AnasAlokla/multilingual_go_emotions_V1.1

🌍 Multilingual GoEmotions Classifier 💬

Table of Contents

• 📖 Overview
• ✨ Key Features
• 💫 Supported Emotions
• 🔗 Links
• ⚙️ Installation
• 🚀 Quickstart: Emotion Detection
• 📊 Evaluation
• 💡 Use Cases
• 📚 Trained On
• 🔧 Fine-Tuning Guide
• 🏷️ Tags
• 💬 Support & Contact

Overview

This repository contains a powerful multilingual, multi-label emotion classification model. It is fine-tuned from the robust bert-base-multilingual-cased model on the comprehensive multilingual_go_emotions dataset. The model is designed to analyze text and identify a wide spectrum of 27 different emotions, plus a neutral category. Its ability to detect multiple emotions simultaneously makes it highly effective for understanding nuanced text from diverse sources.

• Model Name: AnasAlokla/multilingual_go_emotions_V1.1
• Architecture: BERT (bert-base-multilingual-cased)
• Tasks: Multi-Label Text Classification | Emotion Detection | Sentiment Analysis
• Languages: Arabic, English, French, Spanish, Dutch, Turkish

Key Features

• 🌍 Truly Multilingual: Natively supports 6 major languages, making it ideal for global applications.
• 🏷️ Multi-Label Classification: Capable of detecting multiple emotions in a single piece of text, capturing complex emotional expressions.
• 💪 High Performance: Built on bert-base-multilingual-cased, delivering strong results across all supported languages and emotions. See the detailed evaluation metrics.
• 🔗 Open & Accessible: Comes with a live demo, the full dataset, and the complete training code for full transparency and reproducibility.
• V1.1 Improved Version: An updated model is available that specifically improves performance on low-frequency emotion samples.

Supported Emotions

The model is trained to classify text into 27 distinct emotion categories as well as a neutral class:

Emotion	Emoji	Emotion	Emoji
Admiration	🤩	Love	❤️
Amusement	😄	Nervousness	😰
Anger	😠	Optimism	✨
Annoyance	🙄	Pride	👑
Approval	👍	Realization	💡
Caring	🤗	Relief	😌
Confusion	😕	Remorse	😔
Curiosity	🤔	Sadness	😢
Desire	🔥	Surprise	😲
Disappointment	😞	Disapproval	👎
Disgust	🤢	Gratitude	🙏
Embarrassment	😳	Grief	😭
Excitement	🎉	Joy	😊
Fear	😱	Neutral	😐

Links

• Live Demo: Hugging Face Space
• Dataset (Supports 6 Languages): multilingual_go_emotions
• Based Model Used: AnasAlokla/multilingual_go_emotions
• GitHub Code: emotion_chatbot

Installation

Install the required libraries using pip:

pip install transformers torch

Quickstart: Emotion Detection

You can easily use this model for multi-label emotion classification with the transformers pipeline. Set top_k=None to see all predicted emotions above the model's default threshold.

from transformers import pipeline

# Load the multilingual, multi-label emotion classification pipeline
emotion_classifier = pipeline(
    "text-classification",
    model="AnasAlokla/multilingual_go_emotions",
    top_k=None # To return all scores for each label
)

# --- Example 1: English ---
text_en = "I'm so happy for you, but I'm also a little bit sad to see you go."
results_en = emotion_classifier(text_en)
print(f"Text (EN): {text_en}")
print(f"Predictions: {results_en}\n")

# --- Example 2: Spanish ---
text_es = "¡Qué sorpresa! No me lo esperaba para nada."
results_es = emotion_classifier(text_es)
print(f"Text (ES): {text_es}")
print(f"Predictions: {results_es}\n")

# --- Example 3: Arabic ---
text_ar = "أشعر بخيبة أمل وغضب بسبب ما حدث"
results_ar = emotion_classifier(text_ar)
print(f"Text (AR): {text_ar}")
print(f"Predictions: {results_ar}")

Expected Output (structure):

Text (EN): I'm so happy for you, but I'm also a little bit sad to see you go. Predictions: [[{'label': 'joy', 'score': 0.9...}, {'label': 'sadness', 'score': 0.8...}, {'label': 'caring', 'score': 0.5...}, ...]]

Text (ES): ¡Qué sorpresa! No me lo esperaba para nada. Predictions: [[{'label': 'surprise', 'score': 0.9...}, {'label': 'excitement', 'score': 0.4...}, ...]]

Text (AR): أشعر بخيبة أمل وغضب بسبب ما حدث Predictions: [[{'label': 'disappointment', 'score': 0.9...}, {'label': 'anger', 'score': 0.9...}, ...]]

Evaluation

The model's performance was rigorously evaluated on the test set.

Test Set Performance

The following table shows the performance metrics of the fine-tuned model on the test set, broken down by emotion category.

