One area currently lacking in IR services, and it
will become more apparent as content volumes
grow, is current awareness, and that will require
a large degree of filtering. It seems inevitable
that some form of classification will be
necessary for this, since what is needed is some
a priori structuring of content in accordance
with known broad interests, in contrast to search
which allows users to apply a posteriori
customisable filtering and therefore serves a
complementary but not equivalent function. As
Phil says, ArXiv has a subject classification,
and this serves its important alerting service for new content.
How this will be done across IRs is another
matter. Ian Stuart suggests the technology is
lacking, and we could wring our hands over this
and then a developer somewhere will change the
world in two days. Maybe CRIG could hold a
competition. Almost certainly it won't be based
on some central cataloguing based approach, but
perhaps a series of distributed filters
configured at will through a Yahoo Pipes type arrangement.
One approach, adopted in the JISC Gold Dust
project (http://www.hull.ac.uk/golddust/) is to
try and learn more about individual users and the
type of materials they search for and use -
developing personal profiles by monitoring such
behaviour - and apply this as a priori filtering.
IRs are one category of content provider in the Gold Dust framework.
Steve Hitchcock
IAM Group, School of Electronics and Computer Science
University of Southampton, SO17 1BJ, UK
Email: [log in to unmask]
Tel: +44 (0)23 8059 7698 Fax: +44 (0)23 8059 2865
At 09:24 25/06/2008, Phil Barker wrote:
>Arthur Sale wrote:
>>
>>· Nobody uses any subject
>>classification for much at all, anywhere in the
>>world. They are not worth spending much time
>>thinking about, unless you have a government breathing over your shoulder.
>Well that's a shame. We know that academics are
>discipline focused (and students too, to some
>extent). Academic communities are discipline
>focused. So it would seem that services that
>aim to meet the needs of these communities
>should at least have discipline aspect to them
>(but I'm not saying that deposit isn't an
>institutional issue). So how are you going to
>link institutional repositories to discipline
>focussed services without knowing which items relate to which discipline.
>
>BTW, arxive and PubMed seem to make use of
>subject classification. As does Amazon --yes I
>know nobody browses Amazon by subject (but given
>an interface like <http://zoomii.com/> they
>might), but if you buy something from Amazon you
>can be sure that they will let you know about other books on the same subject.
>
>Phil.
>
>--
>Phil Barker Learning Technology Adviser
> ICBL, School of Mathematical and Computer Sciences
> Mountbatten Building, Heriot-Watt University, Edinburgh, EH14 4AS
> Tel: 0131 451 3278 Fax: 0131 451 3327
> Web: http://www.icbl.hw.ac.uk/~philb/
>
>
>
>--
>Heriot-Watt University is a Scottish charity
>registered under charity number SC000278.
|