M

Marefa Ner

Developed by marefa-nlp
A large-scale Arabic named entity recognition (NER) model built on a completely new dataset, capable of recognizing 9 different types of entities
Downloads 4,380
Release Time : 3/2/2022

Model Overview

This model is a text span classification knowledge model specifically designed for named entity recognition in Arabic texts, supporting the identification of various entity types such as persons, locations, and organizations.

Model Features

Multi-category entity recognition
Capable of recognizing up to 9 different types of entities, including persons, locations, organizations, etc.
New training data
Built on a completely reconstructed training dataset with high data quality
High-precision recognition
Excellent performance on test sets, especially achieving an F1 score of 0.93 for person recognition

Model Capabilities

Arabic text processing
Named entity recognition
Text span classification

Use Cases

News analysis
News event analysis
Extract key information such as persons, locations, and organizations from news texts
Can accurately identify entities in sentences like 'At Cairo Stadium, the opening ceremony of the African Cup of Nations was held in the presence of the President of the Republic and the President of FIFA'
Social media analysis
Social media content analysis
Analyze key entities in Arabic social media content
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
© 2025AIbase