We present memory-based learning approaches to shallow parsing and apply these to five tasks: base noun phrase identification, arbitrary base phrase recognition, clause detection, noun phrase parsing and full parsing. We use feature selection techniques and system combination methods for improving the performance of the memory-based learner. Our approach is evaluated on standard data sets and the results are compared with that of other systems. This reveals that our approach works well for base phrase identification while its application towards recognizing embedded structures leaves some room for improvement.

Treating shallow parsing as part-of-speech tagging yields results comparable with other, more elaborate approaches. Using the CoNLL 2000 training and testing material, our best model had an accuracy of 94.88%, with an overall FB1 score of 91.94%. The individual FB1 scores for NPs were 92.19%, VPs 92.70% and PPs 96.69%. I

The performance of machine learning algorithms can be improved by combining the output of different systems. In this paper we apply this idea to the recognition of noun phrases. We generate different classifiers by using different representations of the data. By combining the results with voting techniques described in (Van Halteren et al., 1998) we manage to improve the best reported performances on standard data sets for base noun phrases and ar bitrary noun phrases.

Abstract not found

Abstract not found

In this paper, we define the chunking problem as a classification of words and present a weighted probabilistic model for a text chunking. The proposed model exploits context features around the focus word. And to alleviate the sparse data problem, it integrates general features with specific features. In the training stage, we select useful features after measuring information gain ratio of each features and assign higher weight to more informative feature by adopting the information gain ratio. At the application time, we classify words into chunk labels while checking consistency of the begin and the end of a chunk. The experimental results show that the model combining general and specific features alleviates the sparse data problem. In addition, the weighted probabilistic model based on information gain ratio outperforms the non-weighted model.