Abstract :

This Study Presents An Intelligent Retail Analytics Platform That Integrates Graph Convolution Networks (GCNs) With Explainable Machine Learning Techniques To Enhance Enterprise-level Decision-making. The System Is Designed To Process Heterogeneous Retail Data, Including Customer Demographics, Transactional Records, And Textual Reviews, To Generate Actionable Insights For Business Monitoring, Evaluation, And Strategic Planning. By Combining Advanced Predictive Modeling With Intuitive Analytics, The Platform Enables A Deeper Understanding Of Customer Behavior, Product Performance, And Evolving Sales Trends Across Different Market Segments. The Proposed Architecture Incorporates Multiple Machine Learning Models, Including Tree-based Classifiers And Regressors, Along With A Hybrid Approach That Integrates GCN With Natural Gradient Boosting (NGB). This Combination Significantly Improves Predictive Performance In Both Classification Tasks, Such As Customer Rating Prediction, And Regression Tasks, Such As Total Sales Forecasting. Furthermore, Textual Data Is Effectively Processed Using TF-IDF Vectorization, Enhancing The System’s Ability To Capture Customer Sentiment, Opinions, And Feedback Patterns. To Ensure Transparency And Interpretability, The Platform Integrates Explainable Machine Learning Methods, Allowing Stakeholders To Evaluate Model Outcomes Using Key Performance Metrics Such As Accuracy, Precision, Recall, F1-score, And Error Measures. Visual Analytics Tools, Including Confusion Matrices, ROC Curves, And Correlation Heatmaps, Provide Clear And Interpretable Insights For Informed Decision-making. Additionally, A Web-based Interface Is Developed To Support Secure User Interaction, Role-based Access Control, And Real-time Prediction Capabilities. The Platform Also Facilitates Exploratory Data Analysis Through Interactive Visualizations, Enabling Better Understanding Of Sales Patterns, Demographic Distributions, And Market Segmentation.

Published:

09-4-2026

Issue:

Vol. 26 No. 4 (2026)

Page Nos:

2067-2075

Section:

Articles

License:

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

How to Cite