Monash IT Prof John Grundy to lead prestigious ASE 2020

21 September 2020

Other Organizing Committee members include Dr Chunyang Chen, Dr Aldeida Aleti, A/Prof Rashina Hoda and Australian Research Council Discovery Early Career Researcher Award Fellows Dr Xin Xia, Dr Chakkrit Tantithamthavorn and Dr Li Li. All six are from Monash IT’s Department of Software Systems and Cybersecurity.

Monash IT researchers will have 12 papers presented at ASE 2020. ‘This clearly shows that we’re leading world-class research in automated software engineering,’ said Dr Tantithamthavorn. ‘Our work will help software companies increase the productivity of developers and the quality of their products.’

Dr Xia’s work on the Comment UPdater (CUP) has received an ACM SIGSOFT Distinguished Paper Award. Code comments help developers better understand the program when maintaining software. Using a neural network model, CUP learns comment update patterns to automatically generate a new comment based on a corresponding old one. By making it easier to update comments during code changes, CUP reduces inconsistent, incomplete and outdated comments.

Also highlighted at the conference is Dr Tantithamthavorn’s JITBot, an automated tool that uses artificial intelligence for just-in-time (JIT) prediction of safety-critical software defects. JITBot not only assesses the risk of a code commit, but indicates which features contribute to this risk and how to mitigate it. In doing so, JITBot potentially saves thousands of hours of development time.

Dr Chen’s OwlEye, an innovative deep learning approach to detect graphical user interface display issues, is another ASE 2020 standout. OwlEye targets text overlap, blurred screen and missing images due to device software or hardware incompatibility. It accurately detects and locates these bugs on Google Play and F-Droid apps so that developers can then fix them.

One more notable project is Dr Li’s enhancement of the Jupyter Notebook, an open-source web application for developing and presenting data science projects. Jupyter allows the user to create and share documents that contain live code, equations, visualisations and narrative text. Dr Li has helped develop an ‘Osiris’ prototype to facilitate reproduction and extension of its scientific computations and findings, improving reproducibility by 82.23%.

Join ASE 2020 to network with experts and learn about the latest advances in automated software engineering.