Looking for micro-process in large-scale data

  • Authors:
  • Minghui Zhou

  • Affiliations:
  • Peking University, Beijing, China

  • Venue:
  • Proceedings of the 2nd international workshop on Evidential assessment of software technologies
  • Year:
  • 2012

Quantified Score

Hi-index 0.00

Visualization

Abstract

Micro-processes (or micro-patterns) are fine-scale practices used by software projects to accomplish key tasks: reporting and resolving issues, debugging code, mentoring newcomers, and so forth. It is a challenge to discover micro-patterns that may guide the practice across projects. We propose a large-scale evidence based approach. The first step is to collect data from Internet and build a standardized public repository that mirrors the open source universe. In the second step, we could start from specific projects and then generalize the findings across the universe, or, we could target the whole universe projects first and then cluster the findings. Examples are given about how it is proceeded.