Forecasting Automobile Sales in Urban Areas Using an Extreme Learning Machine With A Self-Attention Mechanism
Kayalvizhi Subramanian, Gunasekar Thangarasu · 2023
This study introduces a novel approach for automobile sales forecasting through the fusion of the Self-Attention Mechanism (SAM) and the Extreme Learning Machine (ELM). The integration of SAM, a powerful deep learning mechanism, with ELM, a robust machine learning algorithm, aims to enhance the accuracy and efficiency of forecasting models in the dynamic and complex domain of automobile sales. SAM facilitates the model ability to capture long-range dependencies and intricate patterns within the sales data, while ELM contributes its fast training and generalization capabilities. The proposed model exhibits superior performance in handling the non-linear relationships inherent in automobile sales data, allowing for more accurate predictions and informed decision-making for stakeholders in the automotive industry. The self-attention mechanism enables the model to focus on relevant features and temporal dependencies, resulting in a more nuanced understanding of the intricate factors influencing sales fluctuations. Additionally, the ELM component ensures the model computational efficiency and scalability, making it suitable for real-time forecasting applications. The experimental validation on real-world automobile sales datasets demonstrates the superiority of the proposed SAM-ELM model over traditional forecasting techniques. The model ability to adapt to changing market dynamics and capture intricate patterns positions it as a valuable tool for industry professionals seeking reliable sales predictions. The experimental results suggest that the proposed SAM-ELM model consistently outperforms existing methods in forecasting automobile sales. The combination of Self-Attention Mechanism and Extreme Learning Machine synergistically enhances the model ability to capture complex patterns and adapt to changing conditions. The observed percentage improvements across various metrics underscore the efficacy of SAM-ELM in real-time forecasting, making it a promising approach for stakeholders in the automotive industry.