Data mining

Authors

  • Tadeusz Morzy Instytut Informatyki, Politechnika Poznańska

Keywords:

data mining, data analysis, evolution of information technology, association analysis, classification, clustering, Web mining

Abstract

Recent advances in data capture, data transmission and data storage technologies have resulted in a growing gap between more powerful database systems and users' ability to understand and effectively analyze the information collected. Many companies and organizations gather gigabytes or terabytes of business transactions, scientific data, web logs, satellite pictures, textreports, which are simply too large and too complex to support a decision making process. Traditional database and data warehouse querying models are not sufficient to extract trends, similarities and correlations hidden in very large databases. The value of the existing databases and data warehouses can be significantly enhanced with help of data mining. Data mining is a new research area which aims at nontrivial extraction of implicit, previously unknown and potentially useful information from large databases and data warehouses. Data mining, also referred to as database mining or knowledge discovery in databases, can help answer business questions that were too time consuming to resolve with traditional data processing techniques. The process of mining the data can be perceived as a new way of querying – with questions such  as ”which clients are likely to respond to our next promotional mailing, and why?”. The aim of this paper is to present an overall picture of the data mining field as well as presents briefly few data mining methods. Finally, we summarize the concepts presented in the paper and discuss some problems related with data mining technology.

Issue

Section

Articles