JusBrasilRec: A large-scale dataset of user sessions for recommendations on the legal domain

1) Jusbrasil

Jusbrasil (https://www.jusbrasil.com.br) is known as the largest legal search portal in Brazil. It provides an online environment where users can find the legal documents that best match their information needs. With millions of user interactions to billions of documents containing different artifacts related to law in Brazil, Jusbrasil appears as a large-scale test bed for advancing research on the still scarce area of legal recommender systems. Thus, we have collected and made available a dataset containing user session data from Jusbrasil.

2) JusBrasilRec Dataset

We collected a 30-day dataset from Jusbrasil spanning between 2021-02-23 and 2021-03-24. This period was chosen to avoid recommendation bias, since it consists of the last 30 days before Jusbrasil launched its first recommender system. 

In the dataset, each line represents a user interaction containing six data fields separated by semicolon: 

- session_id (Integer): session identifier established for each 30 minute interval of user inactivity;
- user_id (Integer): user identifier;
- user_type (String): typer of user, i.e. logged or unlooged user;
- doc_id (Integer): accessed legal document identifier;
- doc_type (String): type of the document, i.e. precedents, articles, news, models and pleadings, and doctrines;
- timestamp (Timestamp): date and hour of the access.

The dataset contains a total of 22,442,232 accesses from 2,310,247 users over 4,225,874 items. The accesses are grouped into 5,415,623 sessions.

3) JusBrasilRec TF-IDF Matrix

Additionally to the user session dataset, we also computed and made available a TF-IDF matrix from the textual content of the documents in Jusbrasil, indicating the importance of a word to its respective document amidst a set of documents. Each line in the matrix contains a pair doc_id/word_id with the respective tf_idf value.

4) How to cite this work:
Marcos Aurélio Domingues, Edleno Silva de Moura, Leandro Balby Marinho and Altigran da Silva. A Large Scale Benchmark for Session-based Recommendations on the Legal Domain. Artificial Intelligence and Law. 2023.