Preference SQL - Design, Implementation, Experiences
Wemer Kiel31ing
Institute of Computer Science
University of Augsburg
D-8613 5 Augsburg
Germany
kiessling@informatik.uni-augsburg.de
Gerhard K6stler
Intershop Communications
Intershop Tower
D-07740 Jena
Germany
g. ko e stl er@ inters hop. com
Abstract
Current search engines can hardly cope ade-
quately with fuzzy predicates defined by
complex preferences. The biggest problem of
search engines implemented with standard SQL
is that SQL does not directly understand the
notion of preferences. Preference SQL extends
SQL by a preference model based on strict
partial orders (presented in more detail in the
companion paper [Kie02]), where preference
queries behave like soft selection constraints.
Several built-in base preference types and the
powerful Pareto operator, combined with the
adherence to declarative SQL programming
style, guarantees great programming producti-
vity. The Preference SQL optimizer does an effi-
cient re-writing into standard SQL, including a
high-level implementation of the skyline opera-
tor for Pareto-optimal sets. This pre-processor
approach enables a seamless application
integration, making Preference SQL available on
all major SQL platforms. Several commercial
B2C portals are powered by Preference SQL. Its
benefits comprise cooperative query answering
and smart customer advice, leading to higher e-
customer satisfaction and shorter development
times of personalized search engines. We report
practical experiences ranging from m-commerce
and comparison shopping to a large-scale
performance test for a job portal.
Permission to copy without fee all or part of this material is granted
provided that the copies are not made or distributed for direct
commercial advantage, the VLDB copyright notice and the title of the
publication and its date appear, and notice is given that copying is by
permission of the Very Large Data Base Endowment. To copy
otherwise, or to republish, requires a fee and~or special permission from
the Endowment
Proceedings
of the 28th VLDB
Conference,
Hong Kong, China, 2002
1. Introduction
When searching for items to be purchased over the
Intemet, customer wishes and preferences are becoming
increasingly important. Just like in real shopping, a
customer has his or her personal criteria and tastes that
guide the search for the ideal product. These criteria can
be classified into two categories: Knock out criteria that
must be fulfilled versus soft criteria that should be
fulfilled as closely as possible. Going to a real shop the
customer expects to encounter a cooperative sales person,
who assists in finding the most suitable item compatible
with the stated hard and soft criteria. The same
expectation for good customer advice carries over to e-
shops in the Internet. However, the state of the art is far
away from this ideal situation. Current B2C or B2B e-
shops cannot cope adequately with real user preferences.
As a consequence, e-shopping sessions frequently leave
frustrated users behind. All too often no or no reasonable
answer is returned though one has tried hard filling out
query forms to match one's personal preferences closely.
Most probably, one has encountered answers before
sounding like "no hotels, vehicles, flights, etc. could be
found that matched your criteria; please try again with
different choices". The case of repeatedly receiving empty
query results turns out to be extremely disappointing to
the user, and it is even more harmful for the electronic
retailer (e-tailer). Studies by leading marketing research
companies like Forrester have revealed that it requires
only very few unsuccessful attempts that the user will
quit, and he or she will not login to this e-shop again for
quite some time.
Such an unpleasant system behavior has been
recognized by e-tailers too, leading to some more or less
ad hoc solutions. The simplest approach is to dictate the
user to radically deviate from his initial preferences by
leaving some entries in the query form unspecified. In
many cases instead of an empty answer the user is then
990

Get Proceedings 2002 VLDB Conference now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.