The table below shows the performance of the test model:

Performance of Test Model (using class weight)

Labels	accuracy	precision	recall	f1	mcc	support	threshold
admiration	0.933	0.598	0.668	0.631	0.596	2790	0.15
amusement	0.967	0.682	0.793	0.733	0.718	1866	0.10
anger	0.952	0.327	0.356	0.341	0.317	1128	0.15
annoyance	0.908	0.223	0.301	0.256	0.211	1704	0.10
approval	0.920	0.351	0.288	0.317	0.276	2094	0.15
caring	0.970	0.381	0.303	0.337	0.325	816	0.20
confusion	0.959	0.359	0.390	0.374	0.353	1020	0.25
curiosity	0.933	0.405	0.552	0.467	0.438	1734	0.10
desire	0.984	0.385	0.420	0.402	0.394	414	0.30
disappointment	0.958	0.278	0.216	0.243	0.224	1014	0.40
disapproval	0.920	0.221	0.343	0.269	0.235	1398	0.10
disgust	0.972	0.302	0.383	0.338	0.326	600	0.15
embarrassment	0.991	0.388	0.346	0.366	0.362	240	0.45
excitement	0.968	0.248	0.333	0.285	0.272	624	0.10
fear	0.985	0.501	0.526	0.513	0.506	498	0.20
gratitude	0.988	0.913	0.894	0.903	0.897	2004	0.35
grief	0.999	0.529	0.250	0.340	0.363	36	0.85
joy	0.959	0.381	0.472	0.422	0.403	1032	0.15
love	0.971	0.715	0.789	0.750	0.736	1812	0.25
nervousness	0.996	0.430	0.283	0.342	0.347	120	0.70
optimism	0.971	0.573	0.423	0.487	0.478	1062	0.45
pride	0.997	0.468	0.262	0.336	0.349	84	0.25
realization	0.967	0.220	0.146	0.176	0.163	792	0.25
relief	0.993	0.117	0.094	0.104	0.102	138	0.10
remorse	0.987	0.586	0.638	0.611	0.605	516	0.20
sadness	0.960	0.415	0.519	0.461	0.444	1062	0.15
surprise	0.975	0.518	0.425	0.467	0.457	828	0.60
neutral	0.733	0.582	0.621	0.601	0.401	10524	0.10

Test Model Performance (Threshold = 0.5)

The table below shows the performance of the test model with a threshold of 0.5:

Labels	accuracy	precision	recall	f1	mcc	support	threshold
admiration	0.939	0.673	0.570	0.617	0.587	2790	0.5
amusement	0.967	0.735	0.666	0.699	0.682	1866	0.5
anger	0.961	0.400	0.264	0.318	0.306	1128	0.5
annoyance	0.940	0.328	0.137	0.194	0.185	1704	0.5
approval	0.931	0.432	0.211	0.283	0.269	2094	0.5
caring	0.973	0.431	0.246	0.314	0.313	816	0.5
confusion	0.963	0.401	0.337	0.366	0.349	1020	0.5
curiosity	0.944	0.463	0.361	0.406	0.380	1734	0.5
desire	0.985	0.409	0.384	0.396	0.389	414	0.5
disappointment	0.961	0.300	0.198	0.239	0.224	1014	0.5
disapproval	0.945	0.293	0.195	0.234	0.212	1398	0.5
disgust	0.978	0.376	0.267	0.312	0.306	600	0.5
embarrassment	0.991	0.392	0.333	0.360	0.357	240	0.5
excitement	0.977	0.348	0.204	0.257	0.255	624	0.5
fear	0.986	0.547	0.468	0.504	0.499	498	0.5
gratitude	0.988	0.925	0.879	0.902	0.896	2004	0.5
grief	0.999	0.400	0.278	0.328	0.333	36	0.5
joy	0.966	0.451	0.367	0.405	0.389	1032	0.5
love	0.971	0.742	0.747	0.744	0.729	1812	0.5
nervousness	0.996	0.382	0.283	0.325	0.327	120	0.5
optimism	0.971	0.583	0.413	0.484	0.477	1062	0.5
pride	0.997	0.500	0.190	0.276	0.308	84	0.5
realization	0.971	0.270	0.124	0.170	0.169	792	0.5
relief	0.995	0.125	0.029	0.047	0.058	138	0.5
remorse	0.988	0.644	0.560	0.599	0.594	516	0.5
sadness	0.968	0.512	0.408	0.454	0.441	1062	0.5
surprise	0.974	0.492	0.430	0.459	0.447	828	0.5
neutral	0.742	0.648	0.440	0.524	0.368	10524	0.5

Use Cases

This model is ideal for applications requiring nuanced emotional understanding across different languages:

Global Customer Feedback Analysis: Analyze customer reviews, support tickets, and survey responses from around the world to gauge sentiment.

Multilingual Social Media Monitoring: Track brand perception and public mood across different regions and languages.

Advanced Chatbot Development: Build more empathetic and responsive chatbots that can understand user emotions in their native language.

Content Moderation: Automatically flag toxic, aggressive, or sensitive content on international platforms.

Market Research: Gain insights into how different cultures express emotions in text.

Trained On

Base Model: AnasAlokla/multilingual_go_emotions - A powerful pretrained model supporting 104 languages.

Dataset: multilingual_go_emotions - A carefully translated and curated dataset for multilingual emotion analysis, based on the original Google GoEmotions dataset.

Fine-Tuning Guide

To adapt this model for your own dataset or to replicate the training process, you can follow the methodology outlined in the official code repository. The repository provides a complete, end-to-end example, including data preprocessing, training scripts, and evaluation logic.

For full details, please refer to the GitHub repository: emotion_chatbot

Tags

#multilingual-nlp #emotion-classification #text-classification #multi-label #bert #transformer #natural-language-processing #sentiment-analysis #deep-learning #arabic-nlp #french-nlp #spanish-nlp #goemotions #BERT-Emotion #edge-nlp #emotion-detection #offline-nlp
#sentiment-analysis #emojis #emotions #embedded-nlp #ai-for-iot #efficient-bert #nlp2025 #context-aware #edge-ml
#smart-home-ai #emotion-aware #voice-ai #eco-ai #chatbot #social-media
#mental-health #short-text #smart-replies #tone-analysis

Support & Contact

For questions, bug reports, or collaboration inquiries, please open an issue on the Hugging Face Hub repository or contact the author directly.

Author: Anas Hamid Alokla

📬 Email: anasaloklahaaa@gmail.com