Transcript PowerPoint Presentation - University of Virginia, Department of

NSF CCLI Showcase
SIGCSE 2008
NSF CCLI Showcase
SIGCSE 2008
Thursday, 10:00 a.m. — 11:30 a.m.
• Project MLExAI: An Innovative Model for Teaching Core AI Concepts
– Ingrid Russell, University of Hartford
– Zdravko Markov, Central Connecticut State University
• Cognitive Robotics with Tekkotsu: A Curriculum for Machines That
See and Manipulate Their World
– David S. Touretzky, Carnegie Mellon University
– Ethan J. Tira-Thompson, Carnegie Mellon University
– Marsha C. Lovett, Carnegie Mellon University
• Supporting Service-Learning Projects in Software Engineering Project
Classes
– Chang Liu, Ohio University
Project MLExAI: An Innovative Model for Teaching
Core AI Concepts
Ingrid Russell, University of Hartford
Zdravko Markov, Central Connecticut State University
Project Goal
Objectives
The goal is to develop a framework for
teaching core AI topics with a unifying
theme of machine learning. A suite of 26
term-long projects are developed, each
involving the design and implementation of
a machine learning system that enhances a
commonly-deployed application.
•Enhance student learning experience by
implementing a unifying theme of machine learning
to tie together core AI topics.
The applications span a large area including
network security, recommender systems,
game playing, intelligent agents,
computational chemistry, robotics,
conversational systems, cryptography, web
document classification, vision, data
integration in databases, bioinformatics,
pattern recognition, and data mining.
•Highlight the bridge that machine learning provides
between AI technology and modern software
engineering.
•Increase student interest and motivation to learn AI
by providing a framework for the presentation of the
major AI topics that emphasizes the strong
connection between AI and computer science.
•Introduce students to an increasingly important
research area, thus motivating them to pursue
further study in this area.
Cognitive Robotics with Tekkotsu:
A Curriculum for Machines That See and Manipulate Their World
Educational robots are becoming more interesting:
• Substantial computing power on board (Pentium, Linux, C++)
• Webcam for color vision; Wireless ethernet; Arms with grippers
So CS robotics courses should be more ambitious:
• Provide appropriate high level tools: easy-to-use computer vision software,
kinematics solvers, map builders, particle filters, etc.
• Teach students to use these to develop interesting robot behaviors.
Carnegie Mellon’s Cognitive Robotics course:
• Software and lecture notes available at Tekkotsu.org.
• Several supported hardware platforms, with more to come.
NSF DUE-0717705
David S. Touretzky, Ethan J. Tira-Thompson, Marsha C. Lovett: Carnegie Mellon
Supporting Service-Learning Projects in
Software Engineering Project Classes
Chang Liu, School of EECS, Ohio University
•
Problem: It is a challenge to manage real-world projects, often based on substantial
code base, in single-term courses.
•
Solution: We developed and integrated a number of learning aids and teaching
techniques, including
– Simulated team project process exercise in 3-D online virtual worlds
– Simulated team software specification exercise in 3-D online virtual worlds
– Adoption of Wiki as a tool to facilitate team communication
– Shared code segments from previous assignments among students
•
Preliminary Results:
– We have applied our approach in a Software Engineering project class at Ohio
University several times.
– The learning aids were effective in motivating students and enhancing their
learning. They facilitated successful adoption of service-learning projects in our
single-quarter project classes.
1
NSF CCLI Showcase
SIGCSE 2008
Friday 10:00 a.m. — 11:30 a.m.
•
Integrating Privacy Ethics into the Undergraduate Database Curriculum
–
•
Personalized Exploratorium for Database Courses
–
–
–
–
–
–
•
Peter Brusilovsky, University of Pittsburgh
Vladimir Zadorozhny, University of Pittsburgh
Danielle H. Lee, University of Pittsburgh
Sergey Sosnovsky, University of Pittsburgh
Michael V. Yudelson, University of Pittsburgh
Xin Zhou, University of Pittsburgh
The Affinity Research Group Model: Creating and Maintaining Effective
Research Teams
–
–
•
Florence Appel, Saint Xavier University
Ann Q. Gates, University of Texas at El Paso
Steve Roach, University of Texas at El Paso
Problets: Practice Exercises for Computer Science I
–
Amruth Kumar, Ramapo College of New Jersey
Integrating Ethics Into the Database Curriculum
• This is an EMD project to support the introduction of privacy
content into the introductory database course
• The centerpiece is a set of privacy modules that can be
introduced systematically throughout the design thread of
the course, or used in a stand-alone context
• Each module addresses privacy issues arising normally
during a given phase of database design
• Modules contain class discussion exercises, homework
assignments, test questions, teaching tips & resources
• A full-service website is also under development
• The goal is to sensitize students to database privacy issues
& enable them to implement privacy safeguards
Florence Appel * Saint Xavier University * DUE 0442637
• Automatically evaluated exercises and interactive examples for
database course
• Parameterized SQL problem generator and evaluator, SQL-KnoT
(SQL Knowledge Tester) for increasing problem solving skills
• Adaptive and personalized guidance to increase student
engagement and success rate
• Comprehensive database course system by the integration with
automatic evaluation system of exercises, SQL-Tutor (University
of Canterbury) and various contents from several universities
NSF DUE NSF- 0633494
Primary PI: Peter Brusilovsky, Associate Professor, School of Information Sciences
Co-PI: Vladimir Zadorozhny, Associate Professor, School of Information Sciences
http://adapt2.sis.pitt.edu/cbum/
AFFINITY RESEARCH GROUP MODEL: Developing
Students Beyond Academe
A comprehensive model comprised of fundamental principles and effective practices for
involving undergraduates in research.
NSF DUE-0443061
 Key Components:
o Core Ideology
o Student Connectedness
o Deliberate Practices
 Results:
o Higher retention
o Development of professional,
research, technical skills
o Higher level of confidence
o Preparation for success in graduate
studies and workforce
Ann Gates Steve Roach Elsa Villa
The University of Texas at El Paso Department of Computer Science
Problets
Provide Practice Exercises for Computer Science I

