ruBert-base-russian-emotion-detection Open Source Model - Accurately Classify 9 Russian Emotions

Rubert Base Russian Emotion Detection

Developed by MaxKazak

A Russian multi-label emotion classification model fine-tuned on ruBert-base, trained on the ru_goemotions dataset, supporting 9 emotion categories

Text Classification

Transformers

OtherOpen Source License:Apache-2.0 #Russian Sentiment Analysis #Multi-label Classification #High AUC Performance

Downloads 1,288

Release Time : 5/28/2023

Model Overview

This model is specifically designed for sentiment analysis of Russian texts, capable of identifying multiple emotions expressed in text, including joy, interest, surprise, sadness, anger, disgust, fear, guilt, and neutral emotions.

Model Features

Multi-label Emotion Classification

Capable of identifying multiple emotions expressed in text simultaneously, rather than a single emotion

High Accuracy

Achieves 92.4% AUC on the evaluation set, demonstrating excellent classification performance

Broad Emotion Coverage

Supports classification of 9 different emotions, covering a wide range of human emotional expressions

Model Capabilities

Russian text sentiment analysis

Multi-label classification

Emotion recognition

Use Cases

Social Media Analysis

User Comment Sentiment Analysis

Analyze the emotions in Russian user comments on social media

Accurately identifies multiple compound emotions

Customer Service

Customer Feedback Emotion Classification

Automatically classify emotional tendencies in Russian customer feedback

Helps quickly identify negative emotional feedback

🚀 ruBert-base-russian-emotions-classifier-goEmotions

This model is a fine - tuned version of [ai - forever/ruBert - base](https://huggingface.co/ai - forever/ruBert - base) on Djacon/ru_goemotions. It is designed for sentiment analysis, emotion classification, and can handle both multi - label and multi - class tasks.

🚀 Quick Start

This model can be used for multilabel text classification. You can input text through the widget provided, such as "Очень рад тебя видеть!" to get emotion classification results.

✨ Features

Multifaceted Classification: Capable of multi - label and multi - class emotion classification.
High - quality Metrics: Achieves an AUC of 0.9240 on the evaluation set and has good performance on various emotion labels.

📚 Documentation

Model Information

Property	Details
Model Type	ruBert - base - russian - emotions - classifier - goEmotions
Base Model	ai - forever/ruBert - base
Training Data	Djacon/ru_goemotions
Metrics	accuracy, roc_auc

Evaluation Results

This model achieves the following results on the evaluation set (2nd epoch):

Loss: 0.2088
AUC: 0.9240

The quality of the predicted probabilities on the test dataset is as follows:

label	joy	interest	surprise	sadness	anger	disgust	fear	guilt	neutral	average
AUC	0.9369	0.9213	0.9325	0.8791	0.8374	0.9041	0.9470	0.9758	0.8518	0.9095
F1 - micro	0.9528	0.9157	0.9697	0.9284	0.8690	0.9658	0.9851	0.9875	0.7654	0.9266
F1 - macro	0.8369	0.7922	0.7561	0.7392	0.7351	0.7356	0.8176	0.8247	0.7650	0.7781

🔧 Technical Details

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 5e - 05
train_batch_size: 16
eval_batch_size: 16
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon = 1e - 08
lr_scheduler_type: linear
num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	AUC
0.1755	1.0	1685	0.1717	0.9220
0.1391	2.0	3370	0.1757	0.9240
0.0899	3.0	5055	0.2088	0.9106

Framework versions

Transformers 4.24.0
Pytorch 2.0.1
Datasets 2.12.0
Tokenizers 0.11.0

📄 License

This project is licensed under the Apache - 2.0 license.

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご