en
0.25
0.5
0.75
1.25
1.5
1.75
2
KMV-Peer: A Robust and Adaptive Peer-Selection Algorithm
Published on 2011-08-092835 Views
The problem of fully decentralized search over many collections is considered. The objective is to approximate the results of centralized search (namely, using a central index) while controlling the c
Related categories
Presentation
KMV-Peer: A Robust and Adaptive Peer-Selection Algorithm00:00
Motivation and Problem Statement00:25
Solution Framework - Indexing01:34
Our Contributions03:46
Agenda04:24
Per-term KMV Statistics04:36
Peer-Scoring Functions06:36
Peer-Scoring Functions - contd08:17
KMV-int: The Peer Intersection Score10:43
KMV-exp: The Peer Expected Score12:46
A Basic Peer-Selection Algorithm13:26
Algorithm Improvements –Save Communication Cost15:34
Algorithm Improvements – Adaptive Ranking17:55
KMV-Peer: The Peer-Selection Algorithm19:32
Experimental Setting19:36
KMV-Peer Compared to State-of-the-Art21:07
Tuning The Parameters of KMV-Peer23:02
Testing Different Variants of KMV-Peer23:36
Testing Different Scoring Functions25:14
Conclusions27:24
Future Work28:02
Thank You!28:26