Integrating Power BI with Machine Learning Models for Predictive Analytics
Desari Nikitha, S Imran Hussain, M Poornima · 2024
Predictive analytics techniques in the past frequently relied on single-machine memory-bound assembly and classical statistical models, serving only as static reporting tools with no real-time assistance and interface with newer machine learning (ML) tools. As a result, these techniques produce less accurate forecasts and take longer to respond. The study provides a novel method that combines Power BI with a variety of difficult machine learning models, including Autoregressive Integrated Moving Average (ARIMA), Long Short-Term Memory (LSTM), and Random Forest (RF). The proposed system enhances predictive analytics through realtime operation and interactive graphic display, resulting in better forecasting capabilities. The results are impressive: ARIMA, LSTM, and RF models outperform traditional systems with higher accuracy forecasts (ARIMA: 85.1%, LSTM: 89.3%, RF: 87.4%) and lower prediction error rates (LSTM: MAE 2.7, RMSE 3.8, MAPE 6.5%). Furthermore, data integration generates insights on its own, providing businesses with both data-driven information and practical decision-making tools.