›› 2011, Vol. 26 ›› Issue (1): 68-80.doi: 10.1007/s11390-011-1112-0

• Special Section on Natural Language Processing • Previous Articles     Next Articles

Using Syntactic-Based Kernels for Classifying Temporal Relations

Seyed Abolghasem Mirroshandel, Gholamreza Ghassem-Sani, and Mahdy Khayyamian   

  1. Department of Computer Engineering, Sharif University of Technology, Tehran 11155-9517, Iran
  • Received:2009-12-31 Revised:2010-11-22 Online:2011-01-01 Published:2011-01-01
  • About author:Seyed Abolghasem Mirroshandel is a Ph.D. candidate at Department of Computer Engineering of Sharif University of Technology in Tehran, Iran. He obtained his B.Sc. degree in computer science from Tehran University in 2005, and his M.Sc. degree in artificial intelligence from Sharif University of Technology in 2007. His main research interests include natural language processing and information retrieval.
    Gholamreza Ghassem-Sani is currently an associate professor of Department of Computer Engineering of Sharif University of Technology in Tehran, Iran. He obtained his B.Sc. degree in computer science from Shahid Beheshti University in Tehran in 1985, and his M.Sc. degree in intelligent knowledge base systems and Ph.D. degree in computer science from University of Essex in U.K. in 1987 and 1992, respectively. His main research interests include natural language processing and AI planning.
    Mahdy Khayyamian received his M.Sc. degree in information technology from Sharif University of Technology in 2008. He obtained his B.Sc. degree in computer science from Tehran University in 2006. His main research interests include natural language processing and information retrieval.

Temporal relation classification is one of contemporary demanding tasks of natural language processing. This task can be used in various applications such as question answering, summarization, and language specific information retrieval. In this paper, we propose an improved algorithm for classifying temporal relations, between events or between events and time, using support vector machines (SVM). Along with gold-standard corpus features, the proposed method aims at exploiting some useful automatically generated syntactic features to improve the accuracy of classification. Accordingly, a number of novel kernel functions are introduced and evaluated. Our evaluations clearly demonstrate that adding syntactic features results in a considerable improvement over the state-of-the-art method of classifying temporal relations.

