Data mining: definition, functions, methods and application examples


The rapid development of technology makes most businesses collect information through data. This data will later be useful to facilitate business processes, assist in decision-making and determine future business development strategies. This is the importance of implementing data mining or data mining for business.

It is the process or activity of gathering important information from data sources or data warehouses. You can discover this information by finding certain patterns using data mining. The purpose of this activity can be very diverse depending on your needs and interests. In the following article, we will explain what data mining is and what its functions, methods and examples are.

What is data mining?

Data mining or data mining is the process of gathering important data or information through large volumes of data (big data). The collection process often uses mathematical methods, statistics, to use artificial intelligence (AI) technology. You can also know data mining by other terms.

These include KDD (Knowledge Discovery in Databases), data analytics, knowledge mining, business intelligence, data mining, information mining, data archeology, etc. The data mining process itself includes several steps and techniques. Starting from data cleaning or data cleaning, data integration, data selection, data transformation to pattern evaluation to get information from data.

Functions of Data Mining There are many functions or benefits that you can get by implementing data mining. In general, its performance is divided into descriptive and predictive categories. However, apart from that, data mining also has other functions such as correlation, classification, clustering, prediction and sequencing.

Below is an explanation of each data mining function:

Descriptive or Descriptive

Descriptive refers to the function of understanding the data you are researching most. The goal of this process is to find patterns and features in the data.

Using this descriptive function, you can find specific patterns or patterns that are initially hidden in a data. That is, if there is a pattern that is repeated and has a value, it means that you can know the characteristics of the data.


Prediction is a function of a process that later reveals a particular pattern of data. You can find this pattern from several variables in the data. Once you find a pattern, you can use that pattern to estimate other variables whose values ​​are still unknown.

This is why predictive functions are considered equivalent to predictive analytics. You can also use prediction to estimate a specific variable that is not in the data.


Next is the function of the forum. This is a data mining function where you can identify the relationship (relationship) of any data. Both data from past and present.


Classification is the conclusion of several definitions of the characteristics in a group or group of data. An example is customer data (customer data) who stop using your product (customer churn) because they think competitors’ products provide them with greater benefits and customer value.


The next function is clustering or clustering. It is the process of identifying a group of products or goods that have certain characteristics.


Fifth, forecasting is a data prediction technique that is useful for obtaining an overview of the value of a data in the future (prediction). You can make this prediction by collecting a large amount of information. An example of the application of forecasting data is to predict the number of requests (demand) for seasonal products in certain seasons (seasonal marketing).


The last function is sequenced. It is the process of identifying any different relationships over a given period of time. An example of sequencing is data from customers who repeat the purchase of a particular product repeatedly throughout the customer’s life cycle.