Subject: Learning Objects: Who will assemble it?
From: Peter Brusilovsky (plb@cs.cmu.edu)
Date: Sun 20 Feb 2000 - 22:35:26 MET
Date: Sun, 20 Feb 2000 16:35:26 -0500 From: Peter Brusilovsky <plb@cs.cmu.edu> Subject: Learning Objects: Who will assemble it?
List address to send message to everyone: ifets-discuss@LISTSERV.READADP.COM
Details of current discussion: http://ifets.ieee.org/discussions/discuss.html
---------------------------------------------------------------------------
An interesting issue is who is going to use learning object to
generate a new course. So far, most of the practical projects that
has an emphasize on using tagged learning objects (like IDEALS/MTS or
ARIADNE) implied that the user is a course author. The author is
working on a kind of assembly like retrieving relevant objects and
creating a course. So, the current approach to tagging (LOM) and to
developing a framework is based on this paradigm.
What is really interesting that this approach has finally got very
close to a much an older approach known as adaptive curriculum
sequencing and developed in the area of intelligent tutoring systems.
First systems by Barr et all (BIP) and Koffman (GCAI) were developed
in 1970-ies. Now several dozens of systems and approaches are known.
A typical ITS with curriculum sequencing has a knowledge base of
educational objects (i.e., a database of objects tagged with some
metadata that transform data into knowledge) and can can use it to
generate an individualized course for every single learner. The
generation can happen on the fly (though it requires an ITS that
could maintain a student model) or could be done in one shot before
the course start like in DCG of Julita Vassileva (then the course is
simply static and can be delivered by any CAI/IMS system). A note to
Margaret Martinez: classic sequencers take into account only current
state of student's knowledge represented by an overlay student model,
but some modern sequencing and adaptive hypermedia system also adapt
to "learning style" and other traits.
The reason I am talking about these two classes is that the answers
to a number of question posted into this forum depend on who
(granularity, interoperability, etc.) depend on who is doing the
sequencing. As long as all tagging is for a human author, we still
can be flexible with granularity and single vocabulary and precise
tagging. After all - metadata is just for a course author to simplify
finding the relevant material. It provides for flexibility. A human
still can check the material behind the set of metadata to see how it
fits the course. However, if we are producing metadata for a computer
sequencer - we will need quite a precise tagging using very
elaborated standard taxonomies. We are quite ready for the first mode
at least on the level of single organizations, but not yet ready to
the second mode. I guess, we will bridge the gap by combining a work
of a human course author and an intelligent computer assistant. At
the moment the only thing that such assistant could do is to select
relevant questions from pools for a custom quiz, but we can expect
that with maturity of metadata and sequencing technologies, bigger
chunks of author's routine works will be supported by assistants.
Peter
--------------------------------------------------------------
Carnegie Technology Education | HCII, School of Computer Science
4615 Forbes Avenue | Carnegie Mellon University
Pittsburgh, PA 15213, USA | E-mail: plb@cs.cmu.edu
Phone: (412) 268-3537 | Fax: (412) 268-3505
WWW: http://www.contrib.andrew.cmu.edu/~plb/home.html
---------------------------------------------------------
Forum website: http://ifets.ieee.org/
Forum's contact person: kinshuk@massey.ac.nz
Info on Join/Leave List: http://ifets.ieee.org/maillist.html
---------------------------------------------------------
This archive was generated by hypermail 2a24 : Sun 20 Feb 2000 - 22:59:14 MET