数据挖掘和BI（Data mining and BI）.docVIP

数据挖掘和BI（Data mining and BI） Data warehouse refers to the regular data obtained from various data sources by means of ETL (extraction, transformation, loading), which is often the way of latitude tables and fact tables; Data mining refers to the discovery of knowledge by clustering, regression and neural networks in the existing data of the data warehouse, and draws conclusions to support decision making. OLAP - OnLine Analysis Process (online analysis process). Generally, there are several ways of implementing online analysis through virtual cubes (Cube). Many BI products offer OLAP tools (such as BO, Congos, etc.). A virtual cube (Cube) is pre created from a database and can be interpreted as a statistical result set containing different dimensions and indexes. For example, a Cube includes the following dimensions: region (province, city, county, district), time (year, month, day), store name, index: sales volume, number of products sold. Based on the generated Cube, users can make a comparative analysis between these different dimensions and metrics. OLAP is basically such a process. It can be said that OLAP is a manual analysis of aggregated data, mainly analyzing trends, situations, distributions, etc.. Data mining is the process of exploring and analyzing large amounts of data by means of automation or semi automation. The purpose of data mining is to discover meaningful patterns or regularities. Its real meaning should be data mining from from data (information). Finally, make a comparison: 1., Data, Mining, and OLAP are quite different. The essence of the OLAP analysis process is a deductive reasoning process, where OLAP analysis establishes a series of hypotheses, and then uses OLAP to prove or disprove these assumptions to arrive at their own conclusions. DM is essentially an inductive process, and DM is not used to verify the correctness of a hypothetical model, but rather to look for a model in its own database. 2. data mining and OLAP have some complemen