gavin russell-rockliff bi technical specialist microsoft bin305

Taking Your Application Design to the Next Level with Data Mining

Gavin Russell-RockliffBI Technical SpecialistMicrosoftBIN305

Please Raise Your Hand If You’ve Ever…

<… Put a Party Reference Here… >Attended a Statistics Lecture ??Got a Statistics Degree ??Used SQL Server Data Mining ??

Agenda

Data Mining – What is it?Data Mining – How do we do it?Demonstrations

VisualisationReportingETLApplication

Q&A

Data Mining – What Is It?

According to EncartaNoun“Search for Hidden Information”“The locating of previously unknown patterns and relationships within data”

Server-Driven DiscoveryUses a combination of statistics, probability analysis and database technologies

DM Enables Predictive Analysis

Predictive Analysis

Presentation Exploration Discovery

Passive

Interactive

Proactive

Business Insight

Canned reporting

Ad-hoc reporting

OLAP

Data mining

Role of Software

Business Scenarios

Forecasting sales

Churn Analysis

Detecting fraud or invalid data

Targeting promotions

Cross-selling

Determine Business Drivers

Our End-to-End BI Offering

END USER TOOLS AND PERFORMANCE MANAGEMENT APPLICATIONS

BI PLATFORM (RDBMS, ETL, OLAP, Reporting)

DELIVERY

Mainframe/ Departmental Systems

The Big Picture

SQL Server Reporting Services

SQL Server Analysis Services

SQL Server DBMS

SQL Server Integration Services

Our End-to-End BI Offering

END USER TOOLS AND PERFORMANCE MANAGEMENT APPLICATIONS

BI PLATFORM (RDBMS, ETL, OLAP, Reporting)

DELIVERY

The Big Picture

SQL Server Analysis Services

SQL Server™ 2008 Data MiningKey Drivers

Keep Development SimpleRetain Full Suite of AlgortihmsManage Large VolumesAllow for Integration

SQL Server™ 2008 AlgorithmsMicrosoft Naïve Bayes

Quick and approachable algorithmUsed for classification

Microsoft Decision TreesPopular data mining techniqueUsed for classification, regression and association

Microsoft Linear RegressionFinds the best possible straight line through a series of pointsUsed for prediction analysis

SQL Server™ 2008 AlgorithmsContinued

Microsoft Neural NetworkMore sophisticated than Decision Trees and Naïve Bayes, this algorithm can explore extremely complex scenariosUsed for classification and regression tasks

Microsoft Logistic RegressionA particular case of the Neural Network algorithm

Microsoft ClusteringFinds natural groupings inside dataSupports segmentation and anomaly detection tasks

SQL Server™ 2008 AlgorithmsContinued

Microsoft Sequence ClusteringGroups a sequence of discrete events into natural groups based on similarity

Microsoft Time SeriesUsed to predict future values from a time seriesHas been improved in SQL Server 2008 to produce more accurate long-term forecasts

Microsoft Association RulesCommonly supports market basket analysis to learn what products are purchased together

Data Mining Algorithm Usage

What is your task?Predict Variable

•Naïve Bayes•Decision Trees•Neural Network•Logistic Regression

Predict Value

•Decision Trees•Linear Regression•Neural Network•Logistic Regression

Marketing Cluster

•Clustering

Forecast Value

•Time Series

Associate

•Association Rules•Decision Trees

Data Mining Process

Define the ProblemData PreperationModel Validation

AccuracyReliabilityUsefulness

Model Visualisation

Describing the Data Mining Process

Design time

Process time

Query timeMining Model

Describing the Data Mining Process

Design time

Process time

Query timeMining Model

Training Data Data Mining Engine

Data Mining Visualization

Model Creation + Processingdemo

Describing the Data Mining Process

Design time

Process time

Query timeMining Model

Training Data Data Mining Engine

Describing the Data Mining Process

Design time

Process time

Query time

Data Mining Engine

Data to PredictPredicted Data

Mining Model

Predicting the Futuredemo

Data Mining for the Developerdemo

question & answer

Related Content

Breakout Sessions

Using MDX for Enhanced Scorecards and Dashboards (BIN 307)

Track Resources

www.sqlserverdatamining.com

www.microsoft.com/sql

twitter.com/gavinrr

www.microsoft.com/teched

Sessions On-Demand & Community

http://microsoft.com/technet

Resources for IT Professionals

http://microsoft.com/msdn

Resources for Developers

www.microsoft.com/learning

Microsoft Certification & Training Resources

Resources

Complete a session evaluation and enter to win!

10 pairs of MP3 sunglasses to be won

© 2009 Microsoft Corporation. All rights reserved. Microsoft, Windows, Windows Vista and other product names are or may be registered trademarks and/or trademarks in the U.S. and/or other countries.The information herein is for informational purposes only and represents the current view of Microsoft Corporation as of the date of this presentation. Because Microsoft must respond to changing market conditions, it should not be interpreted to be a commitment on the part of Microsoft, and Microsoft cannot guarantee the accuracy of any information provided after the date of this presentation. MICROSOFT MAKES NO WARRANTIES, EXPRESS,

IMPLIED OR STATUTORY, AS TO THE INFORMATION IN THIS PRESENTATION.