Index

A

• Active donor
  • about / How it works..., How to do it...
• aggregate
  • used, for creating classification tree financial summary / Creating a classification tree financial summary using aggregate and an Excel Export node, How to do it..., How it works...
• aggregation
  • flag variables, creating for / Creating flag variables for aggregation, How to do it..., There's more...
• allbutfirst function / How to do it...
• analytical results
  • specific objectives, connecting to / Connecting specific objectives to analytical results
• ANN
  • about / Building iterative Neural Network forecasts
• Association Rules
  • used, for interaction detection / feature creation / Using Association Rules for interaction detection/feature creation, How to do it..., How it works..., There's more...
• assumptions
  • reviewing / Reviewing requirements, assumptions, and constraints
• AUC / Success criteria for classification
• Auto Classifier
  • balancing, evaluating with / Evaluating balancing with Auto Classifier, How to do it..., How it works...
  • used, to tune models / Using Auto Classifier to tune models, How to do it..., How it works...

B

• bagged logistic regression models
  • creating / Creating bagged logistic regression models, How to do it..., How it works..., There's more...
• balancing
  • evaluating, with Auto Classifier / Evaluating balancing with Auto Classifier, How to do it..., How it works...
• benefits
  • evaluating / Evaluating costs and benefits
• BI (Business Intelligence) / Automating HTML reports and graphs
• bootstrap sample
  • creating / Creating a bootstrap sample, How to do it..., How it works..., There's more...
• business goals
  • about / Understanding the goals of the business
  • commercial goals / Understanding the goals of the business
  • service goals / Understanding the goals of the business
  • scientific goals / Understanding the goals of the business
• business objectives
  • about / Define business objectives by Tom Khabaza
  • in data mining / The importance of business objectives in data mining
  • business goals, understanding / Understanding the goals of the business
  • client objectives, understanding / Understanding the objectives of your client
  • specific objectives, connecting to analytical results / Connecting specific objectives to analytical results
  • data mining goals, specifying / Specifying your data mining goals
• business understanding
  • about / Introduction, Business understanding

C

• Cartesian product merge
  • performing / Cartesian product merge using key-less merge by key, How to do it..., How it works...
  • performing, dummy keys used / Multiplying out using Cartesian product merge, user source, and derive dummy, How to do it..., There's more...
• categorical values
  • grouping / Grouping categorical values, How to do it..., There's more...
• CDF (cumulative distribution function) / Rolling your own modeling algorithm – Weibull analysis
• CHAID modeling Node
  • used, for selecting variables / Selecting variables using the CHAID Modeling node, How to do it..., How it works..., There's more...
• CHAID stumps
  • using / Using CHAID stumps when interviewing an SME, How to do it..., How it works...
• champion/challenger model
  • implementing / Implementing champion/challenger model management, How to do it..., How it works..., Script section 2
• classification tree financial summary
  • creating, aggregate used / Creating a classification tree financial summary using aggregate and an Excel Export node, How to do it..., How it works...
  • creating, Excel Export node used / Creating a classification tree financial summary using aggregate and an Excel Export node, How to do it..., How it works...
• classification trees
  • used, to explore predictions of Neural Net / Using classification trees to explore the predictions of a Neural Network, How to do it..., How it works...
• clean downstream
  • performing, Filter node used / Performing clean downstream of a calculation using a Filter node
• CLEM expression language
  • functions / How it works...
• CLEM scripting
  • about / Introduction
  • best practices / CLEM scripting best practices
  • shortcomings / CLEM scripting shortcomings
• cluster centers
  • writing, to Excel for conditional formatting / Using aggregate to write cluster centers to Excel for conditional formatting, How to do it..., How it works...
• commercial goals / Understanding the goals of the business
• conditional formatting
  • cluster centers, writing to Excel for / Using aggregate to write cluster centers to Excel for conditional formatting, How to do it..., How it works...
• constraints
  • reviewing / Reviewing requirements, assumptions, and constraints
• contingencies
  • defining / Identifying risks and defining contingencies
• conversion rates
  • calculating / Calculating and comparing conversion rates, How to do it..., How it works..., There's more..., See also
  • comparing / Calculating and comparing conversion rates, How to do it..., How it works..., There's more..., See also
• correlation matrices
  • used, for removing redundant variables / Removing redundant variables using correlation matrices, How to do it..., How it works...
• costs
  • evaluating / Evaluating costs and benefits

D

• data
  • reformatting, for reporting with Transpose node / Reformatting data for reporting with a Transpose node, How to do it..., How it works..., There's more...
  • preparing / Data preparation
• data integration
  • missing data, evaluating / Running a Statistics node on anti-join to evaluate the potential missing data, Getting ready, How to do it..., How it works...
• Data miners
  • about / Introduction
• data mining goals
  • specifying / Specifying your data mining goals
• data understanding
  • about / Data understanding
• date arithmetic
  • performing / How to do it..., How it works...
• datetime_date() function / How it works...
• deployment
  • about / Deployment
• Derive Count nodes / How it works...
• Derive node
  • functions, nesting into / Nesting functions into one Derive node
• Derive State nodes / How it works...
• Dorian Pyle / Using the Feature Selection node creatively to remove or decapitate perfect predictors
• dummy keys
  • used, for performing Cartesian product merge / Multiplying out using Cartesian product merge, user source, and derive dummy, How to do it..., There's more...

