TY - GEN
T1 - Concept hierarchy extraction from textbooks
AU - Wang, Shuting
AU - Liang, Chen
AU - Wu, Zhaohui
AU - Williams, Kyle
AU - Pursel, Bart
AU - Brautigam, Benjamin
AU - Saul, Sherwyn
AU - Williams, Hannah
AU - Bowen, Kyle
AU - Giles, C. Lee
N1 - Funding Information:
We gratefully acknowledge partial support from NSF and useful comments from the reviewers.
Publisher Copyright:
© 2015 ACM.
PY - 2015/9/8
Y1 - 2015/9/8
N2 - Concept hierarchies have been useful tools for presenting and organizing knowledge. With the rapid growth of online knowledge resources, automatic concept hierarchy extraction is increasingly attractive. Here, we focus on concept extraction from textbooks based on the knowledge in Wikipedia. Given a book, we extract important concepts in each book chapter using Wikipedia as a resource and from this construct a concept hierarchy for that book. We define local and global features that capture both the local relatedness and global coherence embedded in that textbook. In order to evaluate the proposed features and extracted concept hierarchies, we manually construct concept hierarchies for three well used textbooks by labeling important concepts for each book chapter. Experiments show that our proposed local and global features achieve better performance than using only keyphrases to construct the concept hierarchies. Moreover, we observe that incorporating global features can improve the concept ranking precision and reaffirms the global coherence in the book.
AB - Concept hierarchies have been useful tools for presenting and organizing knowledge. With the rapid growth of online knowledge resources, automatic concept hierarchy extraction is increasingly attractive. Here, we focus on concept extraction from textbooks based on the knowledge in Wikipedia. Given a book, we extract important concepts in each book chapter using Wikipedia as a resource and from this construct a concept hierarchy for that book. We define local and global features that capture both the local relatedness and global coherence embedded in that textbook. In order to evaluate the proposed features and extracted concept hierarchies, we manually construct concept hierarchies for three well used textbooks by labeling important concepts for each book chapter. Experiments show that our proposed local and global features achieve better performance than using only keyphrases to construct the concept hierarchies. Moreover, we observe that incorporating global features can improve the concept ranking precision and reaffirms the global coherence in the book.
UR - http://www.scopus.com/inward/record.url?scp=84959199116&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84959199116&partnerID=8YFLogxK
U2 - 10.1145/2682571.2797062
DO - 10.1145/2682571.2797062
M3 - Conference contribution
AN - SCOPUS:84959199116
T3 - DocEng 2015 - Proceedings of the 2015 ACM Symposium on Document Engineering
SP - 147
EP - 156
BT - DocEng 2015 - Proceedings of the 2015 ACM Symposium on Document Engineering
PB - Association for Computing Machinery, Inc
T2 - ACM Symposium on Document Engineering, DocEng 2015
Y2 - 8 September 2015 through 11 September 2015
ER -