Unique features:






Adapt to the learning needs of the student

Maximize learning while minimizing the time spent

Proven to help students learn on their own
Explain the step-by-step execution of programs
Usable for:

Closed-lab exercises, after-class assignments, in-class tests


Expressions (Arithmetic, Relational, Logical), if, if-else, while, for, C++
pointers
C, C++, Java and C#

Used over 2600 times, adopted by 24+ teachers in 2007-08
Available for:
Free to use; Easy to adopt, use, and track student progress
Details at www.problets.org
Contact: [email protected]
NSF CCLI Showcase
SIGCSE 2008
Friday 2:30 p.m. — 4:00 p.m.
• Cooperative Learning Methods for Java-Based CS1 Courses
– Leland L. Beck, San Diego State University
– Alexander W. Chizhik, San Diego State University
• Teaching by Collaborating: Toward a Pedagogy Adapted from the
Open Source Culture
– John David N. Dionisio, Loyola Marymount University
– Ray Toal, Loyola Marymount University
• Building and Using an Emulab
– W. David Laverell, Calvin College
– Timothy H. Brom, University of Kentucky
Cooperative Learning Methods for Java-Based CS1 Courses
Leland L. Beck
Alexander W. Chizhik
San Diego State University
• A comprehensive set of cooperative
learning activities for CS1
• Average final exam scores 25-30% higher
for cooperative learning sections than for
lecture sections
Final Exam Scores by Section
Lecture Sections
• We’ll provide instructor’s notes, training,
and consultation
25
Frequency
• Will you join us in exploring these
materials with your own students?
Cooperative Learning Sections
30
20
15
10
5
0
11-20
21-30
31-40
41-50
51-60
61-70
71-80
Please join us here on Friday between 2:30 and 4:00 to see video
showing this approach in action in the classroom!
81-90
91-100
Teaching by Collaborating:
Toward a Pedagogy Adapted from the
Open Source Culture
Open Source Culture ➟ Set of Values
•Source code is available and long-lived
•Accountability implies community
•Responsibilities accompany rights
John David N. Dionisio
Ray Toal
Department of Electrical Engineering & Computer Science
Loyola Marymount University, Los Angeles, California
Curriculum Arc
Year One: Study, Testing,
and Fixing of Pre-Existing
Code
Pre-existing is key here:
conventional teaching at this
level usually involves writing
“toy” programs from scratch.
Year Two: Coding and
Testing Specific Functions
from Scratch
From the open source
culture’s value of community
and accountability, unit tests
serve as an unambiguous
(and easily automated)
mechanism for validating the
correctness of submitted
work.
Year Three: Term-Length,
Focused Projects
The long life of code becomes
apparent: software written at
the junior level and beyond
finds its way back to the
freshmen, as pre-existing
code that must be examined,
fixed, and completed.
Year Four: Capstone Projects
Capstone projects continue to
“live” on beyond the graduations
of their creators. Future
students are exposed to this
code and may contribute to it in
their own courses.
Instructional Techniques
Sample Code Bazaar
Live, organized, searchable,
student-accessible sample
code libraries introduce
students to the creation of
“derived works” from the
sample code.
The Cyclic Life of Code
Student work becomes a
permanent part of a source
code repository, available
across courses and facilitating
incremental improvement or
additional functionality.
Test Infection
Instructors run student test
suites against a variety of
implementations to eliminate
false positives, after which
each student implementation
is run against each set of
tests.
Release Early, Often, & Open
Code-in-progress is visible and
sharable to fellow students.
Students who produce widely
reused code must be duly
acknowledged, just as
widespread adoption can be
viewed as a measure of
success for an open source
project.
Building and Using an Emulab
• Emulation versus dedicated labs and
simulators
• Building an Emulab at Calvin College
• Creating network topologies and running
experiments (Live Demo)
• Conclusions
NSF CCLI Showcase
SIGCSE 2008
Saturday 10:00 a.m. — 11:30 a.m.
• Cultural Challenges in Teaching Alice in Hawaii
– Judith L. Gersting, University of Hawaii at Hilo
– Keith Edwards, University of Hawaii at Hilo
• WeBWorK in Computer Science
–
–
–
–
Christelle Scharff, Pace University
Andy Wildenberg, Cornell College
Olly Gotel, Pace University
Richard Kline, Pace University
• Tablet PC's as Mind Tools for Enhancing Thinking and Learning Skills
– Cheryl Willis, University of Houston
Narrative Programming at Univ. of Hawaii at Hilo
• Uses Alice to animate Hawaiian legends
– Challenges
• Authentic images
• Cultural sensitivity and respect
– What works
• Provides positive introduction to computer science
– What doesn’t
• Does not pave the way for success in CS-1
in CS
Collaborative Research: Adapting and
Extending WeBWorK for Use in the
Computer Science Curriculum
Project Web Page: http://www.csis.pace.edu/~scharff/webwork
WeBWorK Server: http://atlantis.seidenberg.pace.edu/wiki/webwork2
Dr. Christelle Scharff, Dr. Olly Gotel, Dr. Richard Kline
[email protected], [email protected], [email protected]
Pace University, New York
Dr. Andrew Wildenberg
Cornell College, Iowa
[email protected]
Tablet PCs as MindTools to
Enhance Learning Skills
Cheryl L. Willis, Ph.D.
University of Houston
CCLI Showcase