SPSS Modeler: machine learning models V18.2 (0A079G)

• Teams using only automated modeling who want to understand what it picks and why.
• Analysts who have to justify to the business why one model and not another.
• Anyone who needs to segment with no prior label and does not know which algorithm applies.

• Introduction to machine learning models: taxonomy, measurement levels, and how models are built and applied in SPSS Modeler
• CHAID decision trees: categorical and continuous targets, mixed predictors and treatment of missing values
• C&R Tree decision trees: categorical and continuous targets, mixed predictors and missing values
• Evaluation measures for supervised models, for categorical and continuous targets
• Statistical models for continuous targets: linear regression with categorical predictors and missing values
• Statistical models for categorical targets: logistic regression with categorical predictors and missing values
• Association models: sequence detection
• Black box models: neural networks with categorical and continuous predictors
• Black box models: ensembles, boosting, bagging and combining the best models
• Unsupervised models: K-Means and Kohonen networks, with categorical inputs and missing values
• Unsupervised models: TwoStep, anomaly detection and finding the best segmentation automatically
• Association models: Apriori, evaluation measures and missing values
• Preparing data for modeling: quality, selecting important predictors and balancing

• Labs on IBM SPSS Modeler with real data: streams built, run and fixed during the course.
• Taught directly by IBM Champion instructors from the SIXE team.
• Credly badges available: credly.com/organizations/sixe/badges
• Flexible format — online, classroom or onsite — in closed groups, priced per attendee, in the language of your choice (ES / EN / FR).

Is model evaluation covered?
Yes, and it is part of the outline itself: measures for categorical targets and for continuous ones. A model with no evaluation measure cannot be defended in front of anyone.

Does it cover missing values?
In every model, one by one: what CHAID does with a missing value is not what a neural network does, and that changes the result.

Does it cover XGBoost and Random Trees?
Those two are in the advanced course, advanced models (0A039G). Here you get CHAID and C&R Tree, regression, neural networks, ensembles and the unsupervised models.

Group size and language?
Closed groups of four to ten people, in Spanish, English or French. We run it with fewer than four as well, though it usually pays to wait and fill the group.

• 0A069G — Modeler foundations
• 0A039G — advanced models
• 0A0U8G — categorical targets