Quantitative analysis of a private tax rulings corpus

We present quantitative analysis of a large, novel corpus of tax-related documents. The corpus contains more than 360 thousands of private tax rulings and hasn’t been described previously in the literature. We use text mining and graph analysis methods to analyze the corpus and to answer the specific question about the most important VAT cases of the Court of Justice of the European Union (CJEU). The results presented in the paper, although preliminary, already have significant practical value.