A Novel Data Stream Learning Approach to Tackle One-Sided Label Noise From Verification Latency

Abstract

Where a licence is displayed above, please note the terms and conditions of the licence govern your use of this document.When citing, please reference the published version. Take down policy While the University of Birmingham exercises care and attention in making items available there are rare occasions when an item has been uploaded in error or has been deemed to be commercially or otherwise sensitive.

A novel instance‐based method for cross‐project just‐in‐time defect prediction

“…Each of the selected projects has a substantial development history, containing a considerable amount of change data in its repository. To manage the computational complexity and time required for experimentation, we follow the approach adopted in other defect prediction studies 40 . Specifically, we use a subset of the most recent 5000 software changes in each project's repository as the experimental data.…”

“…To manage the computational complexity and time required for experimentation, we follow the approach adopted in other defect prediction studies. 40 Specifically, we use a subset of the most recent 5000 software changes in each project's repository as the experimental data. Table 1 provides a concise summary of each dataset utilized, encompassing details such as the project name (Project), the total number of cases (#Total Commits), the number of selected cases (#Commits), the count of defect-prone changes (#Defect), the defect ratio (%Defect), and the programming language (Language) employed by each project.…”