DATA MINING Desktop Survival Guide by Graham Williams
Desktop Survival Project Home Introduction Getting Started The Business Problem Data Loading Data Exploring Data Interactive Graphics Statistical Tests Models Network Analysis Text Mining Decision Trees Random Forests Boosting Bagging Support Vector Machine Linear Regression Neural Network Naive Bayes Survival Analysis Evaluation and Deployment Transforming Data Deployment Troubleshooting Issues Moving into R R Getting Help Data Graphics in R Understanding Data Preparing Data Issues Evaluating Models Reporting Fraud Analysis Archetype Analysis Algorithms Bayes Classifier K-Nearest Neighbours Linear Models Open Products AlphaMiner Borgelt Data Mining Suite KNime R Rattle Weka Closed Products Clementine Equbits Foresight GhostMiner InductionEngine ODM Enterprise Miner Statistica Data Miner TreeNet Virtual Predict Installing Rattle Projects Bibliography Index Preface Goals Organisation Features Audience Typographical Conventions A Note on Languages Currency Acknowledgements Introduction Data Mining The Business Problem Types of Analysis Data Mining Applications A Framework for Modelling Agile Data Mining R Rattle Why R and Rattle? Data Preparation Number of Algorithms Repeatability Performance Open Source Data Mining Business Case Sample Business Case Pros and Cons Books on R Getting Started Initial Interaction with R Quitting Rattle and R First Contact Loading a Dataset Building a Model Understanding Our Data Evaluating the Model Evaluating the Model Interacting with R Interacting with Rattle Projects Toolbar Menus Interacting with Plots Keyboard Navigation Summary Command Summary The Business Problem Solar Panel Efficiency Water Collection Others Other Business Problems Fraud Detection Loan Approval Documenting the Business Problem Summary Resources Exercises Data Data Nomenclature Loading Data into Rattle CSV Data Datasets Reading Direct from URL Play Golf Weather Data Other Data Sources ARFF Data ODBC Sourced Data Setting Up a Data Source Name Netezza Setup Teradata Setup R Data R Dataset Data Entry Data Tab Options in Rattle Sampling Data Variable Roles Automatic Role Identification Weights Calculator Manipulating Data Loading Data CSV Data Locating and Loading Data Loading the File CSV Options Basic Data Summary ARFF Data ODBC Sourced Data R Dataset R Data Library Data Options Sampling Data Variable Roles Automatic Role Identification Weights Calculator Command Summary Exploring Data Summarising Data Summary Describe Basics Kurtosis Skewness Missing Exploring Distributions Box Plot Histogram Cumulative Distribution Plot Benford's Law Other Digits Stratified Benford Plots Bar Plot Dot Plot Mosaic Plot GGobi Scatterplot Data Viewer Brushing Identify Multivariate Outliers Other Options Quality Plots Using R Further GGobi Documentation Correlation Analysis Hierarchical Correlation Principal Components Single Variable Overviews Interactive Graphics Interactive Visualisations Latticist GGobi Scatterplot Multiple Plots Brushing Other Plots Data Viewer Brushing Identify Multivariate Outliers Other Options Quality Plots Using R Further GGobi Documentation Documenting Interactive Explorations Code Review Chapter Exercises Command Summary Statistical Tests Documenting Interactive Explorations Code Review Further Resources Chapter Exercises Command Summary Models A Framework for Modelling Descriptive Analytics Predictive Analytics Documenting Models Summary Code Review Exercises Further Resources Command Summary Cluster Analysis Summary Clusters Basic Clustering Hot Spots Alternative Clustering Other Cluster Examples KMeans Export KMeans Clusters Discriminant Coordinates Plot Number of Clusters Hierarchical Clusters Other Cluster Algorithms Association Analysis Summary Overview Algorithm Usage Read Transactions file format sep cols rm.duplicates Summary Apriori data parameter appearance control Inspect Examples Video Marketing Survey Data Other Examples Resources and Further Reading Basket Analysis General Rules Network Analysis Documenting Interactive Explorations Code Review Chapter Exercises Command Summary Text Mining Application to Text Text Mining with R Decision Trees Knowledge Representation Search Heuristic Measures Tutorial Example Rattle R Tuning Parameters Min Split (Rarg[]minsplit) Min Bucket (minbucket) Priors (prior) Loss Matrix Complexity (cp) Other Options Simple Example Convert Tree to Rules Predicting Salary Group Issues Summary Code Review Iris Wine Exercises Resources Command Summary Random Forests Formalities Tutorial Example Tuning Parameters Number of Trees Sample Size Number of Variables Summary Overview Algorithm Usage Random Forest importance classwt Examples Resources and Further Reading Summary Overview Example Algorithm Resources and Further Reading Boosting Formalities Tutorial Example Tuning Parameters Summary Overview AdaBoost Algorithm Examples Step by Step Using gbm Extensions and Variations Alternating Decision Tree Resources and Further Reading Documenting Code Review Further Resources Chapter Exercises Command Summary Bootstrapping Summary Usage Further Information Summary Overview Example Algorithm Resources and Further Reading Bagging Support Vector Machine Formalities Tutorial Example Tuning parameters Examples Resources and Further Reading Overview Examples Resources and Further Reading Linear Regression Linear Regression Formalities Tutorial Example Tuning parameters Generalized Regression Formalities Tutorial Example Tuning parameters Logistic Regression Formalities Tutorial Example Tuning parameters Discussion Probit Regression Formalities Tutorial Example Tuning Parameters Multinomial Regression Formalities Tutorial Example Tuning Parameters Neural Network Formalities Tutorial Example Tuning parameters Documenting Code Review Further Resources Chapter Exercises Command Summary Naive Bayes Summary Code Review Resources Exercises Command Summary Survival Analysis Sample Data Simple Lung Descriptive Analysis Regression survreg Simple Lung coxph Simple Lung Apply to New Data More Input Variables Decision Tree Example from Singer and Willett Other Approaches Design Package Random Survival Forests Prediction on Test Data Evaluation