482 Exam 2 Questions and Correct Solutions Graded A+

482 Exam 2 Questions and Complete Solutions Graded A+ What are the three Business Analytics Tiers? - Answer: - Reporting - Basic Analytics - Advanced Analytics Within the Business Analytics Tier of Reporting, what does Basic Reporting ask? - Answer: What Happened? Within the Business Analytics Tier of Reporting, what does Ad Hoc Reporting ask? - Answer: How many, how often, where? Within the Business Analytics Tier of Reporting, what does Dynamic Reporting ask? - Answer: Where exactly are the problems? Within the Business Analytics Tier of Basic Analytics, what does Reporting with Early Warning ask? - Answer: What actions are needed? Within the Business Analytics Tier of Basic Analytics, what does Basic Statistical Analysis ask? - Answer: Why is this happening? What are the three Features of Business Analytics? - Answer: - The Data - The Users and Sponsors - The Methodology Within Features of Business Analytics, The Data is: - Answer: Massive, operational, and opportunistic Within Features of Business Analytics, The Users and Sponsors utilize: - Answer: Business decision support Within Features of Business Analytics, The Methodology includes: - Answer: 1. Computer-intensive ad hockery 2. Multidisciplinary lineage What is Data Mining? - Answer: Advanced methods for exploring and modeling relationships in large amounts of data. _________________ is an essential component of business analytics - Answer: Data mining When Analytics are not helpful: - Answer: - Snap decisions - No previous data is possible - Most important factors are rare (making decisions to work around unlikely obstacles or miracles) - Expert analysis suggests a particular path - Metrics are inappropriate - Naïve implementation of analytics - Confirming what you already know What does SAS Studio Provide? - Answer: a point-and-click interface for managing data and generating reports SAS Studio interface is ______________ and the interactive tools and reports are rendered in HTML5 - Answer: browser-based SAS Studio generates SAS ______________ that are submitted to SAS behind the scenes. - Answer: programs _______________ are returned to SAS Studio and are delivered through the browser-based interface. - Answer: Results __________________________ can easily be saved in .html, .pdf, or .rtf formats. - Answer: Graphical and report output ___________________ also includes a full programming interface that can be used to write, edit, and submit SAS programs. - Answer: SAS Studio _________________________ streamlines the data mining process to create highly accurate predictive and descriptive models based on vast amounts of data gathered from across an organization. - Answer: SAS Enterprise Miner What are the Strengths of SAS EM? - Answer: a. Predictive Modeling b. Pattern Discovery ______________________ brings self-service, ad hoc visual data discovery and exploration to an organization. - Answer: SAS Visual Analytics T/F: SAS Visual Analytics puts lightning-fast insights within everyone's reach: no matter the size of your organization - or your data. - Answer: True What are the Features of SAS Visual Analytics? - Answer: a. In-memory analytics for fast answers b. Autocharting automatically chooses the best graph c. Forecasting, scenario analysis, and other analytic visualizations d. Text analysis, including content categorization e. Interactive dashboarding and reporting f. Design reports once, publish anywhere g. Integration with mapping technologies h. Feature-rich mobile apps for iPad and Android i. Flexible, scalable deployment options What are the Tasks for Business Analytics:? - Answer: i. Identify and access data sources ii. Combine data sources iii. Transform variables iv. Explore and describe data v. Visualize patterns in the data vi. Analyze and model data vii. Validate models viii. Score ix. Update models ______________ consist of commands in the form of DATA steps and PROC steps - Answer: SAS programs For SAS Programs: - Answer: We focus on point-and-click interfaces that write programs and perform additional functions automatically. _____________________ brings together data components that can exist on multiple machines, from different software vendors, throughout the organization. - Answer: Data management ________________ requires data management activities such as data access, movement, transformation, aggregation, and augmentation. - Answer: Business analytics Managing Data can involve many different type of data: - Answer: For example, simple flat files, files with comma-separated values, Microsoft Excel files, SAS tables, and Oracle tables. _____________ likely combines individual transactions, customer