Learning the past tense of English verbs using inductive logic programming
Connectionist, Statistical, and Symbolic Approaches to Learning for Natural Language Processing
ILP '99 Proceedings of the 9th International Workshop on Inductive Logic Programming
Unsupervised learning of the morphology of a natural language
Computational Linguistics
Bootstrapping morphological analyzers by combining human elicitation and machine learning
Computational Linguistics
Multilingual text analysis for text-to-speech synthesis
Natural Language Engineering
Modularity in a connectionist model of morphology acquisition
COLING '94 Proceedings of the 15th conference on Computational linguistics - Volume 1
A Bayesian model for morpheme and paradigm identification
ACL '01 Proceedings of the 39th Annual Meeting on Association for Computational Linguistics
Unsupervised learning of morphology using a novel directed search algorithm: taking the first step
MPL '02 Proceedings of the ACL-02 workshop on Morphological and phonological learning - Volume 6
The relationship between Precision-Recall and ROC curves
ICML '06 Proceedings of the 23rd international conference on Machine learning
Overview and results of Morpho challenge 2009
CLEF'09 Proceedings of the 10th cross-language evaluation forum conference on Multilingual information access evaluation: text retrieval experiments
Unsupervised word decomposition with the promodes algorithm
CLEF'09 Proceedings of the 10th cross-language evaluation forum conference on Multilingual information access evaluation: text retrieval experiments
Paramor: from paradigm structure to natural language morphology induction
Paramor: from paradigm structure to natural language morphology induction
Towards Learning Morphology for Under-Resourced Fusional and Agglutinating Languages
IEEE Transactions on Audio, Speech, and Language Processing
| Hi-index | 0.00 |
This paper demonstrates that the use of ensemble methods and carefully calibrating the decision threshold can significantly improve the performance of machine learning methods for morphological word decomposition. We employ two algorithms which come from a family of generative probabilistic models. The models consider segment boundaries as hidden variables and include probabilities for letter transitions within segments. The advantage of this model family is that it can learn from small datasets and easily generalises to larger datasets. The first algorithm Promodes, which participated in the Morpho Challenge 2009 (an international competition for unsupervised morphological analysis) employs a lower order model whereas the second algorithm Promodes-H is a novel development of the first using a higher order model. We present the mathematical description for both algorithms, conduct experiments on the morphologically rich language Zulu and compare characteristics of both algorithms based on the experimental results.