The Evaluate Tab Confusion Matrix Measures Graphical Measures Risk Charts Cost Curves Lift ROC Curves Area Under Curve Precision versus Recall Sensitivity versus Specificity Predicted versus Observed Scoring Documenting Interactive Explorations Code Review Chapter Exercises Command Summary Transforming Data Rescale Data Recenter Scale [0,1] Rank Median/MAD Peer Relativity Profiling Index Impute Zero/Missing Mean/Median/Mode Constant Remap Binning Indicator Variables Join Categorics Math Transforms Outliers Cleanup Delete Ignored Delete Selected Delete Missing Delete Obs with Missing Other Transformations Removing Duplicates Command Summary Deployment Documenting Deployment Code Review Chapter Exercises Command Summary Troubleshooting Cairo A factor has new levels Issues Model Selection Overfitting Imbalanced Classification Sampling Cost Based Learning Model Deployment and Interoperability SQL PMML XML for Data Bibliographic Notes Documenting Code Review Chapter Exercises Command Summary Moving into R Interacting with R Basic Command Line Windows, Icons, Mouse, Pointer--WIMP The Current Rattle State Samples Projects The Rattle Log Further Tuning Models Emacs and ESS Documenting Code Review Chapter Exercises Command Summary R Evaluation Exercises Assignment Libraries and Packages Searching for Objects Package Management Information About a Package Testing Package Availability Packages and Namespaces Basic Programming in R Principles Folders and Files Flow Control If Statement For Loop Functions Apply Methods Objects System Running System Commands System Parameters Misc Internet Memory Management Memory Usage Garbage Collection Errors Frivolous Sudoku Further Resources Using R Specific Purposes Survey Analysis Getting Help R Documentation Data Data Types Numbers Strings Building Strings Splitting Strings Substitution Trim Whitespace Evaluating Strings Logical Dates and Times Space Data Structures Vectors Arrays Lists Sets Matricies Exercises Data Frames Accessing Columns Removing Columns Exercises General Manipulation Factors Elements Rows and Columns Finding Index of Elements Partitions Head and Tail Reverse a List Sorting Unique Values Loading Data Interactive Responses Interactive Data Entry Available Datasets The Iris Dataset CSV Data Used In The Book The Wine Dataset The Cardiac Arrhythmia Dataset The Adult Survey Dataset Foreign Formats Stata Data Conversions Reading Variable Width Data Saving Data Formatted Output Automatically Generate Filenames Reading a Large File Manipulating Data Manipulating Data As SQL Using SQLite ODBC Data Database Connection Excel Access Clipboard Data Spatial Data Simple Map A Density Map Overlays and Point in Polygon Other Data Formats Fixed Width Data Global Positioning System Documenting a Dataset Common Data Problems Graphics in R Basic Plot Controlling Axes Arrow Axes Legends and Points Tables Within Plots Colour Labels in Plots Axis Labels Legend Labels Within Plots Maths in Labels Multiple Plots MatPlot Multiple Plots Using ggplot2 Using GGPlot Networks Symbols Other Graphic Elements Making an Animation Animated Mandelbrot Adding a Logo to a Graphic Graphics Devices Setup Screen Devices Multiple Devices File Devices Multiple Plots Copy and Print Devices Graphics Parameters Plotting Region Locating Points on a Plot Scientific Notation and Plots Understanding Data Single Variable Overviews Textual Summaries Multiple Line Plots Separate Line Plots Pie Chart Fan Plot Stem and Leaf Plots Histogram Barplot Trellis Histogram Histogram Uneven Distribution Bump Chart Density Plot Basic Histogram Basic Histogram with Density Curve Practical Histogram Multiple Variable Overviews Scatterplot Scatterplot with Marginal Histograms Multi-Dimension Scatterplot Correlation Plot Colourful Correlations Fluctuation Plot Heat Map Projection Pursuit RADVIZ Parallel Coordinates Categoric and Numeric Measuring Data Distributions Textual Summaries Boxplot Multiple Boxplots Boxplot by Class Tuning a Boxplot Boxplot Using Lattice Boxplot Using ggplot Violin Plot What Distribution Miscellaneous Plots Line and Point Plots Matrix Data Multiple Plots Aligned Plots Probability Scale Network Plot Sunflower Plot Stairs Plot Graphing Means and Error Bars Bar Charts With Segments Bar Plot With Means 3d Bar Plot Stacks Versus Lines Multi-Line Title Mathematics Plots for Normality Basic Bar Chart Bar Chart Displays Multiple Dot Plots Alternative Multiple Dot Plots 3D Plot Clustered Box Plot Perspective Plots Star Plot Residuals Plot Waterfall Plots Dates and Times Simple Time Series Multiple Time Series Plot Time Series Plot Time Series with Axis Labels Grouping Time Series for Box Plot Time Series Heatmap Textual Summaries Stem and Leaf Plots Histogram Barplot Density Plot Basic Histogram Basic Histogram with Density Curve Practical Histogram Correlation Plot Colourful Correlations Measuring Data Distributions Textual Summaries Boxplot Multiple Boxplots Boxplot by Class Box and Whisker Plot Box and Whisker Plot Clustered Box Plot Further Resources Map Displays Further Resources Preparing Data Data Selection and Extraction Training and Test Datasets Data Cleaning Review Data Selectively Changing Vector Values Replace Indices By Names Missing Values Remove Levels from a Factor Variable Manipulations Remove Columns Reorder Columns Remove Non-Numeric Columns Remove Variables with no Variance Cleaning the Wine Dataset Cleaning the Cardiac Dataset Cleaning the Survey Dataset Imputation Nearest Neighbours Multiple Imputation Data Linking Simple Linking Record Linkage Data Transformation Aggregation Sum of Columns Pivot Tables Normalising Data Binning Interpolation Variable Selection Classification Classification Classification Issues Incremental or Online Modelling Model Tuning Tuning rpart Unbalanced Classification Building Models Temporal Analysis Evaluation Basics Basic Measures Cross Validation Graphical Performance Measures Lift The ROC Curve Other Examples 10 Fold Cross Validation Area Under Curve Calibration Curves Reporting Generating Open Document Format Getting Started with odfWeave OpenOffice.org Macro Support Generating HTML Generating PDF with LATEX Configuration Figure Sizes Fraud Analysis Archetype Analysis

