Data Mining: Discovery of a valuable treasure

By this moment in time, you've probably heard something about data mining – today’s hottest catchword. To use a plain analogy, it's like finding a treasure buried deeply in an island. In this case, the treasure is that particular piece of intelligence your business requires and the island is the huge data warehouse you've put together over a long period of time.

Data mining is, indeed, a treasure seeker. Even though it is a newly introduced technology, most companies use its great potential to help their businesses focus on the most vital information. In the article An Introduction to Data Mining: Discovering Hidden Value in your Data Warehouse, Kurt Thearling defined data mining as the extraction of hidden predictive information from large databases. Basically, large information is being break down into simpler and essential ones, simultaneously uncovering patterns and forecasting future trends.

The main objective of data mining is to discern previously unknown relationships among the data, especially when the data came from different databases. In order to do that, the following techniques are being applied.

o Artificial neural networks: usually called "neural network" (NN), a mathematical model that aims to replicate the structure or functional aspects of biological neural networks.

o Decision trees: Tree diagram that serves as a decision tool. Specific decision tree methods include Classification and Regression Trees (CART) and Chi Square Automatic Interaction Detection (CHAID).

o Genetic algorithms: Optimization techniques that use process such as genetic combination, mutation, and natural selection in a design based on the concepts of evolution.

o Nearest neighbour method: A technique that classifies each record in a dataset based on a combination of the classes of the k record(s) most similar to it in a historical dataset (where k ³ 1). Sometimes called the k-nearest neighbour technique.

o Rule induction: The extraction of useful if-then rules from data based on statistical significance.

Applying these, businesses can use relationships to develop new marketing campaigns or foresee how well a product will sell. For example, mining a supermarket database could reveal that particular items are procured together, such as beer and chips, cereal and milk, etc. It can reveal that customers with a common set of demographic characteristics will purchase same items, shop at similar times and regularity, be evenly brand loyal or disloyal, purchase related groups of items, or react to a particular type of promotion. These techniques are also used by the government to detect illegal or prohibited activities by individuals, associations, and other governments.


As data have grown in bulk and complexity, manual data analysis has gradually more been amplified with indirect, automatic data processing. Online Analytical Processing (OLAP), also part of Business Intelligence, is a model that allows complex analytical and impromptu queries with a quick execution time. It uses navigational database to look for specific data within the data warehouse for a short time. Reporting, visualization, and other analysis tools can also be applied in decision making and validate the impact of the possible plans.


Aside from choosing the right technology for data mining, it is also crucial to consider two important factors to achieve success in using it. These are:

“large, well-integrated data warehouse and a well-defined understanding of the business process within which data mining is to be applied.”


Since data mining can only bare patterns that are present in the data, the target data set should be broad enough to contain these patterns while being brief and direct enough to be mined in suitable time range. It’s also trouble-free to use data mining if you know the business process well. It’s simple to point out or query explicit data since it what’s needed are already been recognized.


Besides, data mining can also contribute in customer relationship management (CRM) applications. A company can focus its efforts on prospects that are forecasted to have a high possibility of responding to an offer rather than contacting each customer randomly through a call center or sending mail which can be more time consuming. To determined which channel and which offer an individual is most likely to respond to, more sophisticated techniques may be used. It can also optimize assets across operations – across all potential offers.

“Data mining seeks treasure.”

Yes, it does! It seeks relevant data which when turned into information creates and gives useful plans, patterns and future forecasts. These are used by the companies to generate industry-driven decisions in order to increase profitability and boost company performance.

This blog is another activity in Datamin class under Mr. Ramon Duremdes Jr. Data mining is also discussed in my classmatess'blog entries, the links of their names are found in the left corner of this page.:)