Author Description

This book provides self-paced training for Data Scientists and Business Analysts who want to learn more about Teradata Aster. After installing the VMware images that are included with the book via USB thumb drive, you will be able to perform Aster SQL-MapReduce labs using three different data stores (Teradata Aster, Aster and Hadoop). In-depth labs cover the Teradata Unified Data Architecture as well as the top 35 Aster functions. Each Aster function will first go over the correct syntax, followed by the various arguments and their usage. Optionally, we'll cover the output that in the result set. And finally, you will have number of hands-on labs to reinforce your learning. A partial list of functions include: 1. Time-series: Sessionize, Attribution, nPath 2. Associative: Collaborative Filtering, Market Basket 3. Graph: nTree, Degrees, Triangle_finder, Rectangle_Finder, PageRank 4. Predictive: Naive Bayes, Decision Trees, SVM, KNearest Neighbor along with Confusion Matrix 5. Parser: Apache, JSON, XML, XMLRelation 6. Clustering: Canopy, KMeans, KMeansPlot 7. Statistical: Correlation, Histogram, Enhanced Histogram, Linear Regression, GLMPredict 8. Text: Text Parser, nGram, Naive Bayes Text Classifier, Sentiment Analysis 9. Visualization: nPathViz and cFilterViz 10. Teradata Unified Data Architecture: QueryGrid: Aster-Teradata, QueryGrid: Aster-Hadoop, QueryGrid: Teradata-Aster, AnyDatabase2Aster After the hands-on sections, the book covers a number of Use Case scenarios including churn analysis, analyzing Click stream data and recommendation of other tv shows for followers of a particular tv shows. These highlight how you can nest functions within functions to acquire, prepare, analyze and visualize the data all within one SQL-MR statement.