←Back to feed
🧠 AI⚪ NeutralImportance 4/10
Zero- and Few-Shot Named-Entity Recognition: Case Study and Dataset in the Crime Domain (CrimeNER)
arXiv – CS AI|Miguel Lopez-Duran, Julian Fierrez, Aythami Morales, Daniel DeAlcala, Gonzalo Mancera, Javier Irigoyen, Ruben Tolosana, Oscar Delgado, Francisco Jurado, Alvaro Ortigosa||4 views
🤖AI Summary
Researchers have created CrimeNER, a specialized dataset of over 1,500 annotated crime-related documents for training named-entity recognition AI models. The study addresses the lack of quality training data in the crime domain by developing a database from terrorist attack reports and DOJ press notes, defining 22 types of crime-related entities.
Key Takeaways
- →CrimeNER provides over 1,500 annotated crime documents to address the lack of training data for AI models in law enforcement.
- →The dataset defines 5 coarse and 22 fine-grained crime entity types for named-entity recognition tasks.
- →The research focuses on zero-shot and few-shot learning approaches for crime-related NER applications.
- →Training data was extracted from public terrorist attack reports and U.S. Department of Justice press releases.
- →The study evaluates both specialized NER models and general-purpose large language models on crime data.
#named-entity-recognition#crime-detection#dataset#law-enforcement#zero-shot-learning#few-shot-learning#ai-training-data#natural-language-processing
Read Original →via arXiv – CS AI
Act on this with AI
Stay ahead of the market.
Connect your wallet to an AI agent. It reads balances, proposes swaps and bridges across 15 chains — you keep full control of your keys.
Related Articles