A decision-theoretic generalization of on-line learning and an application to boosting
Journal of Computer and System Sciences - Special issue: 26th annual ACM symposium on the theory of computing & STOC'94, May 23–25, 1994, and second annual Europe an conference on computational learning theory (EuroCOLT'95), March 13–15, 1995
BoosTexter: A Boosting-based Systemfor Text Categorization
Machine Learning - Special issue on information retrieval
Conditional Random Fields: Probabilistic Models for Segmenting and Labeling Sequence Data
ICML '01 Proceedings of the Eighteenth International Conference on Machine Learning
Large Margin Methods for Structured and Interdependent Output Variables
The Journal of Machine Learning Research
Collective multi-label classification
Proceedings of the 14th ACM international conference on Information and knowledge management
EMNLP '02 Proceedings of the ACL-02 conference on Empirical methods in natural language processing - Volume 10
Multilabel Neural Networks with Applications to Functional Genomics and Text Categorization
IEEE Transactions on Knowledge and Data Engineering
Prediction of Sequential Values for Debt Recovery
CIARP '09 Proceedings of the 14th Iberoamerican Conference on Pattern Recognition: Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications
Editorial: Hybrid intelligent algorithms and applications
Information Sciences: an International Journal
Learning and inference order in structured output elements classification
ACIIDS'12 Proceedings of the 4th Asian conference on Intelligent Information and Database Systems - Volume Part I
| Hi-index | 0.00 |
The paper describes the method for structured output classification that is able to perform classification task for unknown shape of output structure. In previous work authors provided that the classification of the element in the output structure can be performed using standard input of the instance (its profile) as well as all other preceding output elements (already classified) as learning attributes. Now they present how the order of the score function application to each element of output space is important and may determine the overall accuracy. For that reason the paper addresses the crucial problem of how to order elements in the structured learning process to get greater final accuracy. The learning is performed by means of ensemble, boosting classification method adapted to structured prediction - the AdaBoostSeq algorithm according to several ordering methods. A heuristic method of ordering is worked out as well. Experimental studies were carried out on a number of real financial datasets.