Data mining - what is it ?!

Data mining is used to identify patterns in databases using software tools. This is done in contrast to the targeted evaluation of databases with regard to specific questions (»How high was the turnover for product x in January 2001?") Or the testing of hypotheses ("Did a deviation in turnover in store y lead to a reduction in turnover for product x?" ).

The data mining procedure, also known as non-directional data analysis, should make it possible to extract previously undiscovered relationships from large databases in a largely automated manner and to disclose them to the company ("Whenever product z is advertised on television, the sales of product x fall within two Weeks "). This is achieved through software tools that make various data mining methods available and thus expand existing data warehouse and OLAP (On-Line Analytical Processing) solutions.

The origin of the methods of data mining lies primarily in statistics, neural networks and machine learning. Procedures and algorithms developed in these areas are used for the three main tasks of data mining: the segmentation, classification and association of data.
During segmentation, groups are formed in the database whose group members have properties that are as similar as possible, but are as different as possible from the members of other groups. For example, groups of customers with similar needs can be identified.
The classification assigns new data records to an existing class. For example, credit applications to banks are assigned to one of the two classes “repayment probable” or “repayment unlikely” immediately after all relevant data has been entered, so that a prediction of the current case with a probability is made from the experience of historical cases.

The association represents dependencies between elements. Particularly when analyzing the basket of goods, for example at retailers, statements can be made such as: "If product A is bought, product B is always bought with 75%".

Data mining occupies the central position in a knowledge acquisition process that begins with data preparation, continues with the application of data mining methods and finally leads to the visualization and interpretation of the results found. The generation of valid and useful statements by data mining methods usually requires a trained user with statistical knowledge.

Was the explanation to "Data mining - what is it ?!"Helpful? Rate now:

More explanations about controlling