Weka stands for Waikato Environment for Knowledge Analysis. It is a Java-based tool for data mining and machine learning built by the University of Waikato. The mining algorithms can directly be applied onto the data sets or can be run from the Java code. It has tools for data preprocessing, regression, clustering, classification, and many other techniques with a capability to visualize.
Visit http://www.cs.waikato.ac.nz/~ml/weka/index.html for more details.
Key features of Weka are listed below:
49 data preprocessing tools
76 classification/regression algorithms
Eight clustering algorithms
Three algorithms for finding association rules
15 attribute/subset evaluators and 10 search algorithms for feature selection
The tool primarily has three user interfaces:
Explorer
Experimenter
KnowledgeFlow

The following figure shows Knowledge Explorer user interface from where navigation to various functions for preprocessing, classification, clustering, association rules, and visualizations is available...