Basically, there are already several data mining methods today that are developed some are already running and other are on its way to be implemented, but in general the process are interconnected.
A data mining process has elements such as clustering, association, sequential pattern, classification and producing results that leads can affect decisions. In this, we will examine those methods in sections.
Data Clustering Technique
A data clustering is one of the data methods that execute a meaning process or clustering objects that have inter-connected characteristics by simply using an automatic technique. Clustering methods defines these classes and separates objects based on class criteria.
While classification method, the object are then assigned to a predefined special classes. So to make this concept more clearly, we will use a book management as example (this is a library scenario).
Basically, a library has wide range of available books with that specializes on different topics. The challenge now is how readers locate those books on a certain topic without experiencing hustle.
By simply using a clustering method, the receptionist in the library can keep books that are similar or a shelf that has a label with meaningful identification. Therefore if the reader desires to get books in that category, they just simply go the shelf directly instead of moving around the library looking for something out of nowhere.
Data Association Technique
This is the best known methods in data mining. In the association process when a certain pattern is determined based on relationship on its items connected to transactions. This method is also known as data association. The technique used in the market analysis in order to identify products that community wants.
Most retailers are using this method to identify buying habits of the customers. This is done by looking at the data chart so they can develop the product that accumulates high sale.
Sequential Patterns of Data Technique
A pattern sequential analysis is among the several data mining methods that request information to be discovered by looking at patterns in events, trends in the transaction over business.
When it comes to sales, the historical data will be evaluated so that business can determine products that clients usually buy in a year period. With the information, business can effectively recommend its customers with better deals based from the historical purchase.
Data Classification Technique
This technique is one of the classic methods in data mining and this is based on machine learning. In classification, the data is then classified into predefined groups or classes. The method uses mathematical techniques like linear programming, decision trees, statistics and neural network.
Data Mining Prediction
This is a technique in data mining that identifies the connection between the independent variables, the relationship between those two, the independent and dependent variables.
For example, the technique in the analysis will be used to predict profit in the sales and considers sale as independent variable, the profit will be the dependent variable since it is dependent on the number of sales.