microsoft data mining 2012
DESCRIPTION
Author: William Brown, Microsoft BI Specialist > This slide presentation covers Microsoft Data Mining functionality from the developer to the end user. In the past, data mining belonged to the deep technical specialist, but the current Microsoft stack allows anyone to create very powerful data mining models. Data mining allows users to find insights that are difficult or impossible to discover with traditional analysis. You'll learn * How to get started with Data mining * The various data mining models and where they can be applied * How to create models and surface the data to users * How to use the new Excel Data mining add-inTRANSCRIPT
Introduction to Microsoft Data Mining
Speaker: William BrownMicrosoft Business Intelligence
September 2012
Mark Ginnebaugh, User Group Leader [email protected]
Objectives
William Brown, Microsoft BI Architect September 2012
Data Mining is….
Data mining is the locating of previously unknown patterns and relationships within data using a database application
Identify Cross Selling
Opportunities
Locate and understand profitable customers
Understand and predict
fraud
Forecast sales and inventory
data
Identify and handle
anomalies during data transfer or data loading
William Brown, Microsoft BI Architect September 2012
Data Mining does not….
Data mining is the locating of previously unknown patterns and relationships within data using a database application
Remove your need to
understand your data!
Reduce the work to
prepare and organize the
data
Find simple answers to complex questions
Reduce the impact of dirty data
Magically make your life
easier
William Brown, Microsoft BI Architect
Describing the Data Mining Process
“Putting Data Mining to
Work”
“Doing Data Mining”Business
UnderstandingData
Understanding
Data Preparation
Modeling
Evaluation
Deployment
Data
www.crisp-dm.org
William Brown, Microsoft BI Architect
“Putting Data
Mining to Work”
Data Preparation
William Brown, Microsoft BI Architect
Data Mining Modeling
Design time
Process time
Query time Mining Model
William Brown, Microsoft BI Architect
Data Mining Modeling
Mining Model
Training Data
Data Mining Engine
William Brown, Microsoft BI Architect
Design time
Process time
Query time
Data Mining Modeling
Data Mining Engine
Data to Predict
Predicted Data
Mining Model
William Brown, Microsoft BI Architect
Design time
Process time
Query time
Introducing Analysis Services 2012
William Brown, Microsoft BI Architect
Hides the complexity Includes full suite of algorithms to automatically
identify and store patterns in your data
Intro to SQL Server Data Mining
William Brown, Microsoft BI Architect
Free add‐in for Excel 2010 Works with 32 and 64 bit editions of Office 2010
Requires SQL Server Analysis Services Analyze Tab – simpler to use Data Mining Tab – full power
Data Mining Add‐Ins for Excel
William Brown, Microsoft BI Architect
SQL Server Data Mining Algorithms
William Brown, Microsoft BI Architect
SQL Server Data Mining AlgorithmsContinued
William Brown, Microsoft BI Architect
SQL Server Data Mining AlgorithmsContinued
William Brown, Microsoft BI Architect
SQL Server Data Mining AlgorithmsContinued
Classify
• Decision Trees
• Logistic Regression
• Naïve Bayes
• Neural Networks
Estimate
• Decision Trees
• Linear Regression
• Logistic Regression
• Neural Networks
Cluster
• Clustering
Forecast
• Time Series
Associate
• Association Rules
• Decision Trees
William Brown, Microsoft BI Architect
Data Mining Add‐Ins for Excel
Menu Data mining
Analyze Key Influencers Naïve Bayes
Detect Categories Clustering
Fill from Example Logical Regression
Forecast Time Series
Highlight Exceptions Clustering
Scenario Analysis – Goal Seek Logical Regression
Scenario Analysis – What if Logical Regression
Predicton Calculator Logical Regression
Shopping Basket Association Rules
William Brown, Microsoft BI Architect
SQL Server Data Mining Visualizations
William Brown, Microsoft BI Architect
1. Creating, training, testing data mining models with SSDT2. Using Excel for user driven data mining3. Authoring a Reporting Services report based on a data mining
model4. Automating data validation with data mining
Message for Developers
William Brown, Microsoft BI Architect
http://www.microsoft.com/sqlserver/en/us/solutions‐technologies/business‐intelligence/data‐mining.aspx
http://www.sqlserverdatamining.com
http://www.predixionsoftware.com/predixion/
Technical Resources
William Brown, Microsoft BI Architect September 2012
To learn more or inquire about speaking opportunities, please contact:
Mark Ginnebaugh, User Group [email protected]