A

Anglicisms Spanish Mbert

Developed by lirondos
This is a pre-trained model for detecting unassimilated English lexical borrowings (also known as Anglicisms) in Spanish news.
Downloads 7,991
Release Time : 3/28/2022

Model Overview

The model tags foreign vocabulary (primarily from English) used in Spanish, such as *fake news*, *machine learning*, *smartwatch*, *influencer*, or *streaming*.

Model Features

Multilingual Support
Based on the Multilingual BERT architecture, capable of handling lexical borrowing issues across multiple languages.
High-precision Detection
Achieves an F1 score of 85.19 for English loanwords on the test set.
Professional Corpus Training
Trained using the COALAS corpus, containing 370,000 words and covering various written media in European Spanish.

Model Capabilities

English Loanword Detection
Foreign Word Recognition
Code-switching Analysis

Use Cases

News Analysis
News Text Analysis
Analyze the usage of English loanwords in Spanish news
Identifies unassimilated vocabulary such as *fake news*, *machine learning*, etc.
Linguistic Research
Lexical Borrowing Research
Study the frequency and patterns of English loanwords in Spanish
Provides quantitative data to support language contact research
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
Š 2025AIbase