Data mining, also known as knowledge discovery in database, can automatically extract or discover useful schema knowledge from various types of data in database, data warehouse and other databases.
In most cases, data mining is the first data extracted from the data warehouse to data mining in the library, of course, can also be extracted from other database or flat file, only the data warehouse of data cleaning is the same as the data cleaning and data mining, if the data has been cleaned in import the data warehouse is so when performing data mining does not need to clean up the data, All data can be directly applied to mining algorithms.
Data mining system can be independent of data warehouse system. However, in order to improve mining efficiency, data warehouse is generally used as the basis, and mining algorithm is used to mine potential patterns from the prepared data, so as to help decision-makers adjust market strategy, reduce risks and make correct decisions.
Prediction of the future is not to rely on what magic or book, but the use of scientific methods and advanced Smartbi data mining scientific platform, to analyze and mine hidden secrets in a large number of data, reveal the relationship between the data, the trend of the development of business research and judgment.
Traditional data analysis reveals the known and past data relationships, while data mining reveals the unknown and future data relationships. Traditional data analysis uses computer technology, but data mining not only uses computer technology, but also involves statistics, model algorithm and other technologies. Because the data mining discovery is the future information, so the most important is used: prediction! Predict the company’s future sales volume, forecast the future price of products, etc.
The Smartbi data mining platform provides one-stop data mining services, covering the whole life cycle of data preprocessing, machine learning algorithm application, model training, evaluation, deployment, and service release.
It is widely used in various fields, including enterprise operation, production control, market analysis, engineering design, urban planning and scientific exploration, etc., mining useful information and knowledge from a large amount of data, in order to better guide our work; The function has the following features:
Smart software Smartbi predictive analysis collects more than 50 kinds of data mining algorithm components, mainly including classification, clustering, association rules, regression and other rich algorithm components; Support for Java and Python algorithm extensions can be customized specifically for user scenarios.
Figure: Categories of machine learning algorithms
Business personnel can easily drag and drop the operation of components, carry out visual modeling, complete the construction of model flow, and can publish the model management. Comments can be added to each node to visually illustrate the modeling process.
Figure: Schematic diagram of machine learning algorithm package
Figure: Schematic diagram of visual modeling