A novel framework that converts data into understandable concepts before generating natural language explanations, ensuring transparency in the analysis process.

• Two-stage approach: Data → abstract concepts → natural language text
• Explainable by design: Overcomes the black-box limitations of standard LLMs
• Multilingual capabilities: Generates explanations in various languages
• Domain-adaptable: Particularly valuable for medical data where transparency and accuracy are critical

For medical applications, this system provides trustworthy data interpretations with clear reasoning paths, essential for clinical decision support and patient data analysis where accountability and explainability are non-negotiable.

Data2Concept2Text: An Explainable Multilingual Framework for Data Analysis Narration