WEKA is a popular open-source machine learning software suite written in Java that contains tools for data pre-processing, classification, regression, clustering, visualization, and feature selection. It has a graphical user interface that allows quick setup and operation, and works with data represented as flat files where each data point is described by a fixed number of attributes. WEKA's algorithms can be directly applied to datasets or called from Java code, and it is portable across platforms since it is implemented in Java.