AIbase
EN
Home
Roberta Base Indonesian Sentiment Analysis Smsa
R

Roberta Base Indonesian Sentiment Analysis Smsa

Developed by ayameRushia
This is an Indonesian sentiment analysis model based on the RoBERTa architecture, fine-tuned on the indonlu dataset, specifically designed for sentiment classification tasks in Indonesian SMS texts.
Text Classification
Transformers
Open Source License:MIT#Indonesian Sentiment Analysis#High Accuracy (93.5%)#RoBERTa Fine-tuning
Downloads 21
Release Time : 3/2/2022

Model Overview

The model can perform sentiment analysis on Indonesian texts, determining the emotional tendency of the text, suitable for scenarios such as social media monitoring and customer feedback analysis.

Model Features

High Accuracy
Achieves an accuracy of 93.49% on the evaluation set, demonstrating excellent performance.
Indonesian Language Optimization
Fine-tuned based on an Indonesian pre-trained model, specifically optimized for Indonesian text characteristics.
Lightweight
Based on the RoBERTa-base architecture, relatively lightweight while maintaining performance.

Model Capabilities

Indonesian Text Classification
Sentiment Tendency Analysis
Short Text Processing

Use Cases

Business Analysis
Customer Feedback Analysis
Analyze the sentiment tendency of Indonesian customer reviews
Accurately identifies over 93% of sentiment tendencies
Social Media Monitoring
Public Opinion Monitoring
Monitor public sentiment on Indonesian social media
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
English简体中文繁體中文にほんご
© 2025AIbase