E

• else branch
  • about / How it works...
• elseif branch
  • about / How it works...
• empty aggregate
  • used, to evaluate sample size / Using an empty aggregate to evaluate sample size , How to do it..., How it works..., A modified version
• evaluation
  • about / Evaluation
• Excel Export node
  • used, for creating classification tree financial summary / Creating a classification tree financial summary using aggregate and an Excel Export node, How to do it..., How it works...

F

• >< function / How it works...
• @FIELD function / How it works...
• Feature Selection node
  • used, for detecting model instability / Detecting potential model instability early using the Partition node and Feature Selection node, How to do it..., How it works...
  • used, to remove perfect predictors / Using the Feature Selection node creatively to remove or decapitate perfect predictors, How to do it..., There's more...
• field formatting
  • changing, in Table node / Changing formatting of fields in a Table node, How to do it..., How it works..., There's more...
• Filter node
  • used, for performing clean downstream / Performing clean downstream of a calculation using a Filter node
• First time donor
  • about / How it works..., How to do it...
• flag variables
  • creating, for aggregation / Creating flag variables for aggregation, How to do it..., There's more...
• full data model
  • used, to address missing data / Using a full data model/partial data model approach to address missing data, How to do it..., How it works...
• functions
  • nesting, into one Derive node / Nesting functions into one Derive node

G

• @GLOBAL variable / Imputing in-stream mean or median
• generated filters
  • combining / Combining generated filters, How to do it..., How it works...
• graphs
  • automating / Automating HTML reports and graphs, How to do it..., Script section 1, Script section 2
• Grow button / How to do it...

H

• high skew variable
  • transforming, with multiple Derive node / Transforming high skew and kurtosis variables with a multiple Derive node, How to do it..., How it works..., There's more...
• HTML reports
  • automating / Automating HTML reports and graphs, How to do it..., Script section 1, Script section 2

I

• IBM SPSS Modeler / Introduction
• if branch
  • about / How it works...
• imbalanced target variable
  • prior probabilities, incorporating for / How to do it..., How it works..., There's more...
• in-stream mean
  • imputing / Imputing in-stream mean or median, How to do it..., How it works..., There's more...
• Inactive donor
  • about / How it works..., How to do it...
• initial data
  • sampling from / Evaluating the need to sample from the initial data, Getting ready, How to do it..., There's more...
• interaction detection / feature creation
  • Association Rules, using for / Using Association Rules for interaction detection/feature creation, How to do it..., How it works..., There's more...
• intof() function / How it works...
• iterative Neural network forecasts
  • building / Building iterative Neural Network forecasts, How to do it..., How it works..., There's more...

J

• jackknife method
  • Outliers, detecting with / Detecting outliers with the jackknife method, How to do it..., How it works..., Script section 1, Script section 2, Script section 3

K

• K-means cluster
  • using, as alternative to anomaly detection / Using a single cluster K-means as an alternative to anomaly detection, How to do it..., How it works...
• K-means clustering
  • about / Optimizing K-means cluster solutions
• K-means cluster solutions
  • optimizing / Optimizing K-means cluster solutions, How to do it..., How it works..., Script section 1, Script section 3, There's more...
• KNN
  • used, to match similar cases / Using KNN to match similar cases, How to do it..., How it works...
• KPIs
  • about / Understanding the objectives of your client
• kurtosis variable
  • transforming, with multiple Derive node / Transforming high skew and kurtosis variables with a multiple Derive node, How to do it..., How it works..., There's more...

L

• @LAST_NON_BLANK field / How it works...
• Lapsing donor
  • about / How it works..., How to do it...
• large datasets
  • offers for / Next-Best-Offer for large datasets, How to do it..., How it works..., There's more...
• locchar function / How to do it...
• log transform / How it works...
• look-up table
  • merging / Merging a lookup table, How to do it..., How it works...

