Browse Prior Art Database

An Analytic Engine for Automated Defect Review

IP.com Disclosure Number: IPCOM000236050D
Publication Date: 2014-Apr-03
Document File: 3 page(s) / 94K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is an analytic engine that automates the review process for reported software defects. The proposed system is capable of making intelligent defect review decisions such as identification of duplicate defects, determining the defects that can be deferred, expected time to resolve, etc.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 52% of the total text.

Page 01 of 3

An Analytic Engine for Automated Defect Review

In the event that a large number of defects arise during the development and launch phases of a software solution, the software architect must analyze the defects, identify the duplicates, defer the less important defects, and close any unrelated reports. The problem is that this is a manual process, which is time-consuming and can delay the release of a software solution.

A tool is needed that provides a summary of the individual defects for a given

software product of an enterprise.

The novel contribution is an analytic engine that automates the defect review process. The proposed system is capable of making intelligent defect review decisions such as identification of duplicate defects, determining the defects that can be deferred, expected time to resolve, etc. This relieves the software architect of manually reading all the details of the defect.

While the proposed engine works as a back-end component, the user interface (UI) of a defect report contains a Review button that facilitates the automated defect review process.

Figure 1: UI of the page displaying the defect along with the "Review" button

Once the defect reviewer clicks/hits the Review button, the system analyzes the defect and shows various review metrics as output. Each of these metrics comes

with a probability of occurrence in reality.

Figure 2: View of the page after the analysis is complete

Following are the details of each metric for which the defect is reviewed. Existing text analysis tools can be used for this purpose.

1


Page 02 of 3

Does the defect have adequate information?

• Applicable only to "New" defects • State of the defect: if this is already "Active/Resolved" that means defect owner already knows the problem, so there is adequate information • Output: Yes/No with a probability • The engine checks the following: - Inadequate info can be because of one of the following reasons: Build not specified
Neither logs nor screenshots are attached
Problem statement does not come up with adequate description. For example, somebody mentioned UI did not display properly but there was not supporting statements about

what else happened.

- For each of these cases, run a search in the defect database
Find how many defects were closed because of any of these reasons

Assign a probability based on the history

Can this defect be deferred from the upcoming product release?

• Applicable on...