What question did this study set out to answer?

The aim is to present PRISMORA, an AI-powered dashboard that transforms business data into actionable insights for better decision-making.

March 13, 2026Open Access

PRISMORA: An AIPowered Business Insights Dashboard for Intelligent Decision Making

Key Points

The aim is to present PRISMORA, an AI-powered dashboard that transforms business data into actionable insights for better decision-making.
Developed using the MERN stack integrated with Python-based machine learning models
Enabled data ingestion from CSV files, Google Sheets, and REST APIs
Implemented role-based secure dashboards using JSON Web Tokens (JWT)
Conducted experimental evaluations using synthetic sales datasets
Compared forecasting accuracy of ensemble learning models with classical time series models
Ensemble learning models like XGBoost and Random Forest outperformed classical time series models
PRISMORA demonstrated enhanced forecasting accuracy
The system is scalable, cost-effective, and intelligent for modern business intelligence needs

Abstract

In the era of digital transformation, organizations generate massive volumes of data from diverse business operations. However, transforming this data into meaningful and actionable insights remains a significant challenge, particularly for small and medium enterprises. Traditional business intelligence dashboards mainly provide descriptive analytics and lack predictive and prescriptive capabilities. This paper presents PRISMORA, an AI-powered business insights dashboard designed to support intelligent decision-making through real-time analytics, machine learning-based predictions, anomaly detection, and actionable recommendations. The proposed system integrates the MERN stack with Python-based machine learning models to enable seamless data ingestion from CSV files, Google Sheets, and REST APIs. PRISMORA offers role-based secure dashboards using JSON Web Tokens (JWT) and interactive visualizations for effective analysis. Experimental evaluation using synthetic sales datasets demonstrates that ensemble learning models such as XGBoost and Random Forest outperform classical time series models in forecasting accuracy. The results indicate that PRISMORA provides a scalable, cost-effective, and intelligent platform for modern business intelligence.

Bookmark

View Full Paper