Databases Reference
In-Depth Information
Association rules have two important measurements: Support and Confi dence. Support
is an argument that decides whether the candidate is frequent or not. The frequent path
patterns are identifi ed by their support values. Confi dence is an argument that describes the
believable degree of association rules.
An example of an association rule is, “90% of transactions that contain beer also contain
diapers; 5% of all transactions contain both of these items.” Here 90% and 5% are called the
confi dence level and support level, respectively. The objective is to fi nd association rules
that satifsfy user-specifed minimum support and minimum confi dence threshold. A strong
association rule will have a large support and high confi dence level.
Web Mining
The World Wide Web serves as a huge, widely distributed, global information service
center for news, advertisements, consumer information, fi nancial management, education,
government, e-commerce, and many other information services. The web contains a rich and
dynamic collection of hyperlink information and web page access and usage information,
providing rich sources for data mining (Han & Kamber, 2001). Naturally, a combination of
the data mining and the World Wide Web are referred to as web mining.
Web mining
is broadly defi ned as the discovery and analysis of useful information
Web mining
from the World Wide Web. It describes the automatic search and retrieval of information
and resources available from online databases or web servers. In general, there are three
knowledge discovery domains pertaining to web mining, which are classifi ed into the
Figure 2: Taxonomy of web mining
