A Probabilistic Graphical Model for Recognizing NP Chunks in Texts

We present a probabilistic graphical model for identifying noun phrase patterns in texts. This model is derived from mathematical processes under two reasonable conditional independence assumptions with different perspectives compared with other graphical models, such as CRFs or MEMMs. Empirical results shown our model is effective. Experiments on WSJ data from the Penn Treebank, our method achieves an average of precision 97.7% and an average of recall 98.7%. Further experiments on the CoNLL-2000 shared task data set show our method achieves the best performance compared to competing methods that other researchers have published on this data set. Our average precision is 95.15% and an average recall is 96.05%.