Identification, Categorisation and Forecasting of Court Decisions

Abstract

I will not pretend that getting this dissertation to where it is today was an easy feat, it definitely was not. Diving into a completely new field, learning to navigate between the two disciplines, figuring out the academic world, decrypting legal Dutch, working from home, it was not without challenge. As overwhelming as it may have been at times, it was an exciting and rewarding endeavour, learning, figuring out new things, publishing, presenting and travelling.Throughout the process, I've been guided and supported by Martijn Wieling and Michel Vols. I couldn't wish for better supervisors. I am not able to express how much of an impact they both had on my academic and personal life. Thank you for caring, and for being so involved. Working on every project, paper, and finally the dissertation, I've always felt like a part of the team. Though this dissertation has my name on the cover, it has been a true collaborative effort.I would like to thank my reading committee members Kevin Ashley, Bart Verhij and Janneke Gerards for reading my dissertation and providing such valuable feedback.Being part of two departments, I've had many colleagues over the years who brought endless insights to my work and life. While I will not be able to mention them all here, in no particular order I'd like to thank

The Expanding Methodological Toolbox of the ECHR Scholar

“…For example, to be able to predict the outcome of a case before the Court, an algorithm must be trained on large amounts of information of previous judgments to allow it to learn which factors played a role in reaching that outcome. In using this information, machine learning can help identify such factors and can come to a more detailed categorization of judgments based on new patterns it discovered (Medvedeva, 2022). This may ultimately improve the starting point of the qualitative work of many ECHR scholars and reduce their need to study each and every case individually.…”