Mining Web using Hyper Induced Topic Search Algorithm

There is a lot of data on the Web, some in databases, and some in files or other data sources. The databases may be semi structured or they may be relational, object, or multimedia databases. These databases have to be mined so that useful information is extracted. While we could use many of the data mining techniques to mine the Web databases, the challenge is to locate the databases on the Web. Furthermore, the databases may not be in the format that we need for mining the data. We may need mediators to mediate between the data miners and the databases on the Web. This paper presents the important concepts of the databases on the Web and how these databases have to be mined to extract patterns and trends