TY - GEN
T1 - Named Entity Disambiguation Leveraging Multi-aspect Information
AU - Zhang, Quanlong
AU - Li, Feng
AU - Wang, Fang
AU - Li, Zhoujun
N1 - Publisher Copyright:
© 2015 IEEE.
PY - 2016/1/29
Y1 - 2016/1/29
N2 - Named Entity Disambiguation (NED) aims at dis-ambiguating named entity mentions in a text to their corre-sponding entries in a knowledge base such as Wikipedia. Itis a fundamental task in Natural Language Processing (NLP)and has many applications such as information extraction, information retrieval, and knowledge acquisition. In the pastdecade, a number of methods have been proposed for theNED task. However, most of existing work focuses on exploringmany more useful information to help tackle this problem. Theeffectiveness of different features proposed for the task arenot well-studied in a same platform. In this paper, we extractvarious remarkable features by leveraging statistical, textual andsemantic information, and evaluate various combinations of themulti-aspect features for the disambiguation task in the sameplatform. Specifically, we utilize two learning to rank methods tocombine different features, train and test the combined methodson several standard data sets. Through extensive experiments, we investigate the effects on the quality of the disambiguationof exploiting different features and show which combinations offeatures are the best choices for disambiguation.
AB - Named Entity Disambiguation (NED) aims at dis-ambiguating named entity mentions in a text to their corre-sponding entries in a knowledge base such as Wikipedia. Itis a fundamental task in Natural Language Processing (NLP)and has many applications such as information extraction, information retrieval, and knowledge acquisition. In the pastdecade, a number of methods have been proposed for theNED task. However, most of existing work focuses on exploringmany more useful information to help tackle this problem. Theeffectiveness of different features proposed for the task arenot well-studied in a same platform. In this paper, we extractvarious remarkable features by leveraging statistical, textual andsemantic information, and evaluate various combinations of themulti-aspect features for the disambiguation task in the sameplatform. Specifically, we utilize two learning to rank methods tocombine different features, train and test the combined methodson several standard data sets. Through extensive experiments, we investigate the effects on the quality of the disambiguationof exploiting different features and show which combinations offeatures are the best choices for disambiguation.
UR - https://www.scopus.com/pages/publications/84964705626
U2 - 10.1109/ICDMW.2015.35
DO - 10.1109/ICDMW.2015.35
M3 - 会议稿件
AN - SCOPUS:84964705626
T3 - Proceedings - 15th IEEE International Conference on Data Mining Workshop, ICDMW 2015
SP - 248
EP - 255
BT - Proceedings - 15th IEEE International Conference on Data Mining Workshop, ICDMW 2015
A2 - Wu, Xindong
A2 - Tuzhilin, Alexander
A2 - Xiong, Hui
A2 - Dy, Jennifer G.
A2 - Aggarwal, Charu
A2 - Zhou, Zhi-Hua
A2 - Cui, Peng
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 15th IEEE International Conference on Data Mining Workshop, ICDMW 2015
Y2 - 14 November 2015 through 17 November 2015
ER -