[Colloquium] Reminder: Pham/Dissertation Defense/Jul 21, 2014

Margaret Jaffey margaret at cs.uchicago.edu
Sun Jul 20 08:02:09 CDT 2014



       Department of Computer Science/The University of Chicago

                     *** Dissertation Defense ***


Candidate:  Quan Pham

Date:  Monday, July 21, 2014

Time:  10:00 AM

Place:  Ryerson 277

Title: A Reproducible Framework for Computational Research

Abstract:
In today's world of publishing, reproducing research results has
become challenging as scientific research has become inherently
computational. Encoding a computation-based result in a text-based
paper is nearly impractical, leading to the overarching research
question. ``Can computation-based research papers be reproducible?’’

The aim of this thesis is to describe frameworks and tools, which if
provided to authors can aid in assessing computational
reproducibility. Towards this aim, the thesis proposes a reproducible
framework for creating descriptive and interactive publications by
linking them with associated science objects, such as source codes,
datasets, annotations, workflows, re-executable software package,
process and data provenance, and virtual machine images. To create
science objects in a linkable representation for use within research
papers, the thesis describes a set of tools as part of the framework.
In particular, it focuses on Provenance-To-Use (PTU), an application
virtualization tool that encapsulates source code, data, and all
associated data and software dependencies into a package. We describe
how by capturing data dependencies, PTU allows full and partial
repeatability of the virtualized software; we describe how by
capturing software dependencies, PTU can be used for building and
maintaining software pipelines. Finally, we show how PTU can be used
to provide computational reproducibility in a distributed environment
and in environments where data is resident in databases.

We evaluate and validate the framework by using several publication
use cases and determine the extent to which computational
reproducibility is achievable.

Quan's advisor is Prof. Ian Foster

Login to the Computer Science Department website for details,
including a draft copy of the dissertation:

 https://www.cs.uchicago.edu/phd/phd_announcements#quanpt

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Margaret P. Jaffey            margaret at cs.uchicago.edu
Department of Computer Science
Student Support Rep (Ry 156)               (773) 702-6011
The University of Chicago      http://www.cs.uchicago.edu
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=


More information about the Colloquium mailing list