Sebis Research News

On Thursday, January 24th, 2019, at 6:00 pm, the Seitz Weckbach Fackler & Partner law firm hosted an event organized by the Augsburg Law Society on Smart Contracts and Blockchain - a Huge hype or an actual disruptive technology? At the lecture evening, Prof Dr. Hans Grigoleit and Prof. Dr. Florian Matthes gave a talk on the diverse technical and legal implications of smart contracts and blockchain from the perspective of computer science on the one hand, and the lawyers perspective on the other.

The presentation slides of Prof. Dr. Florian Matthes talk are available here for download.

The paper "Stop Illegal Comments: A Multi-Task Deep Learning Approach" was presented at the ACM 2018 Artificial Intelligence and Cloud Computing Conference (AICCC 2018), Tokyo, Japan. The authors (Ahmed Elnaggar, Bernhard Waltl, Ingo Glaser, Jörg Landthaler, Elena Scepankova and Florian Matthes) outperformed the state of the art results in detecting illegal comments using transfer learning and multi-task deep learning approach. The paper has won the best paper prize for AI and Machine Learning (first place).

The paper Visualizing Business Ecosystems: Applying a Collaborative Modelling Process in Two Case Studies was presented at the Australasian Conference on Information Systems (ACIS 2019), Sydney, Australia. The authors (Anne Faber, Adrian Hernandez-Mendez, Sven-Volker Rehm and Florian Matthes) report from case studies of two companies that have instantiated ecosystem models following a collaborative approach. The paper has won the ACS Best Paper Prize (third place).

The team of Florian Matthes, consisting of Ingo Glaser, and Elena Scepankova, has published their recent results on Classifying Semantic Types of Legal Sentences: Portability of Machine Learning Models at the 31st international conference on legal knowledge and information systems (Jurix). The paper is going to be presented at the conference from 12-14 December 2018, Netherlands, Groningen.

More details can be found here.

The growing amount of textual data in the legal domain leads to a demand for better text analysis tools adapted to legal domain specific use cases. Semantic Text Matching (STM) is the general problem of linking text fragments of one or more document types. The STM problem is present in many legal document analysis tasks, such as argumentation mining. A common solution approach to the STM problem is to use text similarity measures to identify matching text fragments. In this paper, we recapitulate the STM problem and a use case in German tenancy law, where we match tenancy contract clauses and legal comment chapters. We propose an approach similar to local interpretable model-agnostic explanations (LIME) to better understand the behavior of text similarity measures like TFIDF and word embeddings. We call this approach eXplainable Semantic Text Matching (XSTM).