The content and access dynamics of a busy Web site: findings and implications
Proceedings of the conference on Applications, Technologies, Architectures, and Protocols for Computer Communication
A comparison of DFT and DWT based similarity search in time-series databases
Proceedings of the ninth international conference on Information and knowledge management
Replication strategies in unstructured peer-to-peer networks
Proceedings of the 2002 conference on Applications, technologies, architectures, and protocols for computer communications
Measurement, modeling, and analysis of a peer-to-peer file-sharing workload
SOSP '03 Proceedings of the nineteenth ACM symposium on Operating systems principles
Efficient Similarity Search in Streaming Time Sequences
SSDBM '04 Proceedings of the 16th International Conference on Scientific and Statistical Database Management
Identifying similarities, periodicities and bursts for online search queries
SIGMOD '04 Proceedings of the 2004 ACM SIGMOD international conference on Management of data
I tube, you tube, everybody tubes: analyzing the world's largest user generated content video system
Proceedings of the 7th ACM SIGCOMM conference on Internet measurement
Hi-index | 0.00 |
In P2P file sharing systems, time dependent characteristics of query (query trends) for a file become much important to forecast the demand of the file in future. Prediction of future demand would be effective for the efficient use of the caching mechanism, however, the accurate prediction of query trend is difficult because patterns of query trends may differ significantly according to a nature of keyword used in the query. Identification of query pattern is one of important roles for accurate forecast of future query demand. In this paper, we propose a new method to classify measured query trends into some typical query patterns. We first measure query trend for each keyword in the most famous P2P file sharing system in Japan, and analyze the pattern of query trends by using clustering technique with Discrete Fourier Transform. We then apply our method to the measurement results and show that most of keywords can be categorized into one of four typical trend patterns.