Collective Entity Linking and a Simple Slot Filling Method for TAC-KBP 2011
暂无分享,去创建一个
This paper summarize our work in TAC2010 knowledge base population track. We submit result for english entity linking and regular slot filling task. For entity linking we use a frequency based method as baseline and implement a collective method following (Han et al., 2011) for entity linking. For slot filling, we use wikipedia infobox as a source of supervision by mapping back to sentences to generate more training sentences. Evaluation result shows the strength and weakness of both our approaches.
[1] Ian H. Witten,et al. Learning to link with wikipedia , 2008, CIKM '08.
[2] Dan Klein,et al. Accurate Unlexicalized Parsing , 2003, ACL.
[3] Silviu Cucerzan,et al. Large-Scale Named Entity Disambiguation Based on Wikipedia Data , 2007, EMNLP.
[4] Heng Ji,et al. Knowledge Base Population: Successful Approaches and Challenges , 2011, ACL.
[5] Jun Zhao,et al. Collective entity linking in web text: a graph-based method , 2011, SIGIR.