Travelled to:
1 × Brazil
1 × Canada
1 × France
1 × Switzerland
5 × USA
Collaborated with:
G.V.Cormack W.Yih J.Alspector J.K.Kalita ∅ A.Chowdhury H.Hajishirzi X.Sun V.Prabakarmurthi N.Karunanithi N.J.Howard A.M.Tyrrell N.M.Allinson A.J.Jones S.M.Beitzel E.C.Jensen O.Frieder D.A.Grossman D.D.Lewis
Talks about:
base (4) featur (3) classifi (2) select (2) random (2) detect (2) text (2) near (2) high (2) use (2)
Person: Aleksander Kolcz
DBLP: Kolcz:Aleksander
Facilitated 1 volumes:
Contributed to:
Wrote 11 papers:
- SIGIR-2010-HajishirziYK #adaptation #detection #learning #similarity
- Adaptive near-duplicate detection via similarity learning (HH, WtY, AK), pp. 419–426.
- KDD-2009-KolczC #composition #email
- Genre-based decomposition of email class noise (AK, GVC), pp. 427–436.
- SIGIR-2009-CormackK #evaluation
- Spam filter evaluation with imprecise ground truth (GVC, AK), pp. 604–611.
- KDD-2007-KolczY #classification
- Raising the baseline for high-precision text classifiers (AK, WtY), pp. 400–409.
- KDD-2005-Kolcz #classification #naive bayes
- Local sparsity control for naive Bayes with extreme misclassification costs (AK), pp. 128–137.
- SIGIR-2005-BeitzelJFGLCK #automation #classification #query #using #web
- Automatic web query classification using labeled and unlabeled training data (SMB, ECJ, OF, DAG, DDL, AC, AK), pp. 581–582.
- KDD-2004-KolczCA #detection #robust
- Improved robustness of signature-based near-replica detection via lexicon randomization (AK, AC, JA), pp. 605–610.
- KDD-2002-KolczSK #classification #performance #random
- Efficient handling of high-dimensional feature spaces by randomized classifier ensembles (AK, XS, JKK), pp. 307–313.
- CIKM-2001-KolczPK #categorisation #feature model #summary
- Summarization as Feature Selection for Text Categorization (AK, VP, JKK), pp. 365–370.
- PDP-1994-HowardKTAJ #configuration management #logic #named #novel #parallel #using
- Zelig: A Novel Parallel Computing Machine Using Reconfigurable Logic (NJH, AK, AMT, NMA, AJJ), pp. 70–75.
- DL-1998-AlspectorKK #clique #modelling
- Comparing Feature-Based and Clique-Based User Models for Movie Selection (JA, AK, NK), pp. 11–18.