When courts started publishing judgements, big data analysis within the legal domain became possible. By taking data from the European Court of Human Rights as an example, we investigate how natural language processing tools can be used to analyse texts of the court proceedings in order to automatically predict judicial decisions. With an average accuracy of 75% in predicting the violation of 9 articles of the European Convention on Human Rights our approach highlights the potential of machine learning approaches in (...) the legal domain. We show, however, that predicting decisions for future cases based on the cases from the past negatively impacts performance. Furthermore, we demonstrate that we can achieve a relatively high classification performance when predicting outcomes based only on the surnames of the judges that try the case. (shrink)

The automated identification of national implementations of European directives by text similarity techniques has shown promising preliminary results. Previous works have proposed and utilized unsupervised lexical and semantic similarity techniques based on vector space models, latent semantic analysis and topic models. However, these techniques were evaluated on a small multilingual corpus of directives and NIMs. In this paper, we utilize word and paragraph embedding models learned by shallow neural networks from a multilingual legal corpus of European directives and national legislation (...) to develop unsupervised semantic similarity systems to identify transpositions. We evaluate these models and compare their results with the previous unsupervised methods on a multilingual test corpus of 43 Directives and their corresponding NIMs. We also develop supervised machine learning models to identify transpositions and compare their performance with different feature sets. (shrink)

The automated identification of national implementations of European directives by text similarity techniques has shown promising preliminary results. Previous works have proposed and utilized unsupervised lexical and semantic similarity techniques based on vector space models, latent semantic analysis and topic models. However, these techniques were evaluated on a small multilingual corpus of directives and NIMs. In this paper, we utilize word and paragraph embedding models learned by shallow neural networks from a multilingual legal corpus of European directives and national legislation (...) to develop unsupervised semantic similarity systems to identify transpositions. We evaluate these models and compare their results with the previous unsupervised methods on a multilingual test corpus of 43 Directives and their corresponding NIMs. We also develop supervised machine learning models to identify transpositions and compare their performance with different feature sets. (shrink)

Terms of service of on-line platforms too often contain clauses that are potentially unfair to the consumer. We present an experimental study where machine learning is employed to automatically detect such potentially unfair clauses. Results show that the proposed system could provide a valuable tool for lawyers and consumers alike.

This Article explores the costs and benefits of one subset of continuous and discontinuous rules. These expressions are shown to be distinct from the familiar dichotomy expressed as standards versus rules, but they share the difficulty of dividing the world of law in two. Still, regulatory approaches that focus on discontinuities can often be made more continuous, and vice versa. A speed limit is discontinuous in the sense that one drives above or below (or within) the announced limit. But it (...) is often made more continuous—even with more discontinuities— as when the stated limit is different for various kinds of vehicles. This Article works around these definitional problems to show that law often discourages useful disclosures by encouraging parties with information to offer continuous information in order to avoid after-the-fact lawsuits when specific disclosures prove to have inaccuracies. For example, it is common to hear or be warned that a medical procedure poses the risk of death, when a better-informed doctor or hospital could have given the precise percentages attached to various outcomes. Similarly, a corporation is on safe ground when it follows “generally accepted accounting principles,” when investors would have learned more from information about good and bad outcomes put in probabilistic terms. The Article works toward the suggestion that law might create a safe harbor in which probabilistic disclosures are protected when they are, or are certified to be, more useful than the ready alternative of fairly general disclosures. (shrink)