Data Mining in business intelligence applications

XXXX (I don't know why people on cnblogs are so resistant to xxxx, haha ......) Launched" Experience the SQL Server 2005 "Activity, of course, some SQL Server 2005 Article Translated into Chinese. Add it to favorites!

Article 2: Data Mining in business intelligence applications

Smart Application Platform

Over the past two decades, with the rapid economic development, organizations have collected a large amount of commercial data. However, having a large amount of data does not mean having a wealth of business information. Therefore, business intelligence must go to the next step-data mining. Data mining can help you discover potential effective modes while browsing massive amounts of data, and apply these models to business operations.
Microsoft SQL Server 2005 analytics service creates a new data mining component-an easy-to-use, easy-to-scale, easy-to-access, and flexible platform. Organizations that have never considered using data mining before are undoubtedly a very acceptable solution.
The architecture of enterprise-level products is closely integrated with the business intelligence functions of the SQL Server product family, with rich tools, APIs andAlgorithmAll of this allows you to create new business intelligence applications based on SQL Server.ProgramIt provides custom data-driven solutions for various business problems to increase productivity, increase profits, and reduce expenditures. For example:

• Basket Analysis.Analyze the sales results to determine whether the product should be bundled for sale. Use the results of the application to create an online guidance system, or decide how to bundle sales to maximize profits.
• Customer loss analysis.When you create a report, the customer that may cancel the service is displayed.
• Market analysis.Market segments are automatically grouped by customers, and trend analysis is performed based on the results to design market activities.
• Prediction.Predict sales volume and inventory and learn the association between them.
• Data browsing.Models discovered by data mining algorithms can help you better understand your customers. It can compare the differences between high-value customers and low-value customers, or analyze the differences between customers of different brands that love the same product.
• Web site analysis.Analyzes Website user behaviors and summarizes similar usage modes.
• Marketing activity analysis.The money is spent on cutting edge, accurate positioning of effective users, so that we can maximize the effectiveness of every cent of market funds.
• Data quality.When the data is loaded into the data warehouse, check for possible data loss or abnormal data.
• Text Analysis.Analyze the feedback and find the common topics or trends related to your customers or employees.

Integration

SQL Server Data Mining is a member of Microsoft's Business Intelligence Technology family. The addition of Data Mining makes them even more powerful.

• The SQL Server 2005 integration service.Based on the predictive analysis results obtained by data mining algorithms, we can mark exceptions, data categories, and fill in lost data in SQL Server 2005 integration service to create a more powerful data pipeline.
• SQL Server 2005 Analysis Service (OLAP ).Add the data mining dimension and segment data based on the hidden modes in the data to build a more extensive unified space model (UDM ).
• SQL Server 2005 report service.Based on Data Mining and query, you can create more intelligent and insightful reports to present the correct information or select the right Receiving users for reports.

Easy to use

With the built-in data mining wizard and designer, you only need a few clicks to build a very complex and precise model.
Due to its close integration with Microsoft Visual Studio, the SQL server data mining toolkit allows you to view and operate data while designing and editing models. SQL Server Data Mining provides more than a dozen interactive visual interfaces to help you understand the data mining algorithm discovery modes. In addition, before you determine the deployment, you can also refer to the elevation degree and profit chart to compare the differences between different models.

Simplicity and simplicity

In terms of Data Mining Model Application, SQL Server opens a new chapter. The emergence of Data Mining extensions for SQL (DMX) makes it easy for developers and DBAs to create data mining-related applications, in the past, they may only be familiar with creating database-related applications. Now, statements that use the data mining model for prediction are like join queries in SQL queries. That is, the first contact, using tools they are already familiar with and existing knowledge can also be used to operate data mining technology.
Algorithm
Through the joint efforts of Microsoft Research Institute and the SQL server product development team, SQL Server Data Mining contains a variety of effective data mining algorithms, including decision tree, regression tree, time series, aggregation, sequence clustering, association rules, Naive Bayes, neural network and Text Mining.

Scalability

SQL Server Data Mining is completely scalable. Through seamless integration of. Net stored procedures with algorithms and viewer plug-ins, you can take advantage of all the platform features and integration advantages. Using SQL Server Data Mining as your platform means that you will not be limited by the functions of the data mining system, because it can be expanded according to your needs.

Architecture

Applying data mining to organizations of different sizes brings new challenges. Deployment, stability, manageability, and security are all very important factors. SQL Server Data Mining is part of the Analysis Service. It provides all the enterprise-level servers you want.

• Deployment.SQL Server Data Mining is based on the client-server architecture, which allows you to access the model through LAN, Wan, or Internet. Standard APIs, such as ole db, adomd. net, and even XML/a, can provide you with model access functions, regardless of the client location and platform. Once processed, the model is always available, even during its update process.
• Stability.At the beginning of the design, SQL Server Data Mining adopts a parallel architecture to meet enterprise-level data requirements and thousands of concurrent user requests. It can respond to millions of queries every day.
• Easy to manage.SQL Server Data Mining is integrated into the new SQL management studio. It provides an all-in-one solution to help you manage all the attributes of the SQL Server family.
• Security.SQL Server Data Mining provides role-based and fine-grained security functions to ensure that all your business information is securely protected.

More information

SQL Server data mining is an integral part of SQL Server and Windows server system. Windows Server System is a comprehensive and integrated server infrastructure that helps simplify the development, deployment, and operation of flexible business solutions.

For more information about the SQL Server 2005 analytics service, visit the http://www.microsoft.com/ SQL /2005