94376 - Sem. Data Science For Legal Analytic

Learning outcomes

Understanding: at the end of the seminar, students acquire competences in reading and understanding the visualization of data and the quantitative and qualitative analysis of data coming from legal domains (eg Law, legislation, administrative documents). This expertise is aimed at improving the capabilities of legal practices and to make the best use of the so-called LegalTech applications ready in the market that use data analytics, AI, ML.

Skeptical attitude: the student acquires the ability to evaluate and argue the results of data manipulation in order to evaluate them in the light of completeness, correctness, consistency especially using the principles of legal theory. This ability is a precondition for developing a neutral, independent, autonomous opinion and for opposing or confirming algorithmic decisions.

Make: The student is also able to formulate a data-driven hypothesis, to select the appropriate dataset, to normalize and clean them, to manipulate and visualize them using the appropriate statistical and IT tools available at the state of the art.

Course contents

1. Open Data and Digital Legal Sources: PSI directive, FAIR principles, FOIA Act – 4 hours (Understand, Sceptical attitude);
2. Legal Knowledge Representation: LegalXML and AI and Law – 4 hours (Understand, Sceptical attitude);
3. Data lifecycle: selection, purge, normalization, filter, aggregation, metadata, visualization, validation, evaluation, curation. – 6 hours (Understand, Sceptical attitude, Make);
4. Textual manipulation: selection, transformation/conversion, extraction, elaboration. – 6 hours (Understand, Sceptical attitude, Make);
5. Quantitative and qualitative analysis, ethics and XAI – 4 hours (Understand, Sceptical attitude).

Readings/Bibliography

Ashley, Kevin D. Artificial Intelligence and Legal Analytics. Cambridge University Press, 2017.

Livermore, Michael, and Daniel Rockmore, eds. Law as Data: Computation, Text, and the Future of Legal Analysis. SFI Press, 2019.

https://www.datascienceforlawyers.org/

http://www.legalanalyticscourse.com/

https://landers.com.au/legal-insights-news/the-future-of-analytics-in-legal

https://www.knime.com/learning

https://github.com/Liquid-Legal-Institute/Legal-Text-Analytics

https://github.com/echr-od

https://data.europa.eu/euodp/data/dataset/covid-19-documents-on-eur-lex

Teaching methods

We provide slides, readings, exercises, technical manuals using https://virtuale.unibo.it/ and the TEAMS platform.

Considering the interdisciplinary approach and technical content, including the practice sessions, attendance is key. We consider the frequency limit exceeded at 75% of attendance.

This course requires an aptitude for pro-active participation with co-working skills.

Assessment methods

The exam requires you to design, implement, document (4 pages) a short data-driven project using the competences acquired. The project could be published on TEAMS. The project could be carried out in groups of up to three people. The project must be presented orally with the possibility of discussing the theoretical part as well.

Teaching tools

The lessons are conducted by prof. Monica Palmirani with the support of Mr. Salvatore Sapienza, Mr. Francesco Sovrano and Mr. Davide Liga. The course is scheduled for the second semester. The course is based on the learning by doing methodology. After the presentation of the theoretical pillars, we use empirical practices to fix the main concepts. Group activities are encouraged to strengthen co-working capacity.

Office hours

See the website of Monica Palmirani