Weka is a collection of machine learning algorithms for data mining tasks. The algorithms can either be applied directly to a dataset or called from your own Java code. Weka contains tools for data pre-processing, classification, regression, clustering, association rules, and visualization. It is also well-suited for developing new machine learning schemes.
Arff does not support multi-label data and multi-label classification. The only way you can currently do this in Weka is to have one ARFF file for each class value, predicting each class against the rest.
Source: Weka FAQ