Saved in:
| Main Author: | |
|---|---|
| Format: | Recurso digital |
| Language: | English |
| Published: |
Zenodo
2025
|
| Online Access: | https://doi.org/10.5281/zenodo.15259563 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Table of Contents:
- <p><em><span lang="EN-IN">In the era of information technology, the Internet serves as a critical medium for accessing information globally. The World Wide Web (WWW) facilitates a diverse range of Internet-based services, including e-commerce, online banking, entertainment, education, and e-governance. However, the exponential growth in web applications has led to a substantial increase in network traffic, causing congestion and elevating server loads. This, in turn, results in higher response times, thereby negatively impacting user experience. Web caching has emerged as an effective solution to mitigate latency issues by storing frequently accessed web objects closer to end users. Traditional caching strategies, such as Least Recently Used (LRU), Least Frequently Used (LFU), SIZE, GD-Size, and GDSF, have been widely implemented to enhance web system performance. However, recent advancements in machine learning have significantly improved conventional web proxy caching policies. Support Vector Machine (SVM), a robust supervised machine learning algorithm, is extensively utilized for both classification and regression tasks. By integrating conventional caching policies with SVM-based predictive models, intelligent caching approaches have been developed. These models are evaluated using trace-driven simulations, and their performance is systematically compared with traditional web proxy caching techniques. The empirical findings indicate that SVM-enhanced caching strategies yield substantial performance improvements, demonstrating the efficacy of machine learning in optimizing web caching systems.</span></em></p>