Risk Charts

ACTUALLY KNOWN AS CUMULATIVE GAIN CHARTS. PERHAPS LABEL SHOULD BE ``Gain'', NOT ``Risk''.

We have mentioned in passing, particularly in the previous chapter, the idea of a risk chart for evaluating the performance of our models. A risk chart is somewhat similar in concept to a number of the other evaluation approaches, particularly the ROC curves, which have also been mentioned and will be covered in detail in Chapter 22. We formally introduce the risk chart here (rather than in Chapter 22) in order to be able to discuss the model builders in practise, and in particular illustrate their performance. The risk charts can be displayed within Rattle, once we have built our models, by choosing the Risk option of the Evaluate tab.

A risk chart is particularly useful in the context of the audit dataset, and for risk analysis tasks in general. Already we have noted that this dataset has a two class target variable, Adjusted. We have also identified a so called risk variable, Adjustment, which is a measure of the size of the risk associated with each observation. Entities that have no adjustment following an audit (i.e., they are clients who have supplied the correct information) will of course have no risk associated with them (Adjustment $=0$). Entities that do have an adjustment will have a risk associated with them, and for convenience we simply identify the value of the adjustment as the magnitude of the risk

In particular, we can think of revenue (or tax) authorities, where the outcomes of audits include a dollar amount by which the tax obligation of the taxpayer has been changed (which may be a change in favour of the revenue authority or in favour of the taxpayer). For fraud investigations, the outcome might be the dollar amount recovered from the fraudster. In these situations it is often useful to see the tradeoff between the return on investment and the number of cases investigated.

