Back to top

Lexalyze - Interdisciplinary Research Program

Last modified Oct 6, 2019

White Paper on LEXIA Prototype

Motivation

Interestingly, the sphere of law and jurisprudence has been rather skeptical of the digital revolution. While legal science has established legal informatics as a field of research and even if there have been attempts to undertake projects in legal science by employing the means of informatics, the digital revolution has had very little impact on legislation, jurisprudence and contract design. Digital applications only serve as helpful instruments – legal search engines being the most obvious example –, not as transformative agents.

 

Potential of Undetected Synergies

The tools of informatics seem to be highly suited to dealing with issues that arise in the course of legal affairs: The legal system consists of and is confronted with large bodies of complex data that follow rational and objective principles (at least in an ideal world). As far as the legal system itself is concerned, one cannot help the impression that there is a hopeless amount of overregulation; a lot of times, an excessive number of legal norms seem to do nothing but create redundancies and inefficacies. Informatics, on the other hand, is a science that promises to help solve this problem as it has brought about new insights on how to effectively organize und use complex data. Algorithms with excellent precision can perform tasks that involve a lot of data without much effort; in former times, comparable tasks could only be fulfilled by employing knowledgeable people for a long time.

Problem Statement

The increasing complexity of legal systems, respectively texts, has many origins, which are worth a deeper analysis (Katz, 2013; Schuck, 1992). This project is an attempt to unveil the complexity in legal texts driven by structural, lexical and syntactical properties. Thereby we use established quantitative methods from structural network analysis and linguistics into the domain of legal text analysis. Those methods will provide quantitative measurements in order to objectively measure and compare different texts as well as their evolution in time. Therefore, this project aims in a first step, to provide a set of metrics, enabling a structured and objective comparison of legal texts regarding their complexity and understandability.

Evolution: 25 years of the Corporation Stock Act (AktG)

 

Interdisciplinarity and Call for Participation

Since the problems addressed in this project cover a broad spectrum of research areas, the interdisciplinary research group consists of computer scientists, legal scientists, lawyers, and philosophers.

The Lexalyze project follows a dynamical progression and is open to cooperations with suitable partners from academia and industry. There are possibilities of performing a doctoral research (PhD) in connection with the project Lexalyze.

