Data Analytics with SPSS Modeler Course

Lesson 1 • Understanding Streams and Nodes

Explains how streams connect nodes to form executable workflows. Students build their first simple stream to reinforce conceptual understanding.

Lesson 2 • Setting Up Projects and Workspaces

Covers project creation, file organisation, and workspace preferences. Proper setup prevents errors and supports team collaboration later.

Lesson 3 • Overview of Data Analytics Concepts

Covers descriptive, predictive, and prescriptive analytics distinctions. Establishes the analytical mindset needed throughout the course.

Lesson 1 • Connecting to Relational Databases

Covers ODBC configuration and the Database source node for SQL-based imports. Students write basic SQL queries within Modeler to filter source data.

Lesson 2 • Working with Other Data Sources

Introduces IBM Cognos, SAS, and XML source nodes for diverse data environments. Broadens import skills to match real-world enterprise data landscapes.

Lesson 3 • Importing Flat Files and Spreadsheets

Teaches configuration of Variable File and Excel source nodes. Correct delimiter and encoding settings prevent downstream data errors.

Lesson 1 • Deriving and Recoding Fields

Teaches the Derive and Recode nodes to create new variables and bin continuous data. Engineered features often improve predictive model performance significantly.

Lesson 2 • Handling Missing Values

Covers detection, imputation, and exclusion of missing data using the Filler node. Proper handling prevents biased models and invalid outputs.

Lesson 3 • Aggregating and Restructuring Data

Applies Aggregate, Transpose, and Sort nodes to reshape data for analysis. Correct data structure is a prerequisite for accurate modelling and reporting.

Lesson 4 • Filtering and Selecting Records

Uses Select and Sample nodes to subset data by condition or proportion. Targeted record selection reduces noise and improves model accuracy.

Lesson 1 • Feature Importance and Selection

Applies the Feature Selection node to rank predictors by relevance to the target. Reducing irrelevant features improves model efficiency and interpretability.

Lesson 2 • Exploring Relationships Between Variables

Uses scatter plots, matrix plots, and the Means node to examine variable associations. Correlation insights inform feature selection for predictive models.

Lesson 3 • Descriptive Statistics and Data Auditing

Applies the Data Audit and Statistics nodes to profile fields systematically. Audit results guide subsequent cleaning and feature selection decisions.

Lesson 1 • Logistic Regression and Linear Models

Configures Logistic Regression and Linear Regression nodes for binary and continuous targets. Students interpret coefficients and assess model fit using standard metrics.

Lesson 2 • Model Evaluation and Selection

Uses Analysis, Evaluation, and Gains nodes to measure accuracy, lift, and ROC performance. Rigorous evaluation ensures the chosen model meets business acceptance criteria.

Lesson 3 • Neural Network and SVM Models

Applies Neural Network and Support Vector Machine nodes for complex nonlinear patterns. Students tune key hyperparameters and compare performance against simpler models.

Lesson 4 • Auto Classifier and Auto Numeric Nodes

Uses Auto Classifier and Auto Numeric to benchmark multiple algorithms simultaneously. Automated comparison accelerates model selection for classification and regression tasks.

Lesson 1 • Validating Cluster Quality

Evaluates silhouette scores, within-cluster variance, and cluster stability metrics. Validation confirms that segments are distinct, stable, and analytically useful.

Lesson 2 • K-Means Clustering Fundamentals

Configures the K-Means node, selects k, and interprets centroid-based cluster assignments. Understanding distance metrics is essential for meaningful segment creation.

Lesson 3 • Profiling and Interpreting Clusters

Uses graph and statistics nodes to profile each cluster's characteristics after assignment. Meaningful cluster labels drive actionable business segmentation strategies.

Lesson 1 • Evaluating and Deploying Forecasts

Assesses forecast accuracy using MAE, RMSE, and MAPE metrics and exports predictions. Accurate forecasts support inventory, staffing, and budget planning decisions.

Lesson 2 • Time Series Forecasting Fundamentals

Introduces the Time Series node and ARIMA/exponential smoothing models for trend forecasting. Students prepare time-indexed data and configure seasonal parameters correctly.

Lesson 3 • Sequence Detection and CARMA

Uses the Sequence and CARMA nodes to find ordered event patterns in transactional data. Sequential patterns reveal customer journey and process bottleneck insights.

Lesson 1 • Monitoring Model Performance Over Time

Tracks prediction accuracy drift and triggers model refresh cycles using performance benchmarks. Ongoing monitoring prevents silent model degradation in production environments.

Lesson 2 • Exporting Models and Results

Exports models in PMML format and results to databases, flat files, and reports. Standardised export formats enable integration with external systems and BI tools.

Lesson 3 • Automating Streams with Scripts

Uses SPSS Modeler scripting to automate stream execution and parameter updates. Automation reduces manual effort and ensures reproducible, scheduled analytics runs.

Lesson 4 • Scoring New Data with Deployed Models

Applies saved model nuggets to new datasets for batch and real-time scoring. Consistent scoring pipelines ensure predictions remain aligned with trained model logic.