Database Reference
In-Depth Information
the result set in a user-friendly form based on domain knowledge. The user then navigates through this
hierarchy structure in a top-down fashion, exploring the summaries of interest while ignoring the rest.
The remaining of this chapter is organized as follows. In the first part of the chapter, we first survey
techniques that have been proposed in the literature to provide users with effective and efficient ways to
access relational databases, and then propose a categorization of these techniques based on the problem
that they are supposed to address. In the second part of the chapter, we present the
ESRA
algorithm and
the query answering system that supports
ESRA
-based summary hierarchies.
2. ADVANCED QUERY PROCESSING IN DATABASES
As internet becomes ubiquitous, many people are searching their favorite houses, cars, movies, cam-
eras, restaurants, and so on over the Web. Most Web sites use databases to store their data and provide
SQL-based query interfaces for users to interact with databases (Bergman, 2001; Chang, He, Li, Patel
& Zhang, 2004). Database systems provide well-maintained and high-quality structured data. However,
unlike Web search engines that take a few keywords, look up the index and provide a listing of best-
matching Web pages, they expect users to know the name of the relation to query, the field to look in,
and at times even the field type (Nambiar, 2005). Moreover, database query processing models have
always assumed that the user knows what she/he wants and is able to formulate a query that accurately
expresses her/his needs. Therefore, most database systems have always used a boolean model of query
processing where there is a set of answer tuples that exactly satisfy all the constraints of the query and
thus are equally relevant to the query.
While extremely useful for third-party applications and expert users, the above retrieval model is
inadequate for lay users who cannot articulate the perfect query for their needs - either their queries are
very specific, resulting in no (or too few) answers, or are very broad, resulting in too many answers.
Hence, to obtain a satisfactory answer from a database, users must reformulate their queries a number
of times before they can obtain a satisfactory answer. However, this process is frustrating, tedious and
time-consuming.
In following subsections, we review and discuss several research efforts that have attempted to handle
the dual issues of empty and many answers. Although the list of approaches described below is not ex-
haustive, it provides a representative list of some commonly used approaches. In Section 2.1, we review
a number of approaches for handling the
empty-answer
problem, that is, the problem of not being able
to provide the user with any data fitting her/his query. Section 2.2 presents some works addressing the
many-answers
problem, i.e., the situation where the user query results in overabundant answers. Then,
in Section 2.3 we give an overview of flexible and user-friendly querying techniques, the main objec-
tive of which is to provide intelligent interfaces to access databases in a more human-oriented fashion
and hence diminish the risk of both empty and many answers. A discussion is presented in Section 2.4.
2.1 Handling the Empty-Answer Problem
Most probably, one has encountered answers like 'no houses, hotels, vehicles, flights, etc. could be found
that matched your criteria; please try again with different choices'. The case of repeatedly receiving
empty query result turns out to be extremely disappointing to the user, and it is even more harmful for
the e-merchant.
