Clustering-based automatic generation of extraction patterns
Zheng Jia-heng, Xingyi Wang, Fei Li · 2004
Most information extraction (IE) systems adopt a pattern-matching approach. As a result, how to generate extraction patterns has become an essential step. Using single-link clustering, examples of patterns can be clustered into various categories, each of which represents a pattern. We applied the method to Chinese agricultural texts, and achieved a plausible performance.