Data mining tools and technologies for competitive business advantage.

• Author: Chowdhury, Shamsul

1. INTRODUCTION

   Reliable data in a database or a data warehouse could be used for decision-making purposes by appropriately analyzing the data and making them more meaningful and useful. In other words data could be analyzed to find hidden patterns and foresee trends. The process is broadly being called data mining or knowledge discovery in databases. Over the years the focus of computational technology has shifted more from Program-centric to Data-centric to Business-centric. This shift has enabled organizations to learn more and more about their respective businesses and thereby excel. For example in a Customer Relationship Management (CRM) system, if we can learn more about the customers then we can offer better customer services. In order to survive and succeed in the tough and complex business world of today it is important to learn and utilize data mining tools and technologies for competitive business advantage. This work intends to explore if the fundamentals of data mining tools and technologies, particularly the uses of artificial neural networks are appropriate for use in business applications, for examples: i. detection and prevention of fraudulent business transactions/practices, and ii. trend analysis and forecasting in financial market based on predictive modeling and analysis.

2. DATA MINING

   Data mining is a creative process that uses statistical, mathematical, machine-learning and artificial intelligence techniques like artificial neural networks, case-based reasoning and others to extract and identify useful information and subsequent knowledge from large databases. Data mining is about connecting the past through learning to future action. Data mining usually starts with a hypothesis or an assumption and ultimately creates new information or knowledge for future use. The purpose of data mining is to convert the organizational problem into an information system solution for the organization.

   Typically the main goal of all data mining projects is to analyze and eventually predict future behavior in order to gain some sort of competitive advantage. For example, data mining techniques can be used to analyze a customer's purchasing habits which can provide valuable information about that customer and how to better serve the needs of that customer. Data mining is often referred to as knowledge discovery because its goal is to provide information/knowledge that can help improve the way an organization conducts its business.

   There are generally four steps to the data mining process:

1. defining a problem or objective,

2. gathering/preparing data,

3. building models, and

4. using the...