Thu 8 Nov 2018 10:52 - 11:15 at Horizons 10-11 - Models Chair(s): Maurício Aniche

According to
psychological scientists, humans understand models that most match their own internal models,
which they characterize as lists of "heuristic"s (i.e. lists of very succinct rules).
One such heuristic rule generator is the Fast-and-Frugal Trees (FFT) preferred by psychological scientists.
Despite their successful use in many applied domains, FFTs have not been applied in software
Accordingly, this paper assesses FFTs for software analytics.

We find that FFTs are remarkably effective in that their models are very succinct (5 lines or less describing a binary decision tree) while also
outperforming result from very recent, top-level,
conference papers.
Also, when we restrict training data to operational attributes (i.e., those attributes that are frequently changed by developers), the performance of FFTs are not effected
(while the performance of other learners can vary wildly).

Our conclusions are two-fold. Firstly,
there is much that software analytics community could learn from psychological science.
Secondly, proponents of complex methods should always baseline those methods against simpler alternatives.
For example, FFTs could be used as a standard baseline learner against which other software analytics tools are compared.

Conference Day
Thu 8 Nov

Displayed time zone: Guadalajara, Mexico City, Monterrey change

10:30 - 12:00
ModelsResearch Papers / Journal-First at Horizons 10-11
Chair(s): Maurício AnicheDelft University of Technology, Netherlands
The modular and feature toggle architectures of Google Chrome
Applications of Psychological Science for Actionable Analytics
Research Papers
Di ChenNorth Carolina State University, USA, Wei Fu, Rahul KrishnaNC State University, Tim MenziesNorth Carolina State University
Putback-Based Bidirectional Model Transformations
Research Papers
Xiao HeUniversity of Science and Technology Beijing, China, Zhenjiang HuNational Institute of Informatics
Model Transformation Languages under a Magnifying Glass: A Controlled Experiment with Xtend, ATL, and QVT
Research Papers
Regina HebigChalmers University of Technology & University of Gothenburg, Christoph SeidlTechnische Universität Braunschweig, Thorsten BergerChalmers University of Technology, Sweden / University of Gothenburg, Sweden, John Kook PedersenIT University of Copenhagen, Denmark, Andrzej WąsowskiIT University of Copenhagen, Denmark