Software Defect Prediction

What is Software Defect Prediction?

Software Defect Prediction (SDP) is the line of research that concerned with building prediction models, which leverage software metrics to predict defect-prone areas of a software.

Motivation

• Identify software locations (e.g., subsystems, files or functions) that quality assurance efforts should focus on. Examples are:
  • Which code changes should I review first? – Which module should I test first?
• Learn from the past mistakes to improve the software development process. Examples are:
  • Why subsystem A is more bug-prone than another subsystem B?
  • What can we learn from the failures of project C to improve the quality of project D?