Learning Commonsense Knowledge Models for Semantic Analytics

Learning Commonsense Knowledge Models for Semantic Analytics

2016 IEEE Tenth International Conference on Semantic Computing (ICSC)

10.1109/ICSC.2016.12

https://doi.org/10.1109/icsc.2016.12

Hu Shangfeng, Rajaraman Kanagasabai

24 March 2016

Shangfeng, H., & Kanagasabai, R. (2016). Learning Commonsense Knowledge Models for Semantic Analytics. 2016 IEEE Tenth International Conference on Semantic Computing (ICSC), 400–403. https://doi.org/10.1109/icsc.2016.12

Commonsense knowledge is considered the basis for machine understanding of natural language semantics. In this paper, we propose a new method for fine-grained commonsense knowledge extraction using dependency graphs built through an unsupervised method. We implement the method on a large-scale text collection, and show how our knowledge model can be applied to develop a generic pronoun resolution algorithm. In particular, we demonstrate that our method leads to significant improvements in identifying false co-referred mention pairs, and hence could boost the precision of even the state of the art methods. Our method is unsupervised, and thus the performance can be further improved with bigger collections of raw texts.

Publisher Copyright

There was no specific funding for the research done

© 2016 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

https://oar.a-star.edu.sg/communities-collections/articles/22419

978-1-5090-0662-5

Institute for Infocomm Research