Skip to content
Open access

Explainable Machine Learning for Predictive Analytics: Examining the Influence of Feature Optimization and Model Interpretability

Sep 2026 · Journal of Global Social Transformation · 0 citations · 43 references

Abstract

This study examined the application of explainable machine learning (XAI) in predictive analytics by investigating the influence of feature optimization and model interpretability on machine learning performance and transparency. A quantitative experimental and comparative research design was employed using a structured dataset comprising 1,000 observations collected from a machine-learning dataset relevant to predictive analytics. The data were prepared through systematic data collection and preprocessing procedures, including the identification and treatment of missing values, duplicate observations, inconsistent entries, categorical variables, and numerical features. Feature optimization was subsequently performed to identify and retain the most relevant predictors while reducing redundant and less informative variables. Three machine learning models, including Support Vector Machine (SVM), Random Forest (RF), and Gradient Boosting (GB), were trained and comparatively evaluated using training and testing procedures with cross-validation and hyperparameter optimization. Predictive performance was assessed using accuracy, precision, recall, F1-score, and ROC-AUC, while explainability and interpretability were examined using SHAP (SHapley Additive exPlanations) for global feature contribution and LIME (Local Interpretable Model-Agnostic Explanations) for local prediction explanations. The analysis was conducted using Python-based tools, including Pandas, NumPy, Scikit-learn, SHAP, LIME, and Matplotlib. The findings indicated that feature optimization improved the predictive performance of the machine learning models while reducing feature complexity, and the explainability analysis provided clearer insights into the contribution of important predictors and individual model decisions. The study is significant because it integrates predictive performance with transparency and interpretability, demonstrating that explainable machine learning can support the development of more efficient, understandable, and practically useful predictive analytics systems.

Read PDF

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.