The function of data mining

　　

Data mining makes proactive, knowledge-based decisions by predicting future trends and behaviors. The goal of data mining is to discover the hidden and meaningful knowledge from the database, which mainly has the following five kinds of functions.

1. Automatically predict trends and behaviors

Data mining automates the search for predictive information in large databases, and the problems that previously required a lot of manual analysis can now be quickly and directly drawn from the data itself. A typical example is the market forecast problem, where data mining uses past data on promotions to find the most rewarding users of future investments, and other predictable issues include predicting bankruptcies and identifying groups that are most likely to respond to a given event.

2. Correlation analysis

Data Association is a kind of important and discoverable knowledge that exists in the database. If there is a regularity between the values of two or more variables, it is called Association. Association can be divided into simple association, Timing Association, causal Association. The purpose of association analysis is to find out the hidden network of associations in the database. Sometimes it is not known that the associated functions of data in the database, even if known, are uncertain, so the rules generated by the association analysis are credible.

3. Clustering

The records in the database can be divided into a series of meaningful subsets, that is, clustering. Clustering enhances people's understanding of objective reality and is the precondition of conceptual description and deviation analysis. The clustering technique mainly includes the traditional pattern recognition method and the mathematical taxonomy. In the early 80, Mchalski put forward the concept of clustering technology 牞 Its main point is that in the Division of objects not only to consider the distance between objects, but also requires the division of the class has a certain connotation of the description, so as to avoid the traditional technology some one-sidedness.

4. Concept Description

The concept description is a description of the connotation of some kind of object, and summarizes the relevant characteristics of such objects. The concept description is divided into the characteristic description and the distinguishing description, the former describes the common characteristics of some kind of object, the latter describes the difference between different classes of objects. Generating a characteristic description of a class involves only the commonality of all objects in the class object. There are many ways to generate differentiated descriptions, such as decision tree methods, genetic algorithms, and so on.

5. Deviation detection

The data in the database often has some exception records, and it makes sense to detect these deviations from the database. Deviations include a lot of potential knowledge, such as abnormal instances in the classification, special cases that do not meet the rules, deviations from the predicted values of the observations and models, and changes in value over time. The basic method of deviation detection is to find a meaningful difference between the observations and the reference values. The difference between data mining and traditional analysis method

The essential difference between data mining and traditional data analysis (such as query, report, online application analysis) is that data mining is to excavate information and discover knowledge without explicit assumptions. The information obtained by data mining should be unknown, valid and practical three characters.

Previously unknown information refers to the information is not foreseen in advance, data mining is to find those who can not rely on the intuitive discovery of information or knowledge, or even contrary to the intuition of information or knowledge, the more information excavated out of the unexpected, The more valuable it can be. The most typical example of a business application is a chain store that discovers a startling link between a child's diaper and beer through data mining.