Browse Prior Art Database

Self teach / defect classification system

IP.com Disclosure Number: IPCOM000206080D
Publication Date: 2011-Apr-13
Document File: 4 page(s) / 126K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a process for orthogonal defect classification providing an automated classifier and self teaching defect classification system.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 34% of the total text.

Page 01 of 4

Self teach / defect classification system

Disclosed is a process for orthogonal defect classification providing an automated classifier and self teaching defect classification system. An embodiment of the disclosed process provides a tool development teams may deploy for enhancing the software development process. Using the disclosed process, a set of test scenarios is written, and classified using the type of defects the test scenarios are expected to produce. The automated classifier imports the scenarios for analysis for key words and context. New scenarios are then automatically classified using baseline parameters set using the disclosed.

Using the disclosed process of orthogonal defect classification enables defect classification before a defect is written using verbiage contained in the test scenario. Using the disclosed process typically saves test and development team time during test cycle and post cycle when orthogonal defect classification is evaluated

Orthogonal defect classification (ODC) is a scheme enabling capture of semantics of each software defect quickly. The definition and capture of defect attributes enable mathematical analysis and modeling of defect information possible. Analysis of orthogonal defect classification data provides a valuable diagnostics method for evaluating phases of a software life cycle (for example, design, development, test and service) and the maturity of the product. For example, using ODC is similar to diagnostics performed in medicine using a blood sample from a patient to determine existing health conditions enabling a further determination of corrective actions. Orthogonal defect classification enhances a possibility of increasing the understanding and use of defects well beyond quality. Further explanation may be found in a related article Orthogonal Defect Classification - A Concept for In-Process Measurements (Ram Chillarege, Inderpal S. Bhandari, Jarir K. Chaar, Michael J. Halliday, Diane S. Moebus, Bonnie K. Ray, Man-Yuen Wong, IBM Thomas J. Watson Research Center, 1992 http://www.chillarege.com/articles/odc-concept).

Currently before opening an issue, testers typically take time to classify defects in certain select categories. These categories are pertinent data points used to track overall progress and stability of a code base. The time the tester uses when classifying defects and reclassifying historical defects may be better utilized. In the case of opening multiple defects, the process of orthogonal defect classification becomes an even more cumbersome and repetitive task. In addition, instances typically occur in which defects need to be re-classified as a result of errors by the tester when initially opening the issue. Finally, at the end of the software cycle, defects are analyzed using orthogonal defect classification to gage the quality of the code base as new features and defect fixes are introduced.

Using the disclosed process of orthogonal defect classification...