MétaCan
Menu
Back to cohort
Record W4210274906 · doi:10.5216/rfd.v45i2.70086

Application of Artificial Intelligence in the automatic identification and classification repetitive demand resolution incident in the Brazilian Court of Justice

2022· article· en· W4210274906 on OpenAlex
Antônio Pires Castro Júnior, Gabriel Wainer, Wesley Pacheco Calixto

Why this work is in the frame

A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.

affAt least one author lists a Canadian institution in the pinned OpenAlex snapshot.

Bibliographic record

VenueRevista da Faculdade de Direito da UFG · 2022
Typearticle
Languageen
FieldSocial Sciences
TopicArtificial Intelligence in Law
Canadian institutionsCarleton University
FundersDivision of Graduate EducationCiência sem FronteirasConselho Nacional de Desenvolvimento Científico e TecnológicoFundação de Amparo à Pesquisa do Estado de GoiásCoordenação de Aperfeiçoamento de Pessoal de Nível Superior
KeywordsComputer scienceIdentification (biology)Artificial intelligenceEconomic JusticeRecallSimilarity (geometry)Resolution (logic)Precision and recallLegislationLawOperations researchMachine learningPolitical scienceEngineeringPsychology

Abstract

fetched live from OpenAlex

One of the areas of knowledge with several possibilities for applying artificial intelligence is Law. Recent changes in Brazilian legislation have facilitated the use of information technology resources to streamline the progress and judgment of cases, such as repetitive demand resolution incident (IRDRs). The aim of this paper is to develop and apply an AI method that can identify and relate new lawsuits with consolidated repetitive judgments (IRDRs). The datasets used in this research are judges' repetitive judgment documents, and consolidated in IRDRs. Court documents are transformed into weighted vectors. The construction of the weights in the vector is based on the co-occurrence of the terms, calculated from the combination of the term frequency-inverse document frequency and their similarity in the corpus of the same IRDR. Artificial neural networks are trained with these vectors to recognize whether new lawsuits are related to an IRDR. As the methodology obtained 93% accuracy, 97% precision, and 93% in recall in the simulations, the method can streamline the work of the Court of Justice, seeking to solve society’s conflicts as quickly as possible. Although the method can be used in several scenarios, the simulations were carried out in judicial documents.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

Full frame distilled prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.

metaresearch head score (Codex)0.007
metaresearch head score (Gemma)0.001
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Theoretical or conceptual · Consensus signal: none
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.822
Threshold uncertainty score0.406

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0070.001
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0000.001
Science and technology studies0.0010.001
Scholarly communication0.0000.000
Open science0.0010.000
Research integrity0.0000.000
Insufficient payload (model declined to judge)0.0000.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.057
GPT teacher head0.374
Teacher spread0.317 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it