- This event has passed.
Cafe DSL/Centre for Oncology Informatics Seminar
July 10 @ 11:30 am - 12:30 pm
The Risks of Unsound Software Analytics: Why Practitioners Should Care?
Many software organisations (e.g., Microsoft, Facebook, and Google) currently use powerful data analytics (e.g., statistical and machine learning techniques) to predict and explain the risks of software changes that lead to future defects. However, such predictions and explanations may be invalid if practitioners do not consider the risks of unsound software analytics, leading to invalid predictions and explanations. In the past 4 years, I have conducted a series of empirical investigation to evaluate techniques and identify best practices to develop theoretically sound and actionable analytical models to improve software quality and development processes. In this talk, I will discuss why practitioners should care about the risks of building unsound software analytics. Throughout the talk, I will provide concrete examples while proposing best practices that our SE community should follow to avoid such risks.
Dr. Chakkrit Tantithamthavorn is a lecturer at the School of Computer Science, the University of Adelaide, Australia. Prior to that, he was a research fellow at Queen’s University (Canada) and Nara Institute of Science and Technology (Japan). During his Ph.D. study, he won one of the most prestigious funding in Japan, i.e., a JSPS Research Fellowship for Young Researchers and a Grants-in-Aid for JSPS Fellow, and won a “Best Ph.D. Student Award”. His work has been published at several top-tier software engineering venues, such as the IEEE Transactions on Software Engineering (TSE), the Springer Journal of Empirical Software Engineering (EMSE), and the International Conference on Software Engineering (ICSE). His current research aims to address the fundamental issues of analytical modelling for software engineering in order to produce more accurate predictions and explanations. His research interests also include empirical software engineering and mining software repositories (MSR). More about Chakkrit and his work is available online.