IBM SPSS Modeler: Modeler’s New R Nodes

Name: IBM SPSS Modeler: Modeler’s New R Nodes
Rating: 3.2 (9 reviews)

IBM SPSS Modeler Seminar Series

Created bySandy Midili

Last updated 5/2014

English

English [Auto],

What you'll learn

Describe the new features and why they are useful
Learn more R through self study
Add new graphics capability in Modeler
Add new statistics capability in Modeler

Course content

1 section • 20 lectures • 2h 29m total length

The Organization of the Seminar3:47
Explore the seminar on IBM SPSS Modeler’s new R nodes, from downloading essentials and setup to exploring open source R integration, demonstrations, and getting started with modeling.
Overview of the R Integration Package11:56
Learn to integrate R with IBM SPSS Modeler using modular data, modeler data, and a model object to build and score predictions.
Application Example- A Weather ‘Source Node’2:00
Explore building custom nodes in IBM SPSS Modeler using a weather data example, including a user input node and an R API workflow that fetches hourly temperatures.
Writing the R Code to do GeoCoding in Modeler11:04
Geocode addresses in IBM SPSS Modeler with an R geocoding API using the Google geocoding API, add latitude and longitude, and enable distance calculations with haversine.
Building your own Custom R Node7:09
Build a custom R node in IBM SPSS Modeler using the dialog builder, map address, latitude, and longitude to an R script, and install the node for geocoding.
Building your own Modeling Node- Random Forests8:10
Build and evaluate a random forest model within Mosler. Partition data into training and testing, then review text and graphical outputs to assess variable importance and model performance.
The R model building syntax6:32
Learn to build an optimized random forest in R using the caret package, with bootstrap aggregating, cross-validation, and hold-out validation for accurate modeling.
The R scoring syntax3:13
Use the R scoring syntax to score new data with a saved model in SPSS Modeler, ensuring packages and column names, with the model stored in the stream for reuse.
Reviewing the Model Results1:47
Review the console output to see what ran, catch warnings such as a missing package, and understand how partition mode and scoring append results to data.
Building a Dialog for Random Forests12:01
Build a random forest dialogue in IBM SPSS Modeler, selecting target and predictors, with a default 300 trees, using the custom dialog builder and care package to explore other models.
Does Modeler treat an R Model like any other Model2:28
Learn whether SPSS Modeler’s new R nodes behave like other models and how to correctly configure predictor and target fields, metadata, and data partitioning for successful analysis.
String Distance Demonstration6:37
Learn how to use string distance and substring matching in R within IBM SPSS Modeler, comparing strings by keystroke differences with Levenstein distance to find references despite typos.
Iterative Neural Net Forecasts Demontration3:34
Demonstrates iterative neural net forecasts in IBM SPSS Modeler using an R node to predict multiple weeks by rolling data forward, with lagged predictors and random forest or neural options.
Questions about Writing the Neural Net Forecast Node7:46
Explain how marketing spend predicts unit demand and forecast with predictors using neural net nodes in Modeler. Compare legacy scripting to modern approaches, highlighting lagging data and code effort.
Getting Started with R and Modeler19:41
Begin from scratch with R in IBM SPSS Modeler, install R essentials, and use R Studio to run basic commands and practice with datasets like iris.
Getting familiar with R using R Studio7:01
Get familiar with R Studio by starting a new R script, using the console, and exploring the global environment, memory, and vectors; install packages and use help for basic prototyping.
Basic Grammar and Commands in R9:59
Demonstrates basic grammar and commands in R, teaching simple assignment, vector operations, and arithmetic, and showing how to compute mean, standard deviation, and elementwise operations.
Matrices, Data Frames, and Models9:20
Explore matrices and data frames, from creating a 3x3 matrix and previewing the iris dataset with head to converting data files into data frames and building models with lm.
Summary Statistics and GGPlot210:27
Learn to generate descriptive statistics and rich visualizations with ggplot2 in SPSS Modeler. Use the cars and iris datasets to create density plots, histograms, and color-coded discriminant visualizations.
Cbind() and Apply()4:37
Discover how to use cbind to bind new variables to existing data frames and how to apply functions to matrices with apply, enabling row- and column-wise analyses in R.

Requirements

Knowledge or experience with IBM SPSS Modeler or completion of an introductory level data mining course and on the job data mining experience.

Description

IBM SPSS Modeler is a data mining workbench that allows you to build predictive models quickly and intuitively without programming. Analysts typically use SPSS Modeler to analyze data by mining historical data and then deploying models to generate predictions for recent (or even real-time) data.

Overview: Modeler's New R Nodes is a series of self-paced videos. This course is divided into four parts:

·What are the new Modeler R Nodes and why are they an exciting addition?

·What is R and what are some of the best ways to learn more about it?

·Adding new graphics capability with R

·Adding new statistics capability with R

We discuss one of the exiting new features of Modeler 16. We show some R functionality in the R environment itself, but the seminar will culminate in the demonstration of R capabilites in a Modeler stream. Advice will be given on how best to develop more skills in this area, but you will have some working knowledge from these videos alone.

Who this course is for:

Anyone that has experience with IBM SPSS Modeler or has completed an introductory level data mining course and would like to learn how to use Modeler and R together.