The special volume on Non-standard Data Sources in Corpus-based Research was compiled after the NOSDAC workshop held at August 31st, 2012 at the University of Cologne, Germany. The presentationsdiscussed a number of challenges such as spelling variation, encoding, representation and annotation in handling historical, internet and SMS data for linguistic research. This volume aims to register the work carried out in the workshop as well as to provide a snapshot of the research that has been done in this area. The volume is composed of two sections: the first one contains full-papers that serve as the proceedings of the NOSDAC workshop and the second part contains short papers on linguistic resources (e.g. corpora, taggers and tools) developed for non-standard language.