Sexual Harrasment Content
An IndoBERT-based model for identifying sexual harassment content, used to detect related content in text.
Downloads 25
Release Time : 3/19/2024
Model Overview
This model is a fine-tuned version of IndoBERT, specifically designed to identify sexual harassment content in text. It demonstrates high accuracy, precision, recall, and F1 scores on the evaluation dataset.
Model Features
High accuracy
Achieves 91.43% accuracy, precision, recall, and F1 scores on the evaluation dataset.
Based on IndoBERT
Uses the Indonesian pre-trained IndoBERT-base-p1 as the base model, optimized for Indonesian text.
Lightweight fine-tuning
Requires only 2 training epochs to achieve good performance, with a learning rate set to 5e-05.
Model Capabilities
Text classification
Sexual harassment content detection
Indonesian text analysis
Use Cases
Content moderation
Social media content moderation
Automatically detects sexual harassment-related content on social media
Accurately identifies 91.43% of sexual harassment content
Online community management
Helps online communities identify and filter inappropriate content
Featured Recommended AI Models
Š 2025AIbase