A Framework for Enhanced Time Series Forecasting of Internet Traffic Based on AdaBoost-LSTM Integration

• Published in: Proceedings (International Conference on Computer Engineering and Applications. Online) pp. 348 - 353
• Main Author: Zhang, Qicheng
• Format: Conference Proceeding
• Language: English
• Published: IEEE 12.04.2024
• Subjects: AdaBoost; Adaptation models; Computational modeling; Ensemble methods; Experimental validation; Heuristic algorithms; Internet; Internet traffic forecasting; LSTM networks; Prediction algorithms; Predictive models; Time series analysis
• ISSN: 2159-1288
• DOI: 10.1109/ICCEA62105.2024.10604099

Internet traffic is conceptualized as a form of time series data, making algorithms designed for time series forecasting applicable for predicting dynamics of internet traffic. This study introduces for the first time an improved AdaBoost algorithm tailored for the task of website traffic forecasting. By customizing the AdaBoost algorithm to meet the demands of continuous prediction scenarios, its performance in continuous spaces is enhanced. To increase the accuracy of internet traffic forecasts, this research proposes an enhanced algorithmic framework that combines the refined AdaBoost algorithm with Long Short-Term Memory (LSTM) networks, effectively addressing the unique challenges presented by time series forecasting. Ultimately, this improved AdaBoost-LSTM framework is employed for internet traffic prediction, demonstrating its efficacy in handling the complexities of time series data inherent to internet traffic flows.