【摘要】 论述了近年来国内外在常规命名实体识别方面研究工作的进展状况,针对其中最为关键的产品命名实体识别技术,考虑到领域本体对产品命名实体识别的支持,提出了将本体特征融入到统计模型中,结合词性特征、上下文特征,以及本体特征的多特征模型进行产品命名实体识别实验,实验结果证明,该方法能有效地提高产品命名实体识别的性能。更多还原

【Abstract】 In recent years,there are a lot of research achievements in the field of NER(name entity recognition),while PNER(product named entity recognition) is just starting.With the rapid development of e-commerce,PNER becomes one of the key techniques of intelligent e-commerce.In the view of supporting domain ontology to PNER,a method of PNER was proposed,which combined CRF(conditional random field) model with domain ontology as a semantic feature besides word and part-of-speech features.Experiments were conducted to compare the two kinds of feature templates,and results indicate that the method can effectively improve the performance of PNER.更多还原