Unformatted text preview: learning, other AI ﬁelds can potentially contribute signiﬁcantly to
various aspects of the KDD process. We mention a few examples of these areas here:
Natural language presents signiﬁcant opportunities for mining in free-form text, especially for automated annotation and indexing
prior to classiﬁcation of text corpora. Limited
parsing capabilities can help substantially in
the task of deciding what an article refers to.
Hence, the spectrum from simple natural language processing all the way to language understanding can help substantially. Also, natural language processing can contribute
signiﬁcantly as an effective interface for stating hints to mining algorithms and visualizing and explaining knowledge derived by a
P lanning considers a complicated data
analysis process. It involves conducting complicated data-access and data-transformation
operations; applying preprocessing routines;
and, in some cases, paying attention to resource and data-access constraints. Typically,
data processing steps are expressed in terms of
desired postconditions and preconditions for
the application of certain routines, which
lends itself easily to representation as a planning problem. In addition, planning ability
can play an important role in automated
agents (see next item) to collect data samples
or conduct a search to obtain needed data sets.
Intelligent agents can be ﬁred off to collect necessary information from a variety of Articles sources. In addition, information agents can
be activated remotely over the network or
can trigger on the occurrence of a certain
event and start an analysis operation. Finally,
agents can help navigate and model the
World-Wide Web (Etzioni 1996), another area
growing in importance.
Uncertainty in AI includes issues for managing uncertainty, proper inference mechanisms in the presence of uncertainty, and the
reasoning about causality, all fundamental to
KDD theory and practice. In fact, the KDD-96
conference had a joint session with the UAI-96
conference this year (Horvitz and Jensen 1996).
Knowledge representation includes o ntologies, new concepts for representing, storing, and accessing knowledge. Also included
are schemes for representing knowledge and
allowing the use of prior human knowledge
about the underlying process by the KDD
These potential contributions of AI are but
a sampling; many others, including humancomputer interaction, knowledge-acquisition
techniques, and the study of mechanisms for
reasoning, have the opportunity to contribute to KDD.
In conclusion, we presented some deﬁnitions of basic notions in the KDD ﬁeld. Our
primary aim was to clarify the relation between knowledge discovery and data mining.
We provided an overview of the KDD process
and basic data-mining methods. Given the
broad spectrum of data-mining methods and
algorithms, our overview is inevitably limited in scope: There are many data-mining
techniques, particularly specialized methods
for particular types of data and domain. Although various algorithms and applications
might appear quite different on the surface,
it is not unc...
View Full Document
- Spring '14
- Data Mining, KDD