Добавил:
Upload Опубликованный материал нарушает ваши авторские права? Сообщите нам.
Вуз: Предмет: Файл:
SQL Server 2012 Tutorials - Analysis Services Data Mining.pdf
Скачиваний:
38
Добавлен:
26.03.2016
Размер:
1.41 Mб
Скачать

multiple outputs in a logistic regression model, it is easy to experiment with different independent variables and outcomes without having to create many separate models.

Remarks

The Data Mining Add-Ins for Excel 2007 provide logistic regression wizards that make it easy to answer complex questions, such as how many Level Two Operators would be required to improve service grade to a target level for a specific shift. The data mining add-ins are a free download, and include wizards that are based on the neural network or logistic regression algorithms. For more information, see the following links:

SQL Server 2005 Data Mining Add-Ins for Office 2007: Goal Seek and What If Scenario Analysis

SQL Server 2008 Data Mining Add-Ins for Office 2007: Goal Seek Scenario Analysis, What If Scenario Analysis, and Prediction Calculator

Conclusion

You have learned to create, customize, and interpret mining models that are based on the Microsoft Neural Network algorithm and the Microsoft Logistic Regression algorithm. These model types are sophisticated and permit almost infinite variety in analysis, and therefore can be complex and difficult to master.

However, these algorithms can iterate through many combinations of factors and automatically identify the strongest correlations, providing statistical support for insights that would be very difficult to discover through manual exploration of data using Transact-SQL or even PowerPivot.

See Also

Querying a Logistic Regression Model (Analysis Services - Data Mining) Microsoft Logistic Regression Algorithm

Microsoft Neural Network Algorithm

Querying a Neural Network Model (Analysis ServicesData Mining)

Creating and Querying Data Mining Models with DMX: Tutorials (Analysis Services - Data Mining)

After you have created a data mining solution by using Microsoft SQL Server Analysis Services, you can create queries against the data mining models to predict trends, retrieve patterns in the data, and measure the accuracy of the mining models.

The step-by-step tutorials in the following list will help you learn how to build and run data mining queries by using Analysis Services so that you can get the most from your data.

145

In this Section

Bike Buyer DMX Tutorial

This tutorial walks you through the creation of a new mining structure and mining models by using the Data Mining Extensions (DMX) language, and explains how to create DMX prediction queries.

Market Basket DMX Tutorial

This tutorial uses a typical market basket scenario, where you find associations between the products that customers purchase together. This tutorial also demonstrates how to use nested tables when you create a mining structure. You build and train a model based on this structure, and then create predictions using DMX.

Time Series Prediction DMX Tutorial

This tutorial creates a forecasting model to illustrate the use of the CREATE MODEL (DMX) statement. You then add related models and customize the behavior of each by changing the parameters of the Microsoft Time Series algorithm. Finally you create predictions and update the predictions with new data. The ability to update a time series while making predictions was added in SQL Server 2008.

Reference

Data Mining Algorithms (Analysis Services - Data Mining)

Data Mining Extensions (DMX) Reference

Related Sections

Basic Data Mining Tutorial

This tutorial introduces basic concepts, such as how to create a project and how to build mining structures and mining models.

Intermediate Data Mining Tutorial (Analysis Services - Data Mining)

This tutorial contains a number of independent lessons, each introducing you to a different model type. Each lesson walks you through the process of creating a model, exploring the model, and then customizing the model and creating prediction queries.

See Also

Working with Data Mining

Using the Data Mining Tools

Designing and Implementing (Analysis Services - Data Mining)

146

Соседние файлы в предмете [НЕСОРТИРОВАННОЕ]