Logical Clustering and Learning for Time-Series Data

Marcell Vazquez-Chanlatte, Jyotirmoy V. Deshmukh, Xiaoqing Jin, and Sanjit A. Seshia. Logical Clustering and Learning for Time-Series Data. In 29th International Conference on Computer Aided Verification (CAV), pp. 305–325, 2017.

Download

[pdf] 

Abstract

In order to effectively analyze and build cyberphysical systems (CPS), designers today have to combat the data deluge problem, i.e., the burden of processing intractably large amounts of data produced by complex models and experiments. In this work, we utilize monotonic parametric signal temporal logic (PSTL) to design features for unsupervised classification of time series data. This enables using off-the-shelf machine learning tools to automatically cluster similar traces with respect to a given PSTL formula. We demonstrate how this technique produces interpretable formulas that are amenable to analysis and understanding using a few representative examples. We illustrate this with case studies related to automotive engine testing, highway traffic analysis, and auto-grading massively open online courses.

BibTeX

@inproceedings{vazquez-cav17,
  author    = {Marcell Vazquez{-}Chanlatte and
               Jyotirmoy V. Deshmukh and
               Xiaoqing Jin and
               Sanjit A. Seshia},
  title     = {Logical Clustering and Learning for Time-Series Data},
  booktitle = {29th International Conference on Computer Aided Verification (CAV)},
  pages     = {305--325},
  year      = {2017},
  abstract  = {In order to effectively analyze and build cyberphysical systems (CPS), designers today have to combat the data deluge problem, i.e., the burden of processing intractably large amounts of data produced by complex models and experiments. In this work, we utilize monotonic parametric signal temporal logic (PSTL) to design features for unsupervised classification of time series data. This enables using off-the-shelf machine learning tools to automatically cluster similar traces with respect to a given PSTL formula. We demonstrate how this technique produces interpretable formulas that are amenable to analysis and understanding using a few representative examples. We illustrate this with case studies related to automotive engine testing, highway traffic analysis, and auto-grading massively open online courses.},
}