\paper
{Explainable Artificial Intelligence for Trustworthy Decision-Making Systems}
{Kasun Perera, Nadeesha Silva}
{Kasun Perera, Nadeesha Silva}
{Faculty of Computing, University of Sri Jayewardenepura, Sri Lanka\\
	Email: kasun@sjp.ac.lk}
{Artificial Intelligence (AI) systems are increasingly used in critical decision-making domains such as healthcare, finance, and governance. However, the lack of transparency in complex AI models has raised concerns regarding trust, accountability, and ethical decision-making. This study explores the role of Explainable Artificial Intelligence (XAI) in enhancing the interpretability and reliability of AI-driven systems. The objective is to develop approaches that provide clear and understandable explanations for model predictions while maintaining high performance.
	
	The proposed framework integrates machine learning models with explainability techniques such as feature importance analysis, local interpretable model-agnostic explanations (LIME), and SHapley Additive exPlanations (SHAP). These methods are used to interpret model decisions and provide insights into how input features influence predictions. The system is evaluated using real-world datasets across multiple domains to assess its effectiveness in improving user trust and decision transparency.
	
	Experimental results demonstrate that the integration of XAI techniques significantly enhances model interpretability without substantially compromising accuracy. Users are better able to understand, validate, and trust AI-generated outcomes. The study also identifies key challenges, including computational overhead, scalability issues, and the trade-off between model complexity and interpretability.
	
	Furthermore, ethical considerations such as fairness, bias mitigation, and data privacy are examined to ensure responsible AI deployment. The findings suggest that incorporating explainability into AI systems is essential for building trustworthy and human-centered intelligent applications. Future work will focus on developing more efficient and domain-specific explainability techniques to support large-scale real-time systems.
	
	\textbf{Keywords:} Explainable AI, Machine Learning, Trustworthy Systems, Model Interpretability, Ethical AI
}