S3M: Siamese Stack (Trace) Similarity Measure
Automatic crash reporting systems have become a de-facto standard in software development. These systems monitor target software, and if a crash occurs they send details to a backend application. Later on, these reports are aggregated and used in the development process to 1) understand whether it is a new or an existing issue, 2) assign these bugs to appropriate developers, and 3) gain a general overview of the application’s bug landscape. The efficiency of report aggregation and subsequent operations heavily depends on the quality of the report similarity metric. However, a distinctive feature of this kind of report is that no textual input from the user (i.e., bug description) is available: it contains only stack trace information.
In this paper, we present S3M (“extreme”)~— the first approach to computing stack trace similarity based on deep learning. It is based on a siamese architecture that uses a biLSTM encoder and a fully-connected classifier to compute similarity. Our experiments demonstrate the superiority of our approach over the state-of-the-art on both open-sourced data and a private JetBrains dataset. Additionally, we review the impact of stack trace trimming on the quality of the results.
Wed 19 MayDisplayed time zone: Amsterdam, Berlin, Bern, Rome, Stockholm, Vienna change
17:00 - 17:50
|S3M: Siamese Stack (Trace) Similarity Measure
Aleksandr Khvorov JetBrains, ITMO University, Roman Vasiliev JetBrains, George Chernishev Saint-Petersburg State University, Irving Muller Rodrigues Polytechnique Montreal, Montreal, Canada, Dmitrij Koznov Saint-Petersburg State University, Nikita Povarov JetBrainsPre-print
|Mining the ROS ecosystem for Green Architectural Tactics in Robotics and an Empirical Evaluation
Ivano Malavolta Vrije Universiteit Amsterdam, Katerina Chinnappan Vrije Universiteit Amsterdam, Stan Swanborn Vrije Universiteit Amsterdam, The Netherlands, Grace Lewis Carnegie Mellon Software Engineering Institute, Patricia Lago Vrije Universiteit AmsterdamPre-print Media Attached
|Mining Energy-Related Practices in Robotics Software
Michel Albonico UTFPR, Ivano Malavolta Vrije Universiteit Amsterdam, Gustavo Pinto Federal University of Pará, Emitzá Guzmán Vrije Universiteit Amsterdam, Katerina Chinnappan Vrije Universiteit Amsterdam, Patricia Lago Vrije Universiteit AmsterdamPre-print Media Attached
|Mining API Interactions to Analyze Software Revisions for the Evolution of Energy Consumption
Andreas Schuler University of Applied Sciences Upper Austria, Gabriele Anderst-Kotsis Johannes Kepler University, Linz, AustriaPre-print
|Can I Solve it? Identifying the APIs required to complete OSS tasks
Fabio Marcos De Abreu Santos Northern Arizona University, USA, Igor Scaliante Wiese Federal University of Technology – Paraná - UTFPR, Bianca Trinkenreich Northern of Arizona Univeristy, Igor Steinmacher Northern Arizona University, USA, Anita Sarma Oregon State University, Marco Gerosa Northern Arizona University, USAPre-print
|Discussions and Q&A