Development Windows

Weka for Windows

Prepare datasets, build classification and regression models, and explore clustering with Weka, a free, open-source machine-learning workbench for Windows.

WindowsFree and open source; GNU General Public License 3.0Version 3.8.7
Operating system
64-bit Windows
Architecture
x64 (publisher Intel Windows package)
Download package
Self-extracting EXE installer
Download Weka for Windows (x64, bundled Java)

Includes Bellsoft OpenJDK Java VM 25. Publisher lists a separate ARM Windows package. Specific Windows version support was not established for this installer.

Link checked · Download guidance

What is Weka?

Weka is a machine-learning and data-mining workbench for preparing data and comparing predictive models. It provides classification, regression, clustering and association-rule algorithms, with filters for transforming datasets before analysis. Its graphical tools suit exploring a dataset, while Java classes and command-line options support repeatable workflows.

ARFF is Weka's default dataset format and records attribute types alongside the data. CSV input is also supported, but separately loaded training and test CSV files can produce incompatible nominal labels. Keep attribute definitions consistent when evaluating a model; the publisher documents hold-out testing and cross-validation.

This entry links to the publisher's stable 3.8.7 Windows x64 installer, which bundles Bellsoft OpenJDK 25. Program Menu shortcuts use that bundled JVM. Windows on ARM has a separate publisher package; the linked download here is the Intel/x64 option. Additional Weka packages can extend functionality, but downloading and installing them requires an internet connection.

What it offers

  • Prepare and transform datasets with supervised and unsupervised filters.
  • Build classification and regression models, or explore clustering and association rules.
  • Evaluate classifiers with a separate test set or cross-validation.
  • Load ARFF datasets and import CSV data.
  • Use graphical tools or Java command-line classes; extend functionality through the package manager.

Good to know

  • Separate CSV training and test files may infer different nominal labels or label ordering and become incompatible.
  • Downloading and installing additional packages requires an internet connection.
  • Distributed derivative works using Weka under its GPL license must also meet the GPL distribution terms.

Product images

Sources

References used for this listing. Retrieval dates do not indicate hands-on testing.

  1. Downloading and installing Weka - Weka Wiki ↗Retrieved
  2. Primer - Weka Wiki ↗Retrieved
  3. Use csv files - Weka Wiki ↗Retrieved
  4. Commercial applications - Weka Wiki ↗Retrieved
  5. Documentation - Weka Wiki ↗Retrieved
  6. Weka - Download ↗Retrieved

Browser storage

No analytics or advertising cookies. If you dismiss this notice, we remember its acknowledgement in your browser for 180 days. How storage works