Mining the (data) bank

Data mining is the process of discovering meaningful new correlations, patterns, and trends by sifting through large amounts of data stored in repositories, using pattern recognition as well as statistical and mathematical techniques. Many tasks can be accomplished with data mining: classification,...

Full description

Saved in:
Bibliographic Details
Published in:IEEE potentials 2005-10, Vol.24 (4), p.40-42
Main Author: Ataee, P.
Format: Article
Language:English
Subjects:
Banking
Data analysis
Data mining
Equations
MATLAB
Pattern recognition
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Data mining is the process of discovering meaningful new correlations, patterns, and trends by sifting through large amounts of data stored in repositories, using pattern recognition as well as statistical and mathematical techniques. Many tasks can be accomplished with data mining: classification, forecasting, clustering, deviation detection, description, visualization, and link analysis. In this article the unsupervised method was used. The analysis was done by different methods such as windowing, time series, and clustering, which were subsets of the segmentation family. A pattern can be a complicated nonlinear relationship between two variables. Each data-mining technique that could detect this pattern was named as the pattern recognition method. To discover knowledge from raw data, the analyzer program used mathematical functions such as mean, sum, smooth, and normalizes. To attain a more profitable program, the computer program was designed to be applicable to other similar data banks.
ISSN:0278-6648
1558-1772
DOI:10.1109/MP.2005.1549758