Apurba3036/onubhuti-bangla-sentiment

🌟 Onubhuti: Bangla Sentiment Analysis Model

Onubhuti is a sentiment analysis model specifically trained to classify Bangla text into Positive, Neutral, or Negative sentiments. Fine-tuned from FB XLM-Roberta, it leverages state-of-the-art NLP techniques for robust performance. This model can be applied to various natural language processing (NLP) tasks such as customer reviews, social media analysis, and more.

---

📖 Model Details

📋 Model Description

Onubhuti is a transformer-based model fine-tuned on Bangla text for sentiment classification tasks. It has been trained to classify Bangla sentences into three sentiment categories:

• Positive: Sentiments that express approval, happiness, or general positivity.
• Neutral: Sentiments that are neutral or have no clear positive or negative sentiment.
• Negative: Sentiments that express disapproval, dissatisfaction, or general negativity.

The model uses FB XLM-Roberta, a transformer model pre-trained on multilingual data and further fine-tuned specifically for Bangla. This enables the model to understand nuances in sentiment for Bangla text with high accuracy.

• Developed by: Nazmus Sakib Apurba (https://github.com/Apurba3036)
• Shared by: Tamim
• Model type: Transformer-based classification model
• Language(s): Bangla (বাংলা)
• License: Apache 2.0 (or any relevant license)
• Finetuned from model: FB XLM-Roberta

---

🔗 Model Sources

• Huggingface: https://huggingface.co/docs/transformers/model_doc/xlm-roberta
• Paper: [https://proceedings.neurips.cc/paper_files/paper/2017/file/3f5ee243547dee91fbd053c1c4a845aa-Paper.pdf]

---

🚀 Uses

Onubhuti is designed for sentiment analysis tasks and can be used in various contexts:

• Social Media Monitoring: Analyzing tweets, Facebook posts, or other social media content for sentiment.
• Customer Feedback Analysis: Classifying customer reviews, survey responses, or feedback.
• Content Moderation: Identifying positive, negative, or neutral comments in a user-generated content environment.

🛠️ Installation and Usage

1. Installation:
   To use this model, you need to install the necessary libraries:

   pip install transformers torch
   

   # Load model directly
   from transformers import AutoTokenizer, AutoModelForSequenceClassification
   
   tokenizer = AutoTokenizer.from_pretrained("Apurba3036/onubhuti-bangla-sentiment")
   model = AutoModelForSequenceClassification.from_pretrained("Apurba3036/onubhuti-bangla-sentiment")
   

   💡 Example Output:
   
   Bangla text: এই জায়গাটা এত সুন্দর যে ভাষায় বর্ণনা করা কঠিন। পাহাড় আর সবুজ গাছপালার মিশেলে প্রকৃতি যেন তার সেরা রূপ এখানে মেলে ধরেছে। আমি এই জায়গায় বারবার আসতে চাই।
   Predicted sentiment: Positive