Sciweavers

727 search results - page 20 / 146
» Two thresholds are better than one
Sort
View
KDD
2009
ACM
227views Data Mining» more  KDD 2009»
14 years 8 months ago
Efficiently learning the accuracy of labeling sources for selective sampling
Many scalable data mining tasks rely on active learning to provide the most useful accurately labeled instances. However, what if there are multiple labeling sources (`oracles...
Pinar Donmez, Jaime G. Carbonell, Jeff Schneider
FSKD
2007
Springer
126views Fuzzy Logic» more  FSKD 2007»
14 years 1 months ago
Rough Set Model Selection for Practical Decision Making
One of the challenges a decision maker faces is choosing a suitable rough set model to use for data analysis. The traditional algebraic rough set model classifies objects into th...
Joseph P. Herbert, Jingtao Yao
JMM2
2007
127views more  JMM2 2007»
13 years 7 months ago
The Cut Detection Issue in the Animation Movie Domain
— In this paper we are proposing an improved cut detection algorithm adapted to the animation movies domain. A cut is the direct concatenation of two different shots and produces...
Bogdan Ionescu, Patrick Lambert, Didier Coquin, Va...
ETT
2000
91views Education» more  ETT 2000»
13 years 7 months ago
Performance Analysis and Estimation of Call Admission Control Parameters in Wireless Integrated Voice and Data Networks
We propose an admission control policy for wireless multimedia networks that is based on the well known threshold-based guard channel method. The new scheme deals with two differen...
Mahmoud T. El-Hadidi, Khaled M. F. Elsayed, Mohame...
ICPR
2008
IEEE
14 years 2 months ago
Novel image feature alphabets for object recognition
Most successful object recognition systems are based on a visual alphabet of quantised gradient orientations. Here, we introduce two richer image feature alphabets for use in obje...
Martin Lillholm, Lewis D. Griffin