Sciweavers

1139 search results - page 60 / 228
» Boosting strategy for classification
Sort
View
IJIT
2004
13 years 9 months ago
Combining ILP with Semi-supervised Learning for Web Page Categorization
This paper presents a semi-supervised learning algorithm called Iterative-Cross Training (ICT) to solve the Web pages classification problems. We apply Inductive logic programming ...
Nuanwan Soonthornphisaj, Boonserm Kijsirikul
JSAT
2008
63views more  JSAT 2008»
13 years 7 months ago
Whose side are you on? Finding solutions in a biased search-tree
We introduce a new jump strategy for look-ahead based satisfiability (Sat) solvers that aims to boost their performance on satisfiable formulae, while maintaining their behavior o...
Marijn Heule, Hans van Maaren
TIT
2010
88views Education» more  TIT 2010»
13 years 2 months ago
Zero-rate feedback can achieve the empirical capacity
The utility of limited feedback for coding over an individual sequence of DMCs is investigated. This study complements recent results showing how limited or noisy feedback can boos...
Krishnan Eswaran, Anand D. Sarwate, Anant Sahai, M...
ECIS
2004
13 years 9 months ago
Formulating an organizational knowledge strategy: the influence of existing IT infrastructure
This paper questions the influence of a pre-existing information technology (IT) infrastructure on the formulation of an organizational knowledge strategy. We draw on a classifica...
Krishna Venkitachalam, Rens Scheepers
KDD
2002
ACM
157views Data Mining» more  KDD 2002»
14 years 8 months ago
Exploiting unlabeled data in ensemble methods
An adaptive semi-supervised ensemble method, ASSEMBLE, is proposed that constructs classification ensembles based on both labeled and unlabeled data. ASSEMBLE alternates between a...
Kristin P. Bennett, Ayhan Demiriz, Richard Maclin