论文信息 - Pattern Learning and Active Feature Selection for Word Sense Disambiguation

Pattern Learning and Active Feature Selection for Word Sense Disambiguation

We present here the main ideas of the algorithm employed in the SMUls and SMUaw systems. These systems have participated in the Senseval-2 competition attaining the best performance for both English all words and English lexical sample tasks. The algorithm has two main components (1) pattern learning from available sense tagged corpora (SemCor) and dictionary definitions (WordNet), and (2) instance based learning with active feature selection, when training data is available for a particular word.

Rada Mihalcea | Dan I. Moldovan | Rada Mihalcea | D. Moldovan

[1] Thomas G. Dietterich,et al. Learning with Many Irrelevant Features , 1991, AAAI.

[2] Eric Brill,et al. Transformation-Based Error-Driven Learning and Natural Language Processing: A Case Study in Part-of-Speech Tagging , 1995, CL.

[3] Walter Daelemans,et al. TiMBL: Tilburg Memory-Based Learner, version 2.0, Reference guide , 1998 .

[4] Rada Mihalcea,et al. An Automatic Method for Generating Sense Tagged Corpora , 1999, AAAI/IAAI.

[5] Malte Helmert,et al. Transformation-Based Error-Driven Learning: Eine Fallstudie in Part of Speech Tagging , 2000 .

[6] Walter Daelemans,et al. Memory-Based Word Sense Disambiguation , 2000, Comput. Humanit..