ibert-roberta-base-Abusive_Or_Threatening_Speech open-source model - Accurately detect abusive and threatening remarks

Ibert Roberta Base Abusive Or Threatening Speech

Developed by DunnBC22

This model is a fine-tuned version based on ibert-roberta-base, specifically designed for detecting abusive or threatening speech.

Downloads 174.14k

Release Time : 5/5/2023

Model Overview

This model is a fine-tuned version based on kssteven/ibert-roberta-base, focusing on identifying abusive or threatening content in text.

High accuracy

Achieved 97.41% accuracy on the evaluation dataset.

Balanced performance metrics

F1 score 0.7773, recall 0.8610, precision 0.7084, demonstrating good balance.

Efficient training

Only requires 1 training epoch to achieve good performance, with a training loss of just 0.0771.

Text classification

Abusive speech detection

Threatening content identification

Content moderation

Social media content filtering

Automatically identifies and filters abusive or threatening speech on social media.

97.41% accuracy, effectively reducing the spread of harmful content.

Online community management

Helps forums and online communities automatically detect inappropriate speech.

High recall rate (86.1%) ensures most inappropriate speech can be captured.

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Recall	Precision
0.0771	1.0	1828	0.0802	0.9741	0.7773	0.8610	0.7084

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base