M

• mean absolution percent error (MAPE) / Implementing champion/challenger model management
• Mean button / There's more...
• Means node
  • used, for selecting variables / Selecting variables using the Means node, How to do it..., How it works..., There's more...
• Mean Squared Error (MSE) / Success criteria for estimation
• median
  • imputing / Imputing in-stream mean or median, How to do it..., How it works..., There's more...
• Merge node
  • speeding up, cache used / Speeding up merge with caching and optimization settings, How to do it..., How it works...
  • speeding up, optimization settings used / Speeding up merge with caching and optimization settings, How to do it..., How it works...
• missing data
  • exploring, @NULL multiple Derive used / Using an @NULL multiple Derive to explore missing data, How to do it..., How it works...
  • evaluating, during data integration / Running a Statistics node on anti-join to evaluate the potential missing data, Getting ready, How to do it..., How it works...
  • addressing, for binning scale variables / Binning scale variables to address missing data, How to do it..., How it works...
  • addressing, full data model used / Using a full data model/partial data model approach to address missing data, How to do it..., How it works...
  • addressing, partial data model used / Using a full data model/partial data model approach to address missing data, How to do it..., How it works...
• missing values
  • imputing, from uniform distribution / Imputing missing values randomly from uniform or normal distributions, How to do it..., There's more...
  • imputing, from normal distribution / Imputing missing values randomly from uniform or normal distributions, How to do it..., There's more...
• Modeling
  • about / Introduction, Modeling
• model instability
  • detecting, Partition node used / Detecting potential model instability early using the Partition node and Feature Selection node, How to do it..., How it works...
  • detecting, Feature Selection node used / Detecting potential model instability early using the Partition node and Feature Selection node, How to do it..., How it works...
• models
  • building, with outliers / Building models with and without outliers, How to do it..., How it works...
  • building, without outliers / Building models with and without outliers, How to do it..., How it works...
  • tuning, Auto Classifier used / Using Auto Classifier to tune models, How to do it..., How it works...
• Monte Carlo Simulation
  • variable importance, quantifying with / Quantifying variable importance with Monte Carlo simulation, How to do it..., How it works..., Script section 2, There's more...
• MTTF (mean time to failure) / Rolling your own modeling algorithm – Weibull analysis
• multiple Derive nodes
  • transformations, building with / Building transformations with multiple Derive nodes, How it works..., There's more...
  • high skew variable, transforming with / Transforming high skew and kurtosis variables with a multiple Derive node, How to do it..., How it works..., There's more...
  • kurtosis variable, transforming with / Transforming high skew and kurtosis variables with a multiple Derive node, How to do it..., How it works..., There's more...

N

• @NULL multiple Derive
  • used, to explore missing data / Using an @NULL multiple Derive to explore missing data, How to do it..., How it works...
• Neural Net predictions
  • classification trees, using / Using classification trees to explore the predictions of a Neural Network, How to do it..., How it works...
• neural network
  • used, for searching similar records / Searching for similar records using a Neural Network for inexact matching, How to do it..., How it works...
• Neural Network Feature Selection
  • performing / Using Neural Network for Feature Selection, How to do it..., How it works...
• neuro-fuzzy searching
  • used, to find similar names / Using neuro-fuzzy searching to find similar names, How to do it..., There's more...
  • about / Using neuro-fuzzy searching to find similar names
• New donor
  • about / How it works..., How to do it...
• non-standard aggregation
  • performing / Shuffle-down (nonstandard aggregation), How to do it..., How it works...
• nonstandard dates
  • parsing / Parsing nonstandard dates, How to do it..., How it works...
• normal distribution
  • missing values, imputing from / Imputing missing values randomly from uniform or normal distributions, How to do it..., There's more...

O

• OK button / How to do it..., How to do it..., How to do it...
• outlier report
  • creating / How to do it..., How it works...
• Outliers
  • detecting, with jackknife method / Detecting outliers with the jackknife method, How to do it..., How it works..., Script section 1, Script section 2, Script section 3
• outliers
  • models, building with / Building models with and without outliers, How to do it..., How it works...
  • models, building without / Building models with and without outliers, How to do it..., How it works...

P

• parameters
  • used, in calculations / Using parameters instead of constants in calculations
• partial data model
  • used, to address missing data / Using a full data model/partial data model approach to address missing data, How to do it..., How it works...
• Partition node
  • used, for detecting model instability / Detecting potential model instability early using the Partition node and Feature Selection node, How to do it..., How it works...
• PCC / Success criteria for classification
• Percent Correct Classification (PCC) / There's more...
• perfect predictors
  • removing, Feature Selection node used / Using the Feature Selection node creatively to remove or decapitate perfect predictors, How to do it..., There's more...
• Predictors… button / How to do it...
• Preview button / How to do it...
• prior probabilities
  • incorporating, for imbalanced target variable / How to do it..., How it works..., There's more...

