First International Workshop on Software Defect Datasets (SDD 2023)
As technology advances, software plays an ever-increasing role in our daily lives. Unsurprisingly, improving software quality is a central priority of our time. Approaches for understanding, diagnosing, localizing and fixing software defects are usually empirically grounded in datasets of software defects and their fixes. In recent years, a large number of defect datasets have been created, facilitating the evaluation of new techniques and enabling the study of software defects at a scale never seen before. Despite these advances, defect datasets face important challenges for their creation, maintainability and use.
The SDD workshop aims at bringing together researchers and practitioners involved in the creation and use of defect datasets for software engineering research. The goal is to open discussion on best practices for creating defect datasets, current challenges faced by today’s datasets and the needs of the software engineering community with regards to defect datasets. The workshop will feature contributed papers and invited talks in this area.
Workshop Topics
Topics of interest include, but are not limited to:
- Design and development of defect datasets.
- Maintenance and evolution of defect datasets.
- Classification of defects in large datasets.
- Usability of defect datasets.
- Evaluation metrics for defect datasets.
External workshop website
Fri 8 DecDisplayed time zone: Pacific Time (US & Canada) change
09:00 - 10:30 | |||
09:00 30mTalk | SecBench.js: An Executable Security Benchmark Suite for Server-Side JavaScript SDD Michael Pradel University of Stuttgart | ||
09:30 30mTalk | BugSwarm: Overview, Lessons Learned and Opportunities SDD Cindy Rubio-González University of California at Davis | ||
10:00 30mTalk | Using the Normalized Java Resource to Find Software Defects SDD Jens Palsberg University of California, Los Angeles (UCLA) |
10:30 - 11:00 | |||
11:00 - 12:30 | |||
11:00 30mTalk | Defect mining is easy, curation is hard: lessons learned from building the Defects4J benchmark SDD René Just University of Washington | ||
11:30 30mTalk | On the Reproducibility of Software Defect Datasets SDD Hao-Nan Zhu University of California, Davis | ||
12:00 30mTalk | Building Reproducible Flaky Test Datasets SDD Jonathan Bell Northeastern University File Attached |
12:30 - 14:00 | |||
14:00 - 15:30 | |||
14:00 30mTalk | BugsInPy: A Database of Existing Bugs in Python Projects SDD | ||
14:30 20mTalk | Code Revert Prediction with Graph Neural Networks: A Case Study at JP Morgan SDD Yulong Pei J.P. Morgan AI Research, Salwa Alamir J.P. Morgan AI Research, Rares Dolga J.P. Morgan AI Research, Sameena Shah J.P. Morgan AI Research | ||
14:50 20mTalk | Towards Assessing the Real-World Impact of Defects in Blockchain-based Smart Contracts SDD Michael Hettmer University of Duisburg-Essen, Essen, Germany, Benedikt Severin University of Duisburg-Essen, Florian Blum University of Duisburg-Essen, Volker Gruhn University Duisburg-Essen | ||
15:10 20mTalk | Log Summarisation for Defect Evolution Analysis SDD Rares Dolga J.P. Morgan AI Research, Ran Zmigrod JP Morgan - Chase, Rui Silva JP Morgan - Chase, Salwa Alamir JP Morgan - Chase, Sameena Shah J.P. Morgan AI Research |
15:30 - 16:00 | |||
16:00 - 17:00 | |||
16:00 60mPanel | Software defect datasets: experiences, needs and opportunities. SDD Frank Tip Northeastern University, Jens Palsberg University of California, Los Angeles (UCLA), Ridwan Salihin Shariffdeen National University of Singapore |
Accepted Papers
Call for Papers
Authors are invited to submit manuscripts in English structured as short papers at a length of at most 4 pages of content, excluding references. Short papers can be one of the following:
- Position papers that describe issues related to the creation or use of defect datasets for software engineering research,
- Experience papers that describe best practices or challenges encountered when creating or using defects datasets, or
- Extended abstracts that describe work in progress, current challenges and/or opportunities in the area of defect datasets.
Submissions must conform to the ESEC/FSE Format and Submission Guidelines. Papers not following these guidelines will be rejected without review. At least one author of an accepted paper must register for and attend the workshop.