International Journal of Innovative Computer Science and IT Research

E-ISSN: 3067-1108

A Widely Indexed Open Access Peer Reviewed Multidisciplinary Monthly Scholarly International Journal

Call for Paper Volume 2 Issue 6 November-December 2026 Submit your research before last 3 days of this month to publish your research paper in the issue of November-December.

Explainable Data Analytics for Transparent Computational Decision Support

Author(s) Ritesh Ajoodha
Country South Africa
Abstract The increasing use of computational systems for prediction, classification, recommendation, risk assessment, and resource allocation has created a growing requirement for decision processes that are not only accurate but also understandable and transparent. Conventional data analytics frequently produces complex computational outputs without adequately explaining how particular predictions or recommendations are generated. Explainable Data Analytics addresses this limitation by integrating analytical models with techniques that communicate the factors, relationships, and patterns contributing to computational decisions. This paper examines the role of explainable data analytics in developing transparent computational decision-support systems, focusing on model interpretability, feature importance, local and global explanations, surrogate modeling, rule-based explanations, visualization, uncertainty, and human-centered decision support. A conceptual methodological framework is developed consisting of data preparation, analytical modeling, explanation generation, explanation validation, and decision-support integration.
The analysis considers both inherently interpretable models and post-hoc explanation techniques for complex machine-learning systems. The study identifies important application areas including cybersecurity, financial risk assessment, healthcare analytics, education, business intelligence, public administration, and intelligent information systems. Two analytical tables compare explainability approaches and application requirements for transparent decision support. The findings indicate that explainability can improve user understanding, facilitate model auditing, support error identification, and strengthen confidence in computational recommendations.
However, transparency is affected by several challenges, including explanation fidelity, computational complexity, conflicting stakeholder requirements, data quality, model opacity, explanation instability, and the potential for misleading interpretations. The study argues that explainability should be treated as an integral component of analytical system design rather than as an optional layer added after model development. The paper concludes that combining interpretable analytical methods, post-hoc explanation techniques, uncertainty communication, human oversight, and continuous evaluation can contribute to more accountable and trustworthy computational decision-support environments.
Keywords Explainable Data Analytics, Explainable Artificial Intelligence, Transparent Decision Support, Model Interpretability, Computational Decision-Making, Data Science, Feature Importance, Machine Learning, Explainability, Intelligent Systems.
Field Engineering
Published In Volume 2, Issue 4, July-August 2026
Published On 2026-08-24

Share this