold · CPC title

• G06F40/30

  Semantic analysis · CPC title