The BibTip people wrote an article in DLIB a while back which explained
recommender systems in general and their approach in particular:-
http://www.dlib.org/dlib/may08/monnich/05monnich.html
I think Dave got their approach right. Here is what they say:-
'The BibTip recommendation system is based on the behavioral patterns of
users interacting with a library catalog. This so called "implicit"
recommendation service rests upon the observation of user patterns and the
statistical evaluation of the usage data. All the data stored and processed
are anonymous (identification numbers and session IDs). On a technical level
the BibTip architecture may be seen as an agent architecture involving three
software agents: the OPAC Observation Agent, the Aggregation Agent, and the
Recommendation Agent. The OPAC Observation Agent observes the selection of
titles within defined OPAC sessions. These data are transferred to the
Aggregation Agent, which then does computations on the statistical material
to arrive at a list of recommendations. Lastly, the Recommendation Agent
presents the list of recommendations to the user'
UEL are implementing BibTip at the moment--I don't know any other UK
institutions doing so. I *think* ExLibris bX works differently. I think they
used the algorithm from the Mesur project and it's based on the data from
resolvers about which items (journals) are 'used' rather than just searches.
In terms of *aggregating* data (one of the common threads around this
debate) the DLib BibTip article said:
'BibTip, like other recommendation services, works better the larger the
database is, since the statistical data are meaningful only if they are
calculated on a sufficiently large number of transactions. Therefore, an
observation period for the collecting and analyzing of data is necessary
before the first recommendations can be displayed'
And they continue:
'One way to accelerate the build-up of sufficient data to generate
recommendations is to pool the statistical data of several libraries. This
might be particularly feasible for libraries with a common interest in their
choice of titles and similar clientele (and comparable user behavior), for
example, public libraries..'
It might be something data services like EDINA and/or MIMAS would be
interested in?....i.e. aggregating the data from multiple VuFind (or other
services...).
Ken
Tel +44 (0)7788 727 845. Email: [log in to unmask]
www.kenchadconsulting.com
Skype: kenchadconsulting Twitter: @KenChad
Open Library Systems Specifications: http://libtechrfp.wikispaces.com
-----Original Message-----
From: Open source software in libraries [mailto:[log in to unmask]] On
Behalf Of David Friggens
Sent: 22 March 2011 01:31
To: [log in to unmask]
Subject: Re: [LIS-OSS] VuFind and activity data
> but I know of a few European (German, I think?) libraries who are
capturing OPAC usage data.
I think you're referring to BibTip (bibtip.com). They make a point of
using search rather than circulation data.
> Once I get a bit of spare time, I'm hoping to compare recommendations
using the session ID + item ID data we've collected (just over 1 million
instances in the last 12 months) to the recommendations we're currently
using, which are based on circ data analysis.
I've been thinking similar things, but "spare" time hasn't been very
forthcoming. :-) Would be very interested to see what you come up
with.
Cheers
David
|