Oracle Advanced Analytics

Click here to load reader

  • date post

    13-May-2015
  • Category

    Documents

  • view

    3.760
  • download

    7

Embed Size (px)

Transcript of Oracle Advanced Analytics

  • 1.Oracle Advanced AnalyticsOracle R Enterprise & Oracle Data MiningExtending the Database into a Comprehensive Advanced Analytics PlatformCharlie BergerSr. Director Product Management, Data Mining and Advanced AnalyticsOracle [email protected] www.twitter.com/CharlieDataMineR

2. The following is intended to outline our general product direction. It is intended for information purposes only, and may not be incorporated into any contract. It is not a commitment to deliver any material, code, or functionality, and should not be relied upon in making purchasing decisions. The development, release, and timing of any features or functionality described for Oracles products remains at the sole discretion of Oracle.2 Copyright 2012, Oracle and/or its affiliates. All rights reserved. 3. Oracle Advanced Analytics OptionAgendaExtending the Database into a Comprehensive Advanced Analytics Platform Big Data AnalyticsOracle Advanced Analytics Option Oracle Data Mining SQL & PL/SQL focused in-database data mining and predictive analytics Oracle R Enterprise Integrates Open Source R with the Oracle DatabaseR3 Copyright 2012, Oracle and/or its affiliates. All rights reserved. 4. Big Data AnalyticsRBut the bigger you get, the more likely you are to be doingextensive data mining and the more likely you are to beimplementing or moving towards in-database analytics.Jim Kobielus, a senior data management analyst with Cambridge, Mass.-based ForresterResearch Inc. quote from Customary Data Warehouse Concepts vs. Hadoop: ForresterMakes the Call,Mark Brunelli, Senior News EditorThis RSS Reprints Published: 11 Aug 2011http://searchdatamanagement.techtarget.com/news/2240039468/Customary-data-warehouse-concepts-vs-Hadoop-Forrester-makes-the-call?vgnextfmt=print4 Copyright 2012, Oracle and/or its affiliates. All rights reserved. 5. Oracle Advanced Analytics OptionTransforming the Database into a Comprehensive Advanced Analytics Platform Oracle Advanced Analytics Option enables companies to R"bring the algorithms to the data" vs. extracting the data to specialized andexpensive dedicated statistical and data mining servers Oracle Advanced Analytics Option includes: Oracle Data Mining SQL & PL/SQL focused in-database data mining and predictive analytics Oracle R Enterprise Integrates the Open-Source Statistical Environment R with the Oracle Database Data movement is eliminated or dramatically reduced while analytical and computeintensive operations are performed inside the database, where the data resides, toincrease performance, reduce cycle times required to extract information from dataand reduce total cost of ownership over traditional statistical and data miningenvironments5 Copyright 2012, Oracle and/or its affiliates. All rights reserved. 6. Oracle In-Database Advanced AnalyticsComprehensive Advanced Analytics PlatformOracle R Enterprise Oracle Data Mining Popular open source statistical Automated knowledge discovery inside theprogramming language & environmentDatabase Integrated with database for scalability 12 in-database data mining algorithms Wide range of statistical and advanced Text mininganalytical functions Predictive analytics applications R embedded in enterprise appls & OBIEE Exploratory data analysis Rdevelopment environment Star schema and transactional data mining Extensive graphics Exadata "scoring" of ODM models Open source R (CRAN) packages SQL Developer/Oracle Data Miner GUI Integrated with Hadoop for HPCStatisticsAdvanced Analytics Data & Text MiningPredictive Analytics6 Copyright 2012, Oracle and/or its affiliates. All rights reserved. 7. OracleHardware and SoftwareEngineering to Work Together Oracle is the worlds most complete, open, and integrated business software and hardware systems company Data Warehousing, VLDB and ILM Oracle Data MiningNew GUI 12- in-DB data mining algorithms Mine star schemas, text, transactional data In-DB model build & applyExadata scoring 50+ in-DB statistical functions Oracle R EnterpriseNew Run R in-DB; function push down to SQL Wide library of supported in-DB statistical functions Embedded R supports all R packages Oracle has taught the RDBMS how to perform data mining, statistical analysis, adv. analytics, etc.7 Copyright 2012, Oracle and/or its affiliates. All rights reserved. 8. Oracles In-Database Advanced Analytics OptionValue Proposition10-100x PERFORMANCE Integrated features of the Database Perform analytics in-DB to eliminate data movement Reduce information latency: days-wks mins-hours10x LOWER TOTAL COST OF OWNERSHIP Eliminate/minimize expensive annual usage fees associated withtraditional stats/mining packages Leverage Oracle DB, DW & BI technology platform8 Copyright 2012, Oracle and/or its affiliates. All rights reserved. 