Data mining is also known as data exploration and data mining. It is a method of analyzing large amounts of information stored in an enterprise through mathematical patterns to identify different customers or market segments and analyze consumer preferences and behavior. It is a step in database knowledge discovery. Data mining generally refers to the process of automatically searching for information hidden in a large amount of data with special relationships. There are 3 main steps: data preparation, law finding and law representation. The tasks of data mining are correlation analysis, cluster analysis, classification analysis, anomaly analysis, idiosyncratic group analysis and evolutionary analysis. Data mining is usually associated with computer science and achieves these goals through many methods such as statistics, online analytical processing, intelligence retrieval, machine learning, expert systems (relying on past rules of thumb) and pattern recognition.

It is a discipline that uses data to identify and solve problems.

It is usually achieved by exploring, processing, analyzing, or modeling data.

We can see that data mining has the following characteristics:

Based on a large amount of data: it is not that mining is not possible on a small amount of data, in fact, most of the algorithms of data mining can be run on a small amount of data and get results. However, on the one hand, too small a volume of data can be analyzed manually to summarize the rules, on the other hand, small data volumes often do not reflect the general characteristics of the real world.

Non-triviality: The so-called non-triviality refers to the fact that the mined knowledge should be non-trivial, and must not be similar to what a famous sports commentator said, "After my calculation, I found an interesting phenomenon, by the end of this game, the number of goals scored and conceded in this World Cup is the same. What a coincidence!" That kind of knowledge. This seems like a no-brainer, but it's a mistake that many novice data miners who don't have business knowledge often make.

Implicit: Data mining is about discovering knowledge that lies deep within the data, not information that surfaces directly on the surface. Commonly used BI tools, such as reporting and OLAP, are perfectly capable of allowing users to find out this information.

Novelty: The knowledge that is mined should be previously unknown, otherwise it is nothing more than validation of the experience of business experts. Only brand new knowledge can help organizations gain further insights.

Value: The results of mining must be able to bring direct or indirect benefits to the enterprise. Some people say that data mining is just a "dragon-slaying technique", which looks fabulous but is of no use. This is just a misunderstanding. It is undeniable that in some data mining projects, the lack of clear business objectives, inadequate data quality, people's resistance to changing business processes, or the inexperience of the diggers may lead to poor results or even no results at all. But there are plenty of success stories that prove that data mining can be a powerful tool for improving efficiency

.