Sentix AI is a professional-grade sentiment analysis tool that leverages State-of-the-Art (SOTA) NLP models to detect emotional tone in 20+ languages. Designed for both individual text analysis and high-volume CSV processing, Sentix AI provides actionable data visualizations for researchers and marketers.
- Global Reach: Supports English, Spanish, French, Hindi, and more via a multilingual RoBERTa-based transformer.
- Bulk Processing: Upload CSV datasets and receive instant sentiment mapping.
- Visual Analytics: Interactive Pie and Bar charts for sentiment distribution using Plotly.
- High Accuracy: Detects 5 levels of sentiment (from Very Negative to Very Positive) with confidence scores.
- One-Click Export: Download processed results as a clean CSV for further reporting.
- Language: Python 3.10+
- AI/ML: Hugging Face Transformers, PyTorch, RoBERTa
- Frontend: Streamlit
- Data: Pandas, Plotly (Visualizations)
- Clone the repo:
git clone [https://github.com/mansiggit/sentiment-pro.git](https://github.com/mansiggit/sentiment-pro.git) cd sentix-ai - Create a Virtual Environment:
python -m venv .venv source .venv/bin/activate # Windows: .venv\Scripts\activate
- Install Dependencies:
pip install -r requirements.txt
- Run the App:
streamlit run app.py