Judicial precedents search supported by natural language processing and clustering

Calambas, Manuel; Ordóñez, Armando; Chacon, Angela; Ordóñez, Hugo

doi:10.1109/columbiancc.2015.7333448

Cited by 6 publications

(2 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In Colombia, a legal structure has been designed as the formal source of law. Authorities must apply norms and rules with the same factual and legal assumptions on the constitutional court ruling, thus, lawyers must use previous judgments to support their current cases [7]. Legal informatics is a set of tools to facilitates the development of activities corresponding to the legal area, such as searching for documents and transcribing information through a machine.…”

Section: State-of-the-artmentioning

confidence: 99%

DEVA a Methodology based on User-Centered Design for the development of a search system for judicial documents

Ordoñez

Ordóñez²,

Ordóñez³

2022

Investigación e Innovación en Ingenierías

View full text Add to dashboard Cite

Summary: This work provides a methodology to analyze user preferences regarding the components that a judicial search engine should include. We validate this methodology by developing a search system. Starting from initial characteristics, we designed a search prototype. A test was then applied to identify whether the search engine meets the user's needs. The prototype developed was compared to other next-generation search engines in terms of user satisfaction and accuracy. The results show that our proposal outperforms other search engines. Objective: Define and implement the components for the development of a judicial search engine. Methodology: The methodology consists of in: Definition: Includes a review of the state of the art of the technologies associated with the project. Prototype development: Different modules and components of the system are developed here. Evaluation: focused on usability and user satisfaction. Results: The characteristics and elements necessary for the search engine were identified. To this end, Benchmarking on general aspects and components of jurisprudence search engines was carried out. Additionally, a survey was conducted with 79 legal experts. Finally, a use test was carried out. Conclusions: A prototype of the search engine was designed according to the characteristics declared by the users. This prototype showed optimal usability and navigation because it was designed considering the needs of the user's jurisprudence. Its interfaces are easy to understand and its filters make the search process more agile and precise. Finally, the developed search engine offers a good experience to the users.

show abstract

Section: State-of-the-artmentioning

confidence: 99%

DEVA a Methodology based on User-Centered Design for the development of a search system for judicial documents

Ordoñez

Ordóñez²,

Ordóñez³

2022

Investigación e Innovación en Ingenierías

View full text Add to dashboard Cite

show abstract

“…As far as we know, this is the first work to explore the use of BERT to automate case law documents categorization in Brazilian Portuguese. Other works explore the same problem in other languages or dialects of Portuguese [Calambás et al 2015, Gonc ¸alves and Quaresma 2003, Gonc ¸alves and Quaresma 2005, Feinerer and Hornik 2008, Mencía and Fürnkranz 2010, and one work explores a single-label version of the same problem with less data in Brazilian Portuguese [de Colla Furquim and De Lima 2012]. All of these articles studied algorithms created before BERT.…”

Section: Related Workmentioning

confidence: 99%

verBERT: Automating Brazilian Case Law Document Multi-label Categorization Using BERT

Serras

Finger

2021

Anais Do XIII Simpósio Brasileiro De Tecnologia Da Informação E Da Linguagem Humana (STIL 2021)

View full text Add to dashboard Cite

In this work, we carried out a study about the use of attention-based algorithms to automate the categorization of Brazilian case law documents. We used data from the Kollemata Project to produce two distinct datasets with adequate class systems. Then, we implemented a multi-class and multi-label version of BERT and fine-tuned different BERT models with the produced datasets. We evaluated several metrics, adopting the micro-averaged F1-Score as our main metric for which we obtained a performance value of 〈F1〉micro = 0.72 corresponding to gains of 30 percent points over the tested statistical baseline.

show abstract

Creating a Brief Review of Judicial Practice Using Clustering Methods

Taran

Revunkov²,

Gapanyuk³

2022

Studies in Computational Intelligence

View full text Add to dashboard Cite

Judicial precedents search supported by natural language processing and clustering

Cited by 6 publications

References 9 publications

DEVA a Methodology based on User-Centered Design for the development of a search system for judicial documents

DEVA a Methodology based on User-Centered Design for the development of a search system for judicial documents

verBERT: Automating Brazilian Case Law Document Multi-label Categorization Using BERT

Creating a Brief Review of Judicial Practice Using Clustering Methods

Contact Info

Product

Resources

About