260 likes | 437 Views
DAT377 Data Mining In SQL Server 2000 And SQL Server 2005 (Code Named “Yukon”). Paul Bradley Principal, Data Mining Technology Apollo Data Technologies. Data Mining Today. Fastest growing BI Segment (IDC) Data Mining Tools: $1.85B in 2006 Mainstream Emergence
E N D
DAT377Data Mining In SQL Server 2000 And SQL Server 2005 (Code Named “Yukon”) Paul Bradley Principal, Data Mining Technology Apollo Data Technologies
Data Mining Today • Fastest growing BI Segment (IDC) • Data Mining Tools: $1.85B in 2006 • Mainstream Emergence • E-commerce (e.g., Amazon.com) • Search (e.g., Vivisimo.com) • Press: “Airline Gave Government Information on Passengers” [NYT, 1/18/2004] • Politics: Data Mining Moratorium Act • SQL-Server is in a unique position to service market needs
Data Mining Operations • Explore Your Data • Identify Patterns and Trends • Act on Patterns and Trends
ApplicationsEmbedded Data Mining • Make Decisions without Coding • Learn business rules directly from data • Client Customization • Learn logic customized for each client • Automatic Update • Data mining application logic updated by model re-processing • Applications do not need to be rewritten, recompiled, re-deployed
Yukon (SQL Server 2005)BI Platform Development Tools Management Tools Reporting Services Analysis Services OLAP & Data Mining Data Transformation Services ETL SQL Server Relational Engine
Win Leadership • Continue standards and developer effort • Comprehensive feature set • Penetrate the Enterprise • Thought leadership Microsoft Data Mining SQL 2005 SQL 2000 Enter the Game • Create industry standard • Target developer audience • V1.0 product with 2 algorithms
Business Knowledge Data Mining OLAP Relative Business Value Reports (Adhoc) Reports (Static) Easy Difficult Usability Value Of Data Mining SQL-Server 2005
“Enabling” • Provide a toolset that is accessible to a wide range of disciplines • Provide data mining at a price point to achieve wide market penetration • Provide database metaphors for managing, distributing learned knowledge • Provide APIs allowing the embedding of data mining functionality into applications
Complete set of tools Integrated solution Embedded data mining Partner Alliances What Yukon Offers Extensive creation and editing environment Six most popular DM algorithms One unique algorithm 12 embeddable viewers
Complete set of tools Integrated solution Embedded data mining Partner Alliances What Yukon Offers Tight integration with OLAP, Relational, DTS, Reporting Easy integration to Web/office applications
Complete set of tools Integrated solution Embedded data mining Partner Alliances What Yukon Offers Embed DM to LOB applications Complete SQL language-based API Native XMLA support Data Mining as a Web service
Complete set of tools Integrated solution Embedded data mining Partner Alliances What Yukon Offers Partner program to bring in ISVs early Complete Data Mining platform Focus on broadening the market DMX to be industry standard in XMLA
Decision Trees Clustering Time Series Naïve Bayes SequenceClustering Association Neural Net Yukon Algorithms
Case Study Data Mining in Market Research
Market ResearchSurvey Analysis • Financial Security Compliance Survey • Identify Respondent-Level Trends • Perform Segment Analysis • Identify Key Drivers • Profile Respondent Sub-Populations • Support • Marketing Program Recommendations • Upper Management Scorecards
Market ResearchSecurity Compliance Survey • Survey Sponsor: National Financial Security Compliance Agency • 37 Questions • System/Network Security Compliance • Drill-down Questions • 1192 Respondents
Market ResearchYukon Data Mining • Utilize the following Yukon Analysis Services Algorithms • Microsoft Decision Trees (Key Driver Analysis) • Microsoft Clustering (Segment Analysis) • Microsoft Naïve Bayes (Sub-Population Profiles)
Case Study Portfolio Prediction
Portfolio Prediction • Identify Correlations and Trends in Historic Stock Market and Financial Data • Seasonality • Adapt to Disruptive Change • Utilize Correlations and Trends to Estimate Future Movement
Portfolio PredictionYukon Data Mining • Monthly Stock Market Data for 21 Stocks • Data Collected Over 10 Years • Monthly High Value • Utilize Microsoft Time Series to Model Temporal Trends • Predict 6 Months into the “Future”
Conclusion • Data Mining is Making Significant BI Impact • Yukon Provides Powerful Data Mining Platform • Embedding Data Mining in BI Applications • Data Mining Model Management and Security • Capitalize Your Data Assets for Strategic Business Decision-Making
Next Steps: SQL Server 2005 Exclusive TechEd Offer! Receive Beta 2 of SQL Server 2005 Register for SQL Server 2005 Beta 2 at: http://www.msteched.com/SqlBetaBits.aspx Visit the SQL Server 2005 Web site www.microsoft.com/sql/2005 Learn more about SQL Server 2005 at TechEd Hands On Labs Rooms 6E and 6F 13 Hands On Labs Ask the Experts Track Cabanas located around CommNet Experts Available All Week
SQL Server Community sites http://www.microsoft.com/sql/community/default.mspx List of newsgroups http://www.microsoft.com/sql/community/newsgroups/default.mspx Locate Local User Groups http://www.microsoft.com/communities/usergroups/default.mspx Attend a free chat or Web cast http://www.microsoft.com/communities/chats/default.mspx http://www.microsoft.com/usa/webcasts/default.asp
Please fill out a session evaluation on CommNet Q1: Overall satisfaction with the session Q2: Usefulness of the information Q3: Presenter’s knowledge of the subject Q4: Presenter’s presentation skills Q5: Effectiveness of the presentation
© 2004 Microsoft Corporation. All rights reserved. This presentation is for informational purposes only. Microsoft makes no warranties, express or implied, in this summary.