Data Mining

Use:

Development of methods for recording and storing data has led to rapid growth of volume of collected and analyzed information. "Data Mining"is a process of knowledge discovery in databases; it is used for automatic data analysis.

By knowledge we shall basically mean the following data:

• Previously unknown – knowledge that must be new, and not confirming previously obtained information;
• Nonstandard – knowledge that is impossible to see in a usual way (with the help of direct visual analysis of data or calculations of simple static characteristics;
• Practically useful – knowledge, valuable for a researcher or a consumer;
• Interpretable – knowledge that is easily presented visually and explained with the terms of topical area.

Knowledge is to describe new connections between characteristics, predict the values of some features on the basis of other ones, etc. Being applied this knowledge can be beneficial. Knowledge is to be understandable for a user. Otherwise there must be methods of postprocessing that make knowledge interpretable.

Functionality

The algorithms used in "Data Mining", Mining require large amounts of computation. Previously it prevented it from widespread practical use. However today's productivity growth of modern processors and advanced Smilart technologies have solved this problem. As a result, it is possible to conduct qualitative analysis of enormous amount of information in a reasonable period of time.

Problems of Business analysis are differently formulated, but the solution of most of them comes down to "Data Mining" tasks. . For instance, risk assessment is a task solution of regression or classification, market segmentation is clustering, and demand stimulation is association rules. In fact "Data Mining" Mining tasks are the elements which can be used for creating a solution for any real business task.

Smilart uses various methods and algorithms of "Data Mining" to achieve the objectives. As "Data Mining"has been developing at the intersection of disciplines such as mathematical analysis, statistics, information theory, machine learning, database theory, it is quite natural that the majority of the Smilart's own algorithms and "Data Mining" methods are created on the basis of various methods of these disciplines.