EVICT Talk with Masha Medvedeva – To Predict or Not to Predict? Re-thinking the Field of Automatic Court Decision Classification

January 26th, 2021

Masha Medvedeva presented her paper in which she and her co-authors, Prof. Dr. Martijn Wieling and Prof. Dr. Michel Vols, we discuss previous research in automatic prediction of court decisions. They defined the difference between outcome identification, outcome-based judgement categorisation and outcome forecasting, and review how various studies fall into these categories. She discussed how important it is to understand the legal data that one works with in order to determine which task can be performed. Finally, she reflected on the needs of the legal discipline regarding the analysis of court judgments. The paper was published Open Access in AI & Law.

About the speaker

Masha Medvedeva is a computational linguist and a postdoctoral researcher in AI and Law at Radboud university. She defended her PhD ‘Identification, Categorisation and Forecasting of Court Decisions’, under supervision of Michel Vols and Martijn Wieling at the University of Groningen. Her work focused on automatic detection of linguistic patterns within legal data, with particular focus on analysing and developing systems designed to predict court decisions. Much of her work concerned the judgements of the European Court of Human Rights. As a member of EVICT, Masha focused on researching housing rights using data science techniques. As part of her work, she developed machine learning systems for classification and analysis of Dutch case law, as well as specifically eviction cases.

The EVICT Monthly Talk Series takes place every third Thursday of the month. For the next talk, please visit our event page for further information.