Adaptive context trees and text clustering

IEEE Transactions on Information Theory

Volumn 47, Issue 5, 2001, Pages 1884-1901

  Jean Philippe, V ^a  

^a Institute for Chemical Research   (Japan)

Author keywords

Adaptive mixture of models; Context tree weighting method; Mean Kullback risk; Text modeling

Indexed keywords

ADAPTIVE ALGORITHMS; COMPUTER SIMULATION; FORMAL LANGUAGES; PARAMETER ESTIMATION; PROBABILITY; TEXT PROCESSING; TREES (MATHEMATICS);

ADAPTIVE CONTEXT TREES; TEXT CLUSTERING;

MARKOV PROCESSES;

EID: 0035397419     PISSN: 00189448     EISSN: None     Source Type: Journal    
DOI: 10.1109/18.930925     Document Type: Article

References (24)

1. A vector space model for automatic indexing
2. Universal noiseless coding
3. Universal modeling and coding
4. Minimax noiseless universal coding for Markov sources
5. Twice-universal coding
6. The context-tree weighting method: Basic properties
7. Hierarchical universal coding
8. The minimum description length principle in coding and modeling
9. Universal prediction
10. A fast adaptive coding algorithm
11. Laplace's law of succession and universal encoding
12. Instances of exact prediction and a new type of inequalities obtained by anchoring
13. 'Universal' aggregation rules with exact bias boundst
14. The 'progressive mixture' estimator for regression trees
15. Gibbs estimators
16. Prediction of random sequences and universal coding
17. Context weighting for general finite-context sources
18. Are Bayes rules consistent in information?
19. Information-theoretic determination of minimax rates of convergence