Rattle introduces the idea of a risk chart to evaluate the performance of a model in the context of risk analysis.

A risk chart plots performance against caseload. Suppose we had a population of just 100 entities (audit cases). The case load is the percentage of these cases that we will actually ask our auditors to process. The remainder we will not consider any further, expecting them to be low risk, and hence, with limited resources, not requiring any action. The decision as to what percentage of cases are actually actioned corresponds to the X axis of the risk chart - the caseload. A 100% caseload indicates that we will action all audit cases. A 25% caseload indicates that we will action just one quarter of all cases.

For a given testing population we know how many cases resulted in adjustments. We also know the magnitude of those adjustments (the risk). For a population of 100 cases, if we were to randomly choose 50% of the cases for actioning, we might expect to recover just 50% of the cases that actually did require an adjustment and 50% of the risk (or in this case, the revenue) recovered from these adjusted cases. Similarly for every caseload value: for a random caseload of 25% of the population we might expect to recover 25% of the adjustments and revenue. The diagonal black line of the risk chart represents this random selection, and can be thought of as the baseline against which to compare the performance of our models.

Any model that Rattle builds in the two-class paradigm generates a risk score for each observation. This is generally the probably of the case requiring an adjustment. We can use this score to sort all of the cases in decreasing order of the score. In selecting cases to be actioned, we then start with those cases that have the highest score. Thus, in evaluating the performance of our model, the caseload axis represents the sorted list of cases, from the highest scored cases at the left (starting at 0% of the cases actioned), and the lowest at the right (ending with 100% of the cases actioned).

The green (dashed) line of a risk chart then records the percentage of adjusted cases that we would actually expect to identify for a given percentage of the caseload. In the risk chart above, for example, we can see that if we only actioned the top 20% of the cases, as scored by our model, we recover almost 60% of the cases that actually did require an adjustment. Similarly, for an 80% caseload we can recover essentially all of the cases that required adjustment. Hence our model can save us 20% of our caseload (i.e., 20% of our resources) whilst pretty much recovering all of the adjustments.

The red (solid) line similarly records the percentage of the total revenue (or risk) that is recovered for any particular caseload. In our example above we see that the red and green lines essentially follow each other. This is not always the case.

EXPLAIN THE REVENUE AND STRIKE RATES. EXPLAIN THE AUC.

A perfect model performance assessed through a risk chart is then a risk chart that maximises the area under the two curves. We can illustrate such a risk chart by plotting the performance of a random forest model on the training data where a random forest often performs ``perfectly,'' as illustrated in this risk chart. The strike rate over the whole dataset in this case is 24% (as annotated at the right hand end of the strike rate line). That is, only 24% of all of the cases in this dataset of audit cases actually required an adjustment. The perfect model accurately identifies all 24% of the cases (and 24$ of the risk) with 24% of the caseload! Thus we see the performance plot having just two components: the essentially linear progression from a caseload of 0% and performance of 0% up to a caseload of 24% and performance of 100%, and then the flat 100% performance as we increase the caseload to 100%.

When a Risk Chart is generated the text window in Rattle will display the aggregated data that is used to construct the plot. This data consists of a row for each level of the probability distribution that is output from the model, ordered from the lowest probability value to a value of 1. For each row we record the model performance in terms of predicting a class of 1 if the probability cutoff was set to the corresponding value.

For example, we might choose a cutoff to be a probability of 0.28 so that anything predicted to be in class 1 with a probability of 0.28 or more will be regarded as in class 1. Then the number of predicted positives (or the Caseload) will be 30% (0.301667) of all cases. Amongst this 30% of cases are 69% of all true positives and they account for 79% of the total of the risk scores. The strike rate (number of true positives amongst the positives predicted by the model) is 61%. Finally, the measure reports the sum of the distances of the risk and recall from the baseline (the diagonal line). This measure can indicate the optimal caseload in terms of maximising both risk recovery and recall.

Like ROC curves, we need to be careful in the use of risk charts for selecting cut points. We note that the area under the ROC curve is a good measure of predictive discrimination, but in comparing models it may not be sensitive enough.

Support further development through the purchase of the PDF version of the book.
The PDF version is a formatted comprehensive draft book (with over 800 pages).
Brought to you by Togaware. This page generated: Sunday, 22 August 2010