9. Oracle Data MiningBuilding Predictive Analytics Applications Oracle Data Mining provides 12 powerful in-database data mining algorithms for bigdata analytics as a native feature of the databaseDesigned for or big data problems involving discovering patterns and relationships in large amounts of data and oftentimes making predictions based on those patterns, Oracle Data Mining allows data analysts and data miners to mine star schemas, transactional data and unstructured data stored inside the database, build predictive models and apply them to data inside the database--all without moving data. Oracle Data Miner help users mine their data and define, save and share advancedanalytical methodologiesUsers who prefer a Graphical User Interface can use the Oracle Data Miner extension to SQL Developer to develop, build, evaluate, share and automate analytical workflows to solve important data driven business problems. Developers can use the SQL APIs and PL/SQL to build applications to automateknowledge discoveryThe Oracle Data Miner GUI generates SQL code that application developers can use to develop and deploy SQL and PL/SQL based automated predictive analytics applications that run natively inside the Oracle Database.9 Copyright 2012, Oracle and/or its affiliates. All rights reserved. 10. What is Data Mining? Automatically sifts through data to find hidden patterns, discover new insights, and make predictions Data Mining can provide valuable results: Predict customer behavior (Classification) Predict or estimate a value (Regression) Segment a population (Clustering) Identify factors more associated with a businessproblem (Attribute Importance) Find profiles of targeted people or items (Decision Trees) Determine important relationships and market baskets within the population(Associations) Find fraudulent or rare events (Anomaly Detection)10 Copyright 2012, Oracle and/or its affiliates. All rights reserved. 11. In-Database Data Mining Traditional Analytics Oracle Data Mining Data ImportResults Faster time for Data MiningData to InsightsModel Scoring Lower TCOEliminatesData Preparation Data Movementand TransformationSavings Data Duplication Maintains SecurityData Mining Model BuildingModel ScoringData remains in the DatabaseData Prep & Transformation Embedded data preparationCutting edge machine learningModel Scoring algorithms inside the SQL kernel ofData ExtractionEmbedded Data PrepDatabaseModel Building Data PreparationSQLMost powerful language for data Hours, Days or WeeksSecs, Mins or Hourspreparation and transformation SourceDataset Analytic ProcessTargetData s/ WorkAreaal ProcessOutputData remains in the Database ing11Copyright 2012, Oracle and/or its affiliates. All rights reserved. 12. Oracle Data Miner Nodes (Partial List)Tables and Views Transformations Explore Data Modeling Text12 Copyright 2012, Oracle and/or its affiliates. All rights reserved. 13. Oracle Data Mining and Unstructured Data Oracle Data Mining mines unstructured i.e. text data Include free text and comments in ODM models Cluster and Classify documents Oracle Text used to preprocess unstructured text13Copyright 2012, Oracle and/or its affiliates. All rights reserved. 14. Oracle Data Mining Algorithms Problem AlgorithmApplicability ClassificationLogistic Regression (GLM)Classical statistical technique Decision Trees Popular / Rules / transparency Nave BayesEmbedded app Support Vector Machine Wide / narrow data / text RegressionMultiple Regression (GLM)Classical statistical technique Support Vector Machine Wide / narrow data / text Anomaly One Class SVMLack examples of target field Detection AttributeAttribute reduction Minimum Description Length (MDL) Identify useful data Importance A1 A2 A3 A4 A5 A6 A7Reduce data noise AssociationMarket basket analysis Rules AprioriLink analysisProduct grouping ClusteringHierarchical K-Means Text mining Hierarchical O-Cluster Gene and protein analysis FeatureText analysis ExtractionNonnegative Matrix Factorization Feature reductionF1 F2 F3 F414Copyright 2012, Oracle and/or its affiliates. All rights reserved. 15. SQL Developer 3.0/Oracle Data Miner 11g Release 2 GUI Graphical User Interface for data analyst SQL Developer Extension (OTN download) Explore datadiscover new insights Build and evaluate data mining models Apply predictive modelsNew GUI Share analytical workflows Deploy SQL Apply code/scripts15 Copyright 2012, Oracle and/or its affiliates. All rights reserved. 16. Oracle Data Miner 11g Release 2 GUI Churn DemoSimple Conceptual WorkflowChurn models to product and profile likely churners16 Copyright 2012, Oracle and/or its affiliates. All rights reserved. 17. Oracle Data Miner 11g Release 2 GUIChurn DemoSimple Conceptual WorkflowClustering analysis todiscover customer segments based onbehavior, demograhics,plans, equipment, etc. 17 Copyright 2012, Oracle and/or its affiliates. All rights reserved. 18. Oracle Data Miner 11g Release 2 GUI Churn DemoSimple Conceptual WorkflowMarket Basket Analysis to identify potential productbundless18 Copyright 2012, Oracle and/or its affiliates. All rights reserved. 19. Fraud Prediction