Research Activities

2019
[Gl19a]Glaser, I; Landthaler, J.; Matthes,F.: Supporting the Legal Reasoning Process by Classification of Judgments Applying Active Machine Learning, IRIS: Internationales Rechtsinformatik Symposium, Salzburg, Austria, 2019
[La19a]Landthaler, J.; Glaser, I; Lecker, H.; Matthes,F.: User Study on Selection Search and Semantic Text Matching in Tenancy Law, IRIS: Internationales Rechtsinformatik Symposium, Salzburg, Austria, 2019
2018
[Gl18b]Glaser, I.; Scepankova, E.; Matthes, F.: Classifying Semantic Types of Legal Sentences: Portability of Machine Learning Models, Jurix: International Conference on Legal Knowledge and Information Systems, Groningen, Netherlands, 2018
[La18c]Landthaler, J.; Glaser, I.; Matthes, F.: Explainable Semantic Text Matching, Jurix: International Conference on Legal Knowledge and Information Systems, Groningen, Netherlands
[Wa18c]Waltl, B.; Bonczek, G.; Scepankova, E.; Matthes, F.: Semantic Types of Legal Norms in German Laws: Classification and Analysis Using Local Linear Explanations, Artificial Intelligence and Law, 2018
[Wa18b]Waltl, B.; Bonczek, G.; Matthes, F.: Rule-based Information Extraction: Advantages, Limitations, and Perspectives, IRIS: Internationales Rechtsinformatik Symposium, Salzburg, Austria, 2018
[Wa18aWaltl, B.; Vogl, R.: Explainable Artificial Intelligence – the New Frontier in Legal Informatics, in: Jusletter IT 22. Februar 2018
 [Gl18a]Glaser, I.; Waltl, B.; Matthes, F.: Named Entity Recognition, Extraction, and Linking in German Legal Contracts, IRISInternationales Rechtsinformatik Symposium, Salzburg, Austria, 2018 (nominated for LexisNexis best paper award)
[La18a]Landthaler, J.; Scepankova, E.; Glaser, I; Lecker, H.; Matthes, F.: Semantic Text matching of contract clauses and legal comments in tenancy law, IRIS: Internationales Rechtsinformatik Symposium, Salzburg, Austria, 2018
2017
[Wa17f]Waltl, B.; Reschenhofer, T.; Matthes, F.: Modeling, Execution and Analysis of Formalized Legal Norms in Model Based Decision Structures, AICOL Workshop - Artificial Intelligence and the Complexity of Legal Systems, London, 2017 (to appear in 2018) 
[Wa17e]Waltl, B.; Muhr, J.; Glaser, I.; Bonczek, G.; Scepankova, S.; Matthes, F.:Classifying Legal Norms with Active Machine Learning, Jurix: International Conference on Legal Knowledge and Information Systems, Luxembourg, Luxembourg, 2017
[Wa17d]Bernhard Waltl / Jörg Landthaler / Florian Matthes, Differentiation and Empirical Analysis of Reference Types in Legal Documents, in: Jusletter IT Flash 17. August 2017, Full Paper
[Wa17c]Waltl, B.; Reschenhofer, T.; Matthes, F.: Process and Tool-support to Collaboratively Formalize Statutory Texts by Executable Models. DEXA - Database and Expert System Applications, Lyon, France, 2017, doi: 10.1007/978-3-319-64471-4_11
[Wa17b]Waltl, B.; Bonczek, G.; Scepankova, E.; Landthaler, J.; Matthes, F.: Predicting the Outcome of Appeal Decisions in Germany’s Tax Law, International Federation for Information Processing (IFIP): Policy Modeling and Policy Informatics, St. Petersburg, Russia, 2017, doi: 10.1007/978-3-319-64322-9_8
[Wa17a]Waltl, B.; Landthaler, J.; Scepankova, E.; Matthes, F.; Geiger, T.; Stocker, C.; Schneider, C.: Automated extraction of semantic information from german legal documents, IRIS: Internationales Rechtsinformatik Symposium, Salzburg, Austria, 2017 (nominated for LexisNexis best paper award)
2016
[Wa16b] Waltl, B.; Matthes, F.; Waltl, T.; Grass, T.: LEXIA - A Data Science Environment for Semantic Analysis of German Legal Texts, IRIS: Internationales Rechtsinformatik Symposium, Salzburg, Austria, 2016
[La15]   Landthaler, J.; Waltl, B.; Matthes, F.: Unveiling References in Legal Texts - Implicit versus Explicit Network Structures, IRIS: Internationales Rechtsinformatik Symposium, Salzburg, Austria, 2016
[Wa16a] Waltl, B.; Altamirano, Laura S.; Matthes, F.: Applying Lexical Knowledge to Support Search and Navigation in Legal Databases, IRIS: Internationales Rechtsinformatik Symposium, Salzburg, Austria, 2016
2015
[Wa15d] Waltl, B.; Zec, M.; Matthes, F.: LEXIA: A Data Science Environment for Legal Texts, Jurix: International Conference on Legal Knowledge and Information Systems, Braga, Portugal, 2015
[Wa15c]   Waltl, B.; Matthes, F.: Supporting the Legal Subsumption Process- Determination of Concreteness and Abstractness in German Laws using Lexical Knowledge, Proceedings of ICAIL Multilingual Workshop on AI & Law 2015, San Diego, USA
[Wa15a]Waltl, B.; Matthes, F.:
Comparison of Law Texts: An Analysis of German and Austrian Legislation regarding Linguistic and Structural Metrics, IRIS: Internationales Rechtsinformatik Symposium, Salzburg, Austria, 2015
2014
[Wa14c]

Waltl, B.; Matthes, F.:
Towards Measures of Complexity: Applying Structural and Linguistic Metrics to German Laws, Jurix: International Conference on Legal Knowledge and Information Systems, Krakow, Poland, 2014

[Wa14b]

Waltl, B.; Schneider, A. W.; Matthes, F.:
Deriving and Modelling Compliance Requirements from Legal Audits, EICAR: Trust and Transparency in IT Security, Frankfurt am Main, Germany, 2014