| Publisher | University of Waterloo | ||
|---|---|---|---|
| Format | 196.0KB WORD | Date added | 20 Aug 1999 |
| Topics | Data Mining / Analysis | ||
| Downloads | 43 | ||
Clustering is one of the important techniques in Data Mining. The objective of clustering is to group objects into clusters such that the objects within a cluster are more similar to each other than objects in different clusters. The density-based clustering algorithm DBSCAN is applicable to either a metric space or a spatial space. In a metric space, the similarity between two objects is defined by a distance function, e.g., the Euclidean distance, which satisfies the triangular inequality. Distance calculation is computationally very expensive in metric spaces and many algorithms are proposed to reduce the distance calculations due to the triangular inequality, but none of them benefits from the gradual addition of new dimensions when clustering. While comparing several clustering algorithms, we noticed that we often begin clustering on a small number of attributes, e.g., two. If the result is partially satisfying, we continue clustering with a higher number of attributes, sometimes up to a large number, e.g., ten. In this paper, we propose gradual clustering algorithms, which progressively clusters objects from a small number to a possibly large one.
Did you find this white paper useful?
25 out of 50 users found this white paper useful
Related white papers
Sybase Business Intelligence and Data Warehousing Solutions for Sybase IQ
This white paper looks at new methods of approaching data warehouses and the technology that supports these data warehouses. Particular focus is made to alternative approaches to the organization...
Common MDX Mistakes and Solutions in Microsoft SQL Server 2000 Analysis Services
This Microsoft Support webcast will discuss solutions to the ten most common mistakes that developers and database administrators make when they use Multidimensional Expressions language (MDX) to create queries that...
BP's Carson Unit Refines Contractor Management With WorkTech's Contractor Cost Tracking System (CCTS)
BP is a leading oil, gas, and energy business with operations in 100 countries around the world. The goals of the company were alignment of processes across all business areas,...
TechNet Webcast: ShopTalk - Measuring IT Health with an IT Scorecard
Business intelligence is as important in the IT department as it is in the Finance department. Making good IT business decisions requires aligning with corporate strategy, measurable performance goals and...
Eleven Steps to Success in Data Warehousing
Navigates 11 key steps to implementing a data warehousing solution that wins new customers, develops new products, and reduces costs. With the average cost of a system valued at $1.8...
Data Mining and Customer Relationships by Kurt Thearling
The way in which companies interact with their customers has changed dramatically over the past few years. A customer's continuing business is no longer guaranteed. As a result, companies have...
Where’s My Customer? Spatial Modeling for Promotion Distribution
HYPERparallel, a leading data mining company, offers a set of data mining applications targeted to customer relationship marketing (CRM). This recipe integrates three technologies: HYPERparallel’s //Discovery (pronounced HYPERparallel Discovery) suite...
