A probabilistic model with multi-dimensional features for object extraction

Abstract

To identify recruitment information in different domains, we propose a novel model of hierarchical treestructured conditional random fields (HT-CRFs). In our approach, first, the concept of aWeb object (WOB) is discussed for the description of special Web information. Second, in contrast to traditionalmethods, the Boolean model and multirule are introduced to denote a one-dimensional text feature for a better representation of Web objects. Furthermore, a two-dimensional semantic texture feature is developed to discover the layout of a WOB, which can emphasize the structural attributes and the specific semantics term attributes of WOBs. Third, an optimal WOB information extraction (IE) based on HT-CRF is performed, addressing the problem of a model having an excessive dependence on the page structure and optimizing the efficiency of the model’s training. Finally, we compare the proposed model with existing decoupled approaches forWOB IE. The experimental results show that the accuracy rate of WOB IE is significantly improved and that time complexity is reduced.

Keywords

feature extraction / conditional random fields (CRFs) / information extraction (IE)

References

[1] Cui H, Kan M Y, Chua T S. Soft pattern matching models for definitional question answering. ACM Transactions on Information Systems, 2007, 25(2)

[2] Nyberg E, Mitamura T, Callan J, Carbonell J, . The JAVELIN question-answering system at TREC 2003: a multi-strategy approach with dynamic planning. In: Proceedings of the 12th Text Retrieval Conference. 2003

[3] Mooney R J, Bunescu R. Mining knowledge from text using information extraction. ACM SIGKDD Explorations Newsletter, 2005, 7(1): 3-10

[4] Kobayashi N, Iida R, Inui K, Matsumoto Y. Opinion mining on the web by extracting subject-attribute-value relations. In: Proceedings of AAAI-CAAW’06. 2006

[5] Loth R, Battistelli D, Chaumartin F, . Linguistic information extraction for job ads. In: Proceedings of the 9th International Conference on Adaptivity Personalization and Fusion of Heterogeneous Information. 2010

[6] Ye S, Chua T. Learning object models from semistructured web documents. IEEE Transactions on Knowledge and Data Engineering, 2006, 18(3): 334-349

[7] Jinlin C, Ping Z, Cook T. Detecting web content function using generalized hidden Markov model. In: Proceedings of the IEEE 5th International Conference on Machine Learning and Applications. 2006, 279-284

[8] Freitag D, McCallum A. Information extraction with HMM structures learned by stochastic optimization. In: Proceedings of American Association for Artificial Intelligence (AAAI-00). 2000, 584-589

[9] Haileong C, Hweetou N. A maximum entropy approach to information extraction from semi-structured and free text. In: Proceedings of American Association for Artificial Intelligence (AAAI-02). 2002, 786-791

[10] Finn A, Kushmerick N. A multi-level boundary classification approach to information extraction. In: Proceedings of the 15th European Conference on Machine Learning. 2004, 111-122

[11] Zhu Z. Weakly-supervised relation classification for information extraction. In: Proceedings of the 13th ACM International Conference on Information and Knowledge Management. 2004, 581-588

[12] Wallach H. Conditional random fields: an introduction. University of Pennsylvania CIS Technical Report MS-CIS-04-21. 2004

[13] Kristjansson T, Culotta A, Viola P, McCallum A. Interactive information extraction with constrained conditional random fields. In: Proceedings of American Association for Artificial Intelligence (AAAI-04). 2004, 412-418

[14] Lafferty J, Xiaojin Z, Yan L. Kernel conditional random fields: representation and clique selection. In: Proceedings of the 21st International Conference on Machine Learning (ICML-2004). 2004

[15] Trevor C, Blunsom P. Semantic role labelling with tree conditional random fields. In: Proceedings of the 9th Conference on Computational Natural Language Learning (CoNLL). 2005, 169-172

[16] Chen M M, Chen Y X, Brent M R, Tenney A E. Constrained optimization for validation-guided conditional random field learning. In: Proceedings of the 15th ACM SIGKDD international conference on Knowledge discovery and data mining. 2009, 189-198

[17] Xiao J H, Wang X L, Liu B Q. The study of a nonstationary maximum entropy Markov model and its application on the pos-tagging task. In: Processings of ACM Transactions on Asian Language Information. 2007, 6(2)

[18] Lafferty J, Mccallum A, Pereira F. Conditional random fields: probabilistic models for segmenting and labeling sequence data. In: Proceedings of the 18th International Conference on Machine Learning (ICML 2001). 2001, 282-289

[19] Cohn T, Blunsom P. Semantic role labelling with tree conditional random fields. In: Proceedings of the 9th Conference on Computational Natural Language Learning (CoNLL). 2005, 169-172

[20] Xu Z T. Hierarchical conditional random fields for Chinese part-ofspeech tagging. Midterm Report for National Undergraduate Innovational Experimental Program. 2007

[21] Peng F C, Feng F F, McCallum A. Chinese segmentation and new word detection using conditional random fields. In: Proceedings of the 20th international conference on Computational Linguistics. 2004

[22] Peng F C,McCallum A. Accurate information extraction from research papers using conditional random fields. In: Proceedings of the Human Language Technology Conference on the North American Chapter of the Association for Computational Linguistics (HLT- NAACL 2004). 2004, 329-336

[23] Li W, McCallum A. Rapid development of Hindi named entity recognition using conditional random fields and feature induction. Journal ACM Transactions on Asian Language Information Processing (TALIP), 2003, 2(3): 290-294

[24] Zhu J, Nie Z Q, Wen J R, Ma W Y. 2D conditional random fields for web information extraction. In: Proceedings of the 22nd International Conference on Machine Learning. 2005, 1044-1051

[25] Tang J, Hong M C, Li J Z, Liang B. Tree-structured conditional random fields for semantic annotation. In: Proceedings of the 5th International Semantic Web Conference (ISWC 2006). 2006, 4273(5): 640-653

[26] Zhu J, Zhang B, Nie Z Q, Wen J R, Hong H W. Webpