Behavior modeling using bigram frequencies for client-side link prefetching

Authors:
A. Georgakis;H. Li;M. Gordan
Affiliations:
Digital Media Laboratory (DML), Department of Applied Physics and Electronics, Umeå University, Umeå, Sweden;Digital Media Laboratory (DML), Department of Applied Physics and Electronics, Umeå University, Umeå, Sweden;Basis of Electronics Department, Technical University of Cluj-Napoca, Cluj-Napoca, Romania
Venue:
IMSA'06 Proceedings of the 24th IASTED international conference on Internet and multimedia systems and applications
Year:
2006

Citing 7
Cited 0

Using statistical testing in the evaluation of retrieval experiments

SIGIR '93 Proceedings of the 16th annual international ACM SIGIR conference on Research and development in information retrieval
Foundations of statistical natural language processing

Foundations of statistical natural language processing
A re-examination of text categorization methods

Proceedings of the 22nd annual international ACM SIGIR conference on Research and development in information retrieval
Machine learning in automated text categorization

ACM Computing Surveys (CSUR)
Caching on the World Wide Web

IEEE Transactions on Knowledge and Data Engineering
Proxy Cache Algorithms: Design, Implementation, and Performance

IEEE Transactions on Knowledge and Data Engineering
The design and evaluation of web prefetching and caching techniques

The design and evaluation of web prefetching and caching techniques

Quantified Score

Hi-index	0.00

Visualization

Abstract

The perceived latency for a user surfing the Internet is the target of a transparent and speculative algorithm that relies on a user behavior model. The model is based on past user behavior and in combination with a weighting scheme for the outbound links of a particular web page, aims at reducing the perceived latencies. The assistance is in the form of prefetching some linked web pages and storing them in the browser's cache. A comparison between the proposed algorithms against two other prefetching algorithms yield improved cache-hit rates given a moderate bandwidth overhead. Furthermore, the experimental results are proven to be statistically significant.