[1] Mani I, Marc V, Wellner B, Lee C, Pustejovsky J. Machine earning of temporal relations. In Proc. ACL 2006, Sydney, ustralia, Jul. 17-21, 2006, pp.753-760.
[2] Tatu M, Srikanth M. Experiments with reasoning for temporal elations between events. In Proc. Coling 2008, Manchester, K, Aug. 18-22, 2008, pp.857-864.
[3] Allen J F. Towards a general theory of action and time. Arificial Intelligence, 1984, 23(2): 123-154.
[4] Pustejovsky J, Hanks P, Sauri R, See A, Gaizauskas R, Seter A, Radev D, Sundheim B, Day D, Ferro L, Lazo M. The imeBank corpus. In Proc. Corpus Linguistics, Lancaster, K, Mar. 28-31, 2003, pp.647-656.
[5] Chklovski T, Pantel P. Global path-based refinement of noisy raphs applied to verb semantics. In Proc. IJCNLP 2005, eju Island, Korea, Oct. 11-13, 2005, pp.792-803.
[6] Torisawa K. Acquiring inference rules with temporal contraints by using Japanese coordinated sentences and nounerb co-occurrences. In Proc. HLT NAACL 2006, Main conerence, New York City, USA, Jun. 4-9, 2006, pp.57-64.
[7] Inui T, Inui K, Matsumoto Y. What kinds and amounts of ausal knowledge can be acquired from text by using conective markers as clues? In Proc. 6th International Confernce on Discovery Science, Sapporo, Japan, Oct. 17-19, 2003, p.180-193.
[8] Zanaotto F M, Pennacchiotti M, Pazienza M T. Discovering symmetric entailment relations between verbs using selecional preferences. In Proc. ACL 2006, Sydney, Australia, ul. 17-21, 2006, pp.849-856.
[9] Chambers N, Wang S, Jurafsky D. Classifying temporal reations between events. In Proc. ACL 2007, Prague, Czech, un. 23-30, 2007, pp.173-176.
[10] Bethard S, Martin J H. Learning semantic links from a corpus f parallel temporal and causal relations. In Proc. ACL 2008, olumbus, USA, Jun. 15-20, 2008, pp.177-180.
[11] Bethard S, Corvey W, Klingenstein S, Martin J H. Building a orpus of temporal-causal structure. In Proc. Language Reources and Evaluation Conference (LREC2008), Morroco, ay 28-30, 2008, pp.908-915.
[12] Bethard S, Martin J H, Klingenstein S. Finding temporal tructure in text: Machine learning of syntactic temporal relaions. International Journal of Semantic Computing (IJSC), Dec. 2007, 1(4): 441-457.
[13] Lapata M, Lascarides A. Learning sentence-internal temporal elations. Journal of Artificial Intelligence Research, 2006, 7: 85-117.
[14] Bethard S, Martin J H, Klingenstein S. Timelines from text: dentification of syntactic temporal relations. In Proc. ICSC, rvine, USA, Sept. 17-19, 2007, pp.11-18.
[15] Savova G, Bethard S, Styler W, Martin J H, Palmer M, asanz J,Ward W. Towards temporal relation discovery from he clinical narrative. In Proc. American Medical Informatics ssociation, San Francisco, USA, Nov. 14-18, 2009, pp.568- 72.
[16] Lapata, M. Lascarides A. Inferring sentence-internal tempoal relations. In Proc. NAACL 2004, Boston, USA, May 2-7, p.153-160.
[17] Bethard S, Martin J H. CU-TMP: Temporal relation classiication using syntactic and semantic features. In Proc. SeEval 2007, Prague, Czech, Jun. 23-24, 2007, pp.129-163.
[18] Chambers N, Jurafsky D. Jointly combining implicit contraints improves temporal ordering. In Proc. EMNLP2008, awaii, USA, Oct. 25-27, 2008, pp.698-706.
[19] Puscasu G. Wvali: Temporal relation identification by yntactico-semantic analysis. In Proc. SemEval 2007, rague, Czech, Jun. 23-24, pp.484-487.
[20] Bramsen P, Deshpande P, Lee Y K, Barzilay R. Inducing emporal graphs. In Proc. EMNLP2006, Sydney, Australia, ul. 22-23, 2006, pp.189-198.
[21] Boguraev B, Ando K. Timeml-compliant text analysis for emporal reasoning. In Proc. IJCAI 2005, Edinburgh, UK, ul. 30-Aug. 5, 2005, pp.997-1003.
[22] Bethard S. Finding event, temporal and causal structure in ext: A machine learning approach [Ph.D. Dissertation]. Uniersity of Colorado at Boulder, 2007.
[23] Mirroshandel S A, Khayyamian M, Ghassem-Sani G R. Using ree kernels for classifying temporal relations between events. n Proc. PACLIC 2009, Hong Kong, China, Dec. 3-5, 2009, p.355-364.
[24] Mirroshandel S A, Khayyamian M, Ghassem-Sani G R. Eventime temporal relation classification using syntactic tree kerels. In Proc. the 4th Language and Technology Conference LTC2009), Poznan, Poland, Nov. 6-8, 2009, pp.300-304.
[25] Abe S, Inui K, Matsumoto Y. Two-phased event relaion acquisition coupling the relation-oriented and argumentriented approaches. In Proc. Coling 2008, Manchester, UK, ug. 18-22, 2008, pp.1-8.
[26] Harris Z. Mathematical Structure of Language. New York: ohn Wiley & Sons, 1968.
[27] Lin D, Pantel P. DIRT | Discovery of inference rules from ext. In Proc. the 7th ACM SIGKDD, San Francisco, USA, ug. 26-29, 2001, pp.323-328.
[28] Szpektor I, Tanev H, Dagan I. Scaling Web-based acquisition f entailment relations. In Proc. EMNLP2004, Barcelona, pain, Jul. 25-26, 2004, pp.41-48.
[29] Pekar V. Acquisition of verb entailment from text. In Proc. LT NAACL Main Conference, New York, USA, Jun. 4-9, 006, pp.49-56.
[30] Torisawa K. Acquiring inference rules with temporal contraints by using Japanese coordinated sentences and nounerb co-occurrences. In Proc. the Human Language Technolgy Conference of the NAACL, Main Conference, New York, New York, USA, Jun. 4-9, 2006, pp.57-64.
[31] Boser B E, Guyon I M, Vapnik V N. A training algorithm for ptimal margin classifiers. In Proc. COLT1992, Pittsburgh, SA, Jul. 27-29, 1992, pp.144-152.
[32] Cortes C, Vapnik V. Support-vector networks. In Proc. Conerence of Machine Learning, Tahoe City, USA, Jul. 9-12, 1995, pp.273-297.
[33] Collins M, Duffy N. Convolution kernels for natural language. In Proc. NIPS 2001, Vancouver, Canada, Dec. 3-8, 2001, p.625-632.
[34] Khayyamian M, Mirroshandel, S A, Abolhassani H. Syntactic ree-based relation extraction using a generalization of Collins nd Duffy convolution tree kernel. In Proc. HLT NAACL, Boulder USA, May 31-Jun. 5, 2009, pp.66-71.
[35] Moschitti A. Syntactic kernels for natural language learning: he semantic role labeling case. In Proc. HLT NAACL, New ork, USA, Jun. 4-9, 2006, pp.97-100.
[36] Zhang M, Zhang J, Su J, Zhou G D. A composite kernel to exract relations between entities with both ffiat and structured eatures. In Proc. ACL 2006, Sydney, Australia, Jul. 17-21, 2006, pp.825-832.
[37] Mani I, Wellner B, Verhagen M, Pustejovsky J. Three aproaches to learning Tlinks in TimeML. Technical Report, S-07-268, Brandeis University, Walthan, USA, 2007.
[38] Chang C C, Lin C J. Libsvm: A library for support vector achines. 2001, http://www.csie.ntu.edu.tw/cjlin/libsvm.
[39] The Stanford Natural Language Processing Group, http://nlp.stanford.edu/software/index.shtml.

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!

ISSN 1000-9000(Print)

         1860-4749(Online)
CN 11-2296/TP

Home
Editorial Board
Author Guidelines
Subscription
Journal of Computer Science and Technology
Institute of Computing Technology, Chinese Academy of Sciences
P.O. Box 2704, Beijing 100190 P.R. China
Tel.:86-10-62610746
E-mail: jcst@ict.ac.cn
 
  Copyright ©2015 JCST, All Rights Reserved