Vous êtes sur la page 1sur 24

Module 9: Introduction to Data Mining

Overview
Overview of Data Mining Creating a Data Mining Solution

Validating Data Mining Models

Lesson 1: Overview of Data Mining


What Is Data Mining? Data Mining Concepts

Data Mining Algorithms

Discussion: Why Mine Data?


Forecasting Sales Targeted Advertising

Credit Ratings

What Is Data Mining?


Data Mining is:
A statistical analysis of data Used to identify trends and patterns

Data Mining Concepts


Data Mining Structure
The central component of a data mining solution

Case Table
Store the source data for the data mining models

Data Mining Model


Defines the data mining algorithm

Data Mining Algorithms


Data Mining Algorithms:
Microsoft Decision Trees Microsoft Time Series* Microsoft Clustering

Microsoft Association
Microsoft Sequence Clustering Microsoft Neural Network Microsoft Nave Bayes Microsoft Linear Regression Microsoft Logistic Regression

Lesson 2: Creating a Data Mining Solution


Data Mining Tools Using the Data Mining Wizard

Using the Data Mining Designer


Viewing a Data Mining Model What Is DMX?

Querying a Cube Using DMX

Data Mining Tools

Data Mining Wizard Data Mining Designer

Using the Data Mining Wizard


Steps to complete the Data Mining Wizard:

1 Specify the definition method 2 Specify the data mining technique 3 Specify the Data Source View

4 Specify table types


5 Specify training data 6 Specify column content and data types

Using the Data Mining Designer

Data Mining Designer tabs Adding models to a data mining structure

Viewing a Data Mining Model


Mining model viewers:

Microsoft Tree Viewer


Microsoft Cluster Viewer Microsoft Time Series Viewer Microsoft Nave Bayes Viewer Sequence Cluster Viewer Microsoft Association Rules Viewer Microsoft Neural Network Viewer

Demonstration: Using Data Mining


In this demonstration you will learn:
How to review a data mining structure

How to view a data mining model

What Is DMX?
DMX is:
Data Mining Extension language An extension of the SQL language

Querying a Cube Using DMX


To build prediction queries:
Use Read Prediction Query Builder Use Read Query Contingent Editor (syntax is similar to T-SQL) DMX Read/Write templates Mining Model Prediction tab of Business Intelligence Range restricted by MDX statement Development Studio

Lesson 3: Validating Data Mining Models


Overview of Data Mining Validation Accuracy Charts

Viewing Accuracy Charts

Overview of Data Mining Validation


Validate and Compare Mining Models Compare the results of the mining model to known data Display the accuracy of the models using accuracy charts

Accuracy Charts

Classification Matrix Lift Charts

Viewing Accuracy Charts

Model Accuracy Lift Chart Predicted Profit

Lab: Implementing Data Mining


Exercise 1: Creating a Data Mining Structure Exercise 2: Adding a Data Mining Model

Exercise 3: Exploring Data Mining Models


Exercise 4: Validating Data Mining Models

Logon information

Virtual machine User name Password

NY-SQL-01 Administrator Pa$$w0rd

Estimated time: 60 minutes

Lab Scenario
You have been asked to add additional features to an

existing demonstration data mining solution. You will add new data mining structures to this project. You will also add two new data mining models to an existing data mining structure and then validate these models.

Lab Review
What is the difference between a data mining structure

and a data mining model? inputs?

Why do some algorithms only allow certain columns as What is a training set?

Module Review and Takeaways


Review Questions Common Issues and Troubleshooting Tips

Course Evaluation

Vous aimerez peut-être aussi