This project is ongoing. Descriptions, methods and results may be updated after further validation.
Educational and informational content. It is not legal advice, guidance for a specific case or an institutional position. Examples and analyses use only legitimate sources and public data, aggregated or properly anonymised.
Research question
Can patterns of abusive litigation be identified in an explainable, auditable way that is compatible with human oversight and due process?
Motivation
Repetitive and artificial litigation consumes judicial resources and harms those with legitimate claims. Screening tools exist, but they are rarely explainable or auditable. The project proposes a framework in which every signal is interpretable and the decision remains human.
Data
- Public procedural metadata (DataJud and STJ), without personal identifiers
Methodology
- Interpretable procedural-pattern indicators
- Explainable machine learning with human-review safeguards
Limitations
- Early stage - only data probes have been carried out.
- Statistical patterns are not equivalent to unlawful conduct; any use requires human oversight and due process.
Implications
- Transparency and auditability criteria for judicial screening tools.
Planned applied products (possible outreach, subject to registration)
- Annotation protocol and public documentation of the framework