Q

• quirk reports / Creating an Outlier report to give to SMEs

R

• random() / How it works...
• random()function / There's more...
• random imputation
  • used, to match variable distribution / Using random imputation to match a variable's distribution, How to do it..., How it works..., There's more...
• Read Values button / How to do it..., How to do it..., How to do it..., How to do it...
• Read Vaues button / How to do it...
• Records field / How to do it...
• redundant variables
  • removing, correlation matrices used / Removing redundant variables using correlation matrices, How to do it..., How it works...
• requirements
  • reviewing / Reviewing requirements, assumptions, and constraints
• resources inventory
  • taking / Taking inventory of resources
• Return On Investment (ROI) / Other customized success criteria
• risks
  • identifying / Identifying risks and defining contingencies
• Run button / How to do it..., How to do it...

S

• sample size
  • evaluating, empty aggregate used / Using an empty aggregate to evaluate sample size , How to do it..., How it works..., A modified version
• sampling
  • need for, evaluating / Evaluating the use of sampling for speed, How to do it..., How it works..., There's more...
• scale variables
  • binning, to address missing data / Binning scale variables to address missing data, How to do it..., How it works...
• scientific goals / Understanding the goals of the business
• Select button / How to do it..., How to do it...
• Sequence function @DIFF1 / How it works...
• Sequence function @OFFSET / How it works...
• Sequence function @SINCE / How it works...
• sequence processing
  • using / Sequence processing, How to do it...
  • working / How it works..., There's more...
• service goals / Understanding the goals of the business
• similar cases
  • matching, KNN used / Using KNN to match similar cases, How to do it..., How it works...
• similar names
  • finding, neuro-fuzzy searching used / Using neuro-fuzzy searching to find similar names, How to do it..., There's more...
• similar records
  • searching, neural network used / Searching for similar records using a Neural Network for inexact matching, How to do it..., How it works...
• Single-Antecedent association rules
  • used, for selecting variables / Selecting variables using single-antecedent Association Rules, How to do it..., How it works..., There's more...
• Soundex codes
  • producing / Producing longer Soundex codes, How to do it..., How it works...
• specific objectives
  • connecting, to analytical results / Connecting specific objectives to analytical results
• Star donor
  • about / How it works..., How to do it...
• Subject Matter Experts (SMEs) / Introduction
• substring() function / How it works...
• substring_between() function / There's more...
• success criteria
  • for classification / Success criteria for classification
  • for estimation / Success criteria for estimation
• SuperNode button / How to do it...
• Support Vector Machines (SVMs) / Using Auto Classifier to tune models

T

• Table node
  • field formatting, changing / Changing formatting of fields in a Table node, How to do it..., How it works..., There's more...
• target variables
  • specifying / The key to the translation – specifying target variables
• terminology
  • defining / Defining terminology
• time-aligned cohorts
  • creating / Creating time-aligned cohorts, How to do it..., How it works..., There's more...
• TIMELAG_null variable / How it works...
• time series forecasts
  • automating / Automating time series forecasts, How to do it..., How it works..., Script section 2, There's more...
• timestamp variable
  • parsing / Parsing and performing a conversion on a complex stream, How to do it..., How it works...
• Tom Khabaza
  • URL / Introduction
• to_integer() function / There's more..., How it works...
• to_string() function / How it works...
• transformations
  • building, with multiple Derive nodes / Building transformations with multiple Derive nodes, How it works..., There's more...
• Transpose node
  • data reformatting, for reporting with / Reformatting data for reporting with a Transpose node, How to do it..., How it works..., There's more...

U

• uniform distribution
  • missing values, imputing from / Imputing missing values randomly from uniform or normal distributions, How to do it..., There's more...

V

• validation
  • need for / How (and why) to validate as well as test, How to do it..., How it works...
• variable distribution
  • matching, random imputation used / Using random imputation to match a variable's distribution, How to do it..., How it works..., There's more...
• variable importance
  • quantifying, with Monte Carlo Simulation / Quantifying variable importance with Monte Carlo simulation, How to do it..., How it works..., Script section 2, There's more...
• variable names
  • changing, Derive node used / Changing large numbers of variable names without scripting, How to do it..., How it works...
• variables
  • selecting, CHAID modeling Node used / Selecting variables using the CHAID Modeling node, How to do it..., There's more...
  • selecting, Means node used / Selecting variables using the Means node, How to do it..., How it works..., There's more...
  • selecting, Single-Antecedent association rules used / Selecting variables using single-antecedent Association Rules, How to do it..., How it works..., There's more...

W

• Wealth2_null variable / How it works...
• Weibull analysis
  • about / Rolling your own modeling algorithm – Weibull analysis
  • rolling / Rolling your own modeling algorithm – Weibull analysis, How to do it..., How it works..., Script section 1