PhoneticAffectIntensityMultilingual 🎙️💬🌍

Multilingual sentiment analysis combining phonesthetic features and emotional dimensions (valence, arousal, dominance) for polarity detection in microblogging sources.

---

📌 Overview

PhoneticAffectIntensityMultilingual is a research tool that goes beyond traditional sentiment polarity by incorporating:

• 🔊 Phonesthetic features — sound-symbolic patterns linked to emotion
• 📐 Affective dimensions — valence, arousal, and dominance (VAD model)
• 🌎 Multilingual support — designed for cross-lingual applicability
• 📱 Microblogging focus — optimized for short, informal texts (Twitter/X)

---

🗂️ Project Structure

PhoneticAffectIntensityMultilingual/
├── data/          # Datasets and lexicons
├── logic/         # Feature extraction and model logic
├── poc/           # Proof-of-concept experiments
├── run/           # Experiment execution scripts
├── root.py        # Main entry point
├── setup.py       # Package configuration
├── requirements.txt
└── README.md

---

⚙️ Installation

git clone https://github.com/EdwinPuertas/PhoneticAffectIntensityMultilingual.git
cd PhoneticAffectIntensityMultilingual
pip install -r requirements.txt

Requirements: Python 3.7+, see requirements.txt for full dependencies.

---

🚀 Usage

python root.py

Explore the poc/ and run/ directories for Jupyter notebooks with step-by-step experiments.

---

🔬 Research Background

Traditional sentiment analysis focuses on binary polarity (positive/negative). This project enriches the analysis by integrating:

Dimension	Description
Valence	Positive ↔ Negative emotional tone
Arousal	Calm ↔ Excited activation level
Dominance	Weak ↔ Strong sense of control
Phonesthemes	Sound clusters with consistent emotional associations

This multi-dimensional approach improves polarity detection, especially in short, context-scarce texts like tweets.

---

📊 Cite This Work

If you use this software in your research, please cite:

@software{puertas2020phoneticaffect,
  author       = {Puertas, Edwin and Alvarado Valencia, Jorge Andrés},
  title        = {PhoneticAffectIntensityMultilingual},
  year         = 2020,
  version      = {v1.0.0},
  doi          = {10.5281/zenodo.4301068},
  url          = {https://doi.org/10.5281/zenodo.4301068}
}

---

👥 Authors

Name	Institution	Contact
Edwin Puertas, PhD	Universidad Tecnológica de Bolívar	epuertas@utb.edu.co
Jorge Andrés Alvarado Valencia, PhD	Pontificia Universidad Javeriana	jorge.alvarado@javeriana.edu.co

Edwin Puertas:

---

📄 License

This project is licensed under the GPL-3.0 License — see the LICENSE file for details.

---

If you find this project useful, please consider giving it a ⭐ and citing our work.