summaries, product summaries, or other levels of data granularity - or some combination of those things. - Answer: data What is a Data Mart? - Answer: Logical subset of the complete data warehouse, or a sort of pie-wedge of the whole data warehouse What are the Reasons for Creating a Data Mart? - Answer: - Ease of creation - Lower cost that implementing a full data warehouse - Easy access to frequently used data - Is less cluttered What is the First Advantage of a Data Mart? - Answer: There is one version of the truth. What is the Second Advantage of a Data Mart? - Answer: Downstream tables are updated as source data is updated, so analyses are always based on the latest information. What is the Third Advantage of a Data Mart? - Answer: The problem of a proliferation of spreadsheets is avoided. What is the Fourth Advantage of a Data Mart? - Answer: Information is clearly identified by standardized variable names and data types. What is the Fifth Advantage of a Data Mart? - Answer: Multiple users can access the same data. What is a Data Warehouse? - Answer: Collection of constituent data marts Individual data marts are targeted to particular business groups, but the collection of all data marts form an integrated whole, is called the _______________________________ - Answer: enterprise data warehouse. Data Warehouses are Corporate/Enterprise-wide, while Data Marts are: - Answer: Departmental Data Warehouses are Union of all data marts, while Data Marts are: - Answer: A single Business process Data Marts are Technology optimal for data access and analysis, while Data Warehouses are good for: - Answer: Data received from staging areas Data Marts are Structured to suit the departmental view of data, while Data Warehouses are Structured for: - Answer: Corporate view of data SAS Studio can be used for data management, as well as a wide variety of other tasks: - Answer: i. data exploration ii. querying and reporting iii. graphical analysis iv. statistical analysis v. scoring Issues that need to be fixed in SAS Studio: - Answer: i. Files with fields named differently ii. Different file types iii. Combining data files Data Mining: - Answer: Is an iterative process of creating predictive and descriptive models, by uncovering previously unknown trends and patterns in vast amounts of data, in order to support decision making. Data mining is a __________ of Business Analytics. - Answer: subset What are the Two Types of Data Mining Analysis? - Answer: - Discovery - Predictive Modeling Within Data Mining Analysis, Discovery includes: - Answer: Unsupervised classification - Cluster analysis - Association rules Within Data Mining Analysis, Predictive Modeling includes: - Answer: Supervised classification - Linear regression - Logistic regression - Decision trees For Type of Targets, Logistic Regression Includes: - Answer: 1. Event/no event (binary target) 2. Class label (multiclass problem) For Type of Targets, Regression includes: - Answer: Continuous Outcomes For Type of Targets, Survival Analysis includes: - Answer: Time-to-event (possibly censored) What are the steps in a Data Mining Project? - Answer: i. Define or refine business objective ii. Select data iii. Explore input data iv. Prepare and repair data v. Transform input data vi. Build models vii. Deploy models viii. Assess results Where is the Menu Bar in SASEM? - Answer: Where is the Project Panel in SASEM? - Answer: Where is the Properties Panel in SASEM? - Answer: Where is the Help Panel in SASEM? - Answer: Where is the Node in SASEM? - Answer: Where is the Process Flow in SASEM? - Answer: Steps to create a new project in SAS EM: - Answer: 1. Name the project. 2. Specify location to store project. 3. Create a diagram after project is created. 4. Define data library after project is created. 5. Link the data source to Enterprise Miner. 6. Create metadata. 7. Create a process flow. What does an Input Data Node do? - Answer: i. Where you can upload a SAS file What does an Input File Node do? - Answer: i. Where you can upload a non SAS file (like an Excel file) What is Metadata? - Answer: data about data What are the Three Purposes for Metadata? - Answer: 1. Define variable roles (such as input, target, or ID). 2. Define measurement levels (such as binary, interval, or nominal). 3. Define table role (such as raw data, transactional data, or scoring data). What Does SEMMA Stand for? - Answer: - Sample - Explore - Modify - Model -Asses What Functionalities are included in the Sample tab? - Answer: 1. Append 2. Data Partition a. Training, Validation, Test 3. File Import 4. Filter 5. Input Data 6. Merge 7. Sample What Functionalities are included in the Explore tab? - Answer: 1. Association 2. Cluster 3. DMDB 4. Graph Explore 5. Link Analysis 6. Market Basket 7. Multiplot 8. Path Analysis 9. SOM/Kohomen 10. StatExplore 11. Variable Clustering 12. Variable Selection The ______________ is an advanced visualization tool that enables you to explore large volumes of data graphically to uncover patterns and trends and to reveal extreme values in the database. - Answer: Graph Explore node With Graph Explorer, you can: - Answer: analyze univariate distributions, investigate multivariate distributions, create scatter and box plots, constellation and 3-D charts, and so on. The MultiPlot node creates the following types of charts: - Answer: - Bar Charts - Scatter Plots Bar Charts includes: - Answer: 1. Histogram of each input and target 2. Bar chart of each input versus each class target 3. Bar chart of each input grouped by each interval target Scatter Plots includes: - Answer: 1. Plot of each interval input versus the target 2. Plot of each class input versus the target What is the StatExplore node used for? - Answer: is a multipurpose tool that you use to examine variable distributions and statistics in your data sets. The _____________node generates summarization statistics. - Answer: StatExplore Variable Clustering Selects: - Answer: Variables or cluster components - Removes collinearity ________________ decreases variable redundancy - Answer: Variable Clustering Variable Clustering Reveals: - Answer: Underlying structure of input variables Variable Selection includes: - Answer: - Reduces the number of inputs not related to target - R-Square - Chi-square What Functionalities are included in the Modify tab? - Answer: 1. Drop 2. Impute 3. Interactive Binning 4. Principal Components 5. Replacement 6. Ruler Builder 7. Transform Variables What is Impute used for? - Answer: to replace missing values For Impute, many modeling techniques will: - Answer: drop rows of data that have any missing values Impute creates: - Answer: imputation indicator variables What is Interactive Binning? - Answer: A grouping tool to model non-linear functions of multiple modes of continuous distributions For Interactive Binning, computes initial bins by: - Answer: quantiles, enabling interactively splitting and combining the initial bins What is Replacement used for? - Answer: to interactively specify replacement values for class and interval levels Replacement is good for: - Answer: Trimming Outliers and Replace Missing ________________ is used to generate score code to process unknown levels when scoring - Answer: Replacement Transform Variables enables you to: - Answer: create new variables or transform existing variables ________________ creates interaction variables - Answer: Transform Variables Interval Transformations include: - Answer: - Best - Multiple - Log - Log 10 - Square Root - Inverse - Square - Exponential - Centering - Standardize - Range - Bucket - Quantile - Optimal Binning - Maximum Normal - Maximum Correlation - Equalize - Optimal Max. Equalize - None Class Transformations include: - Answer: Group rare levels What Functionalities are included in the Model tab? - Answer: 1. AutoNeural 2. Decision Tree 3. Dmine Regression 4. DMNeural 5. Ensemble 6. Gradient Boosting 7. LARS 8. MBR 9. Model Import 10. Neural Network 11. Partial Least Squares 12. Regression 13. TwoStage What is AutoNeural used for? - Answer: Used to automatically configure a neural network What does AutoNeural implement? - Answer: a search algorithm to incrementally select activation functions for a variety of multilayer networks What do Decision Trees classify? - Answer: observations based on the values of nominal, binary, or ordinal targets ________________ predict outcomes for interval targets - Answer: Decision Trees Decision Trees are: - Answer: Easy to interpret and Interactive What Does Dmine Regression compute? - Answer: a forward stepwise least squares regression ___________________ (Optional) Computes all two-way interactions of classification variables - Answer: Dmine Regression Dmine Regression (Optional) Uses AOV16 variables to: - Answer: identify non-linear relationships between interval variables and the target variable Dmine Regression Groups variables to: - Answer: reduce the number of levels of classification variables __________________ is another modeling node that you can use to fit an additive nonlinear model. - Answer: DMNeural For DMNeural, The additive nonlinear model uses: - Answer: bucketed principal components as inputs to predict a binary or an interval target variable with automatic selection of an activation function. What does Ensemble create? - Answer: new models by combining the posterior probabilities (for class targets) or the predicted values (for interval targets) from multiple predecessor models The Three Methods that Ensemble includes are: - Answer: - Average - Maximum - Voting Gradient Boosting utilization: - Answer: a. Sequential ensemble of many trees b. Extremely good predictions c. Very effective at variable selection What does LARS stand for? - Answer: Least Angle Regression What does LASSO stand for? - Answer: Least Absolute Shrinkage and Selection Operator ____________ can be used for modeling or variable selection - Answer: LARS _________________ has an non-linear relationship between inputs and output - Answer: Neural Network For Neural Network, prediction is: - Answer: more important than ease of explaining model Neural Networks enable you to: - Answer: construct, train, and validate multilayer feed for neural networks. For Neural Network, users may select from several predefined: - Answer: architectures or manually select input, hidden, and target layer functions and options. ______________________ enables you to model a class target and an interval target. The interval target variable is usually the value that is associated with a level of the class target. - Answer: TwoStage TwoStage allows you to choose between: - Answer: Sequential or Concurrent modeling For TwoStage, you can choose the type of model at each stage: - Answer: Tree, Regression, or Neural The score code of the _______________ is a composite of the class and value models. - Answer: TwoStage node What Functionalities are included in the Assess tab? - Answer: 1. Cutoff 2. Decisions 3. Model Comparisons 4. Score 5. Segment Profile _____________________ provides a common framework for comparing models - Answer: Model Comparison Model Comparison includes: - Answer: many diagnostic charts Within Model Comparison, Model Assessment Criterion includes: - Answer: - For decision prediction - For ranking predictions - For estimate predictions - User can override default criterion for determining the champion Within Model Assessment, For decision prediction includes: - Answer: 1. Accuracy and Misclassification 2. Profit or loss 3. Kolmogorov-Smirnov (KS) Statistic Within Model Assessment, For ranking predictions includes: - Answer: 1. ROC index a. The validation data is sorted from high to low (either scores or estimates). Each point on the ROC chart corresponds to a specific fraction of the sorted data. 2. GINI coefficient Within Model Assessment, For estimate predictions includes: - Answer: 1. Average squared error 2. Schwarz's Bayesian Criterion (SBC) What is Score Ranking? - Answer: The validation data is sorted from highest to lowest (either by prediction rankings or estimates). Each point on the ROC chart corresponds to a specific fraction of cases. Scoring enables you to score, that is: - Answer: predict or classify, an observation from a data set whose role is score. For Score, Each node in the process flow generates: - Answer: scoring code, and this code is compiled and passed to the Score node. This enables the scoring data set to be scored outside of Enterprise Miner Modifying Data for Regression Modeling includes: - Answer: 1. Manage missing values 2. Interpret the model 3. Handle extreme or unusual values 4. Use nonnumeric inputs 5. Account for nonlinearities SAS Rapid Predictive Modeler generates: - Answer: predictive models quickly and accurately SAS Rapid Predictive Modeler provides: - Answer: self-sufficiency to business users SAS Rapid Predictive Modeler creates: - Answer: Easy-to-understand charts and reports SAS Rapid Predictive Modeler integrates: - Answer: analytics and BI for better decisions SAS Rapid Predictive Modeler enables: - Answer: business users, without prior statistical knowledge, to build predictive models quickly and effectively. ___________________________ results can be consumed in simple and easy-to-understand charts to take better decisions. - Answer: SAS Rapid Predictive Modeler SAS Rapid Predictive Modeler requires: - Answer: SAS EM to work SAS Rapid Predictive Modeler allows you to choose from: - Answer: Basic, Intermediate, or Advanced prebuilt models. SAS Rapid Predictive Modeler automatically: - Answer: handles outliers, missing values, rare target events, skewed data, variable selection and model selection. For SAS Rapid Predictive Modeler, analytic results are: - Answer: presented in easy to understand business terms: scorecard, lift charts, listing of key variables in the model. Analytic experts can further customize and improve models developed in: - Answer: SAS Rapid Predictive Modeler using SAS Enterprise Miner. For SAS Rapid Predictive Modeler, models are registered in SAS metadata to: - Answer: a. Automate the execution of score code b. Ease deployment to other systems