kr brunch!
A Semi-Regular Student Reading Group in Logic, Ontologies, and Reasoning
What Is It?
There are many people in the local community who either are, soon
will be, or might possibly be interested in applying knowledge
representation and reasoning to their work. This reading group is
intended to be a forum for those people to get together, learn more
about existing work and literature, and discuss their own efforts.
Most importantly, refreshments will be provided. The group will also
most likely very rarely actually meet during decent brunching hours.
Schedule
UPCOMING!
Monday, June 25, 2007: Logic and Databases.
Where: Bossone 303, 4pm, 2k7/6/25
Topics: Relationships between databases and logic, the relational algebra
and fragments of first order logic, and query classes.
• A. Chandra.
Theory of Database Queries.
In Proceedings of the Seventh ACM SIGACT-SIGMOD-SIGART Symposium on Principles of Database Systems, pgs. 1--9. Austin, Texas, USA, 1988. ACM Press.
This is available from the ACM
here; a local copy is
here.
• M. Vardi.
Complexity of Relational Query Languages.
In Proceedings of the 14th ACM Symposium on Theory of Computing, pgs 137--146.
1982. ACM Press.
This is available from the author
here; a local copy is
here.
• V. Tannen.
Friendly Logics Course Notes. 2005.
Val's finally put these up,
here. They're quite good.
• Any good refence on the relational algebra. If you're so inclined,
looking at something short on finite model theory may also be useful.
PAST
Monday, April 9, 2007: Friendly Logics---Description Logics and Other FOL Fragments.
Where: Bossone 303, 4pm, 2k7/4/9
Topics: Expressiveness and computational properties of description logics
and other fragments of first order logic. OWL-Lite, OWL-DL, OWL-Full.
Pizza & pop courtesy
GICL@Drexel.
• Brachman and Levesque, "Knowledge Representation and Reasoning."
Morgan Kaufman, 2004. Chapter 16: "The Tradeoff Between Expressiveness and Tractability," specifically the first half---pages 327--336.
Definitely a good book. This part provides
a very understandable example of the computational difference between two description logics.
Ask Joe to borrow his copy or convince the library to get a copy.
• F. Baader, S. Brandt, and C. Lutz. Pushing the EL Envelope. In Proceedings of the Nineteenth International Joint Conference on Artificial Intelligence IJCAI-05, Edinburgh, UK, 2005. Morgan-Kaufmann Publishers.
You can get a copy of this
here, or a local copy
here. This is part of a series of papers from the authors and
others on standard and non-standard inferences on extensions of the EL
language.
Monday, March 19, 2007: The Semantic Web---XML, RDF, OWL, and Description Logic.
Where: Bossone 304, 4pm, 2k7/3/19
Topics: What are description logics? What kind of knowledge can they
express? What is the Semantic Web and why does it get so much attention?
How is this more than just XML?
Pizza & pop courtesy
GICL@Drexel.
• D. Nardi, R. J. Brachman. "An Introduction to Description Logics."
In the Description Logic Handbook (Chap. 1), edited by F. Baader, D. Calvanese, D.L. McGuinness, D. Nardi, P.F. Patel-Schneider, Cambridge Univ. Press, 2002, pgs. 5--44.
You can borrow a copy of this chapter
here
(or local copy
here). This book is
again a mostly excellent one, there are several very good chapters in
this. The chapter's long, so don't be afraid to skim the background and
reasoning parts. For now we're mostly interested in description logic
language essentials.
• F. Manola and E. Miller. "The RDF Primer." W3C, 2004.
This is available on the web
here, and a cached PDF copy is available
here.
We are only interested for now in Sections 1, 2, and 3. Don't be afraid
to skim and don't be scared off by the relatively dense looking examples.
• M. K. Smith, C. Welty, and D. L. McGuiness. "OWL Web Ontology Language Guide."
W3C, 2004.
This is available on the web
here, and a cached PDF copy is available
here.
We are only interested in Section 3 for now.
Slides: Available
here; gives a basic KR example, transitioning from first order logic to description logic, and presenting RDF and OWL versions.
Monday, February 12, 2007: Introductions.
Where: Drexel's Bossone Building, Room 303, 4pm, 2k7/2/12.
Topics: What is knowledge representation & is it useful for anything?
What does "formal" mean? Possible review of first order logic. Pizza
& pop courtesy
GICL@Drexel.
• Brachman and Levesque, "Knowledge Representation and Reasoning."
Morgan Kaufman, 2004. Chapter 1: "Introduction."
You can "borrow" an (unfortunately annotated; sorry!) copy of Chapter 1
here. This book
is excellent however; it may be well worth your effort to acquire a
copy. A particular feature is that it explicitly
focuses on the tradeoff between expressiveness and
computability/tractability.
• Ray. "Interoperability Standards in the Semantic Web." Journal of
Computing and Information Science in Engineering. March 2002, volume 2. Pgs 65--69.
A paper motivating the need for rigorous ontology to enable correct,
flexible, automated software integration. Notably, this is from the
head of Manufacturing Systems Integration Division at NIST, which leads and
participates in a great many international standardization efforts
related to mechanical engineering software.
Available from ASME Journals, or cached
here.
• Any first order logic refresher.
Enderton, Russell and Norvig, Brachman and Levesque, and many others all have good primers.
Possible Future Topics
Ontology Engineering: Realistic, in-use ontologies, modelling principles, effectiveness
metrics.
Fragments of FOL & Computability/Tractability:
Interesting description logics, their relationships
to first order logic, and their computational properties.
The tradeoff and sharp divide between expressiveness and
tractability.
DL Reasoning: Topics: Structural algorithms, tableaus, instance stores.
Knowledge in Action: Situation calculus, action logics, PSL.
• Note related list:
readings in process modeling
and semantic integration
Reasoning about Knowledge: Modal logic, epistemic logics, protocol verification.
Ontology Integration
Probability and Uncertainty
Audience
This reading group is intended for computer science MS and PhD
graduate students, but is open to anyone with an interest.
Undergraduates, academics, and practitioners are more than welcome to
participate. Although details
and more difficult material will not be shied away from, the group
also aims to include those new to
knowledge representation. An explicit goal is to bring
up to speed new students and researchers with an interest in knowledge
representation.
If your advisor has ever told you "Yo, you should
look into this Semantic Web stuff. I hear it's smokin'." and
rambled on about XML, logic, and the Internet while you
just looked at them blankly, you should join this group. You
should also come if you're ready to debate the comparative merits of
$\mathcal{FL}^_(\sqcap, ^-)$ and $\mathcal{FL}^-(\sqcap)$.
Scheme
The plan is to meet roughly every 3 weeks. There will be a loose
progression from introductory to more detailed material which will
periodically reboot in order to return to the basics from a new
direction and simultaneously provide a good place for new people to
jump in.
For each meeting, one or a few papers will be chosen for
participants to read. At the meeting, a previously designated
attendee will briefly summarize the paper and lead the discussion, and
another will take notes. In addition to having read the chosen items,
participants should also think about potential future readings or
directions to explore. At the close of each meeting, readings,
discussion leaders, and dates will be chosen for the subsequent
meeting.
Much of the style, format, and conduct of the sessions will
be up to the discussion leaders, the attendees, and their wishes/needs.
Please let Joe know if you'll be attending a session so he may coordinate food appropriately.
Where Is That Exactly?
Drexel, in its infinite wisdom, took down the linkable
campus map in favor of a useless Flash animation. The Bossone
Building is the combination of the new research and the old
Commonwealth buildings, at 31st and Market, directly north from the
Main building. It's outlined here. Room 303 is on the east side of the third floor (roughly near the markers in the figure).