Data mining in evaluation of internet path performance

Authors:
Leszek Borzemski
Affiliations:
Wroclaw University of Technology, Institute of Control and Systems Engineering, Wroclaw, Poland
Venue:
IEA/AIE'2004 Proceedings of the 17th international conference on Innovations in applied artificial intelligence
Year:
2004

Citing 0
Cited 5

Complementing search engines with text mining

IEA/AIE'2005 Proceedings of the 18th international conference on Innovations in Applied Artificial Intelligence
Using adaptive fuzzy-neural control to minimize response time in cluster-based web systems

AWIC'05 Proceedings of the Third international conference on Advances in Web Intelligence
Using the geographic distance for selecting the nearest agent in intermediary-based access to internet resources

KES'05 Proceedings of the 9th international conference on Knowledge-Based Intelligent Information and Engineering Systems - Volume Part III
Mining internet data sets for computational grids

KES'05 Proceedings of the 9th international conference on Knowledge-Based Intelligent Information and Engineering Systems - Volume Part III
Lessons from the application of domain-independent data mining system for discovering web user access patterns

KES'06 Proceedings of the 10th international conference on Knowledge-Based Intelligent Information and Engineering Systems - Volume Part III

Quantified Score

Hi-index	0.00

Visualization

Abstract

The data mining algorithms and tools have been employed in many engineering and non-engineering applications. This paper proposes a new practical application of data mining in the study of problems related to the Internet performance. Well-predicted performance is a key engineering issue in Internet-based processing, e.g. in peer-to-peer systems. The measurements of round-trip-time performed by the traceroute probing technique are used for Internet path performance evaluation. We show how data mining can be used by the end-users to evaluate the performance of the communication path between their location and a specific Internet host in a long-term performance prediction scale. The mining data is mined using the IBM Intelligent Miner for Data system. We discover how the round-trip times of the packets and the number of hops they pass vary with the day of the week and the time of the measurement and use this knowledge to built the decision tree that can be a useful guide to the future characteristics of relevant properties of given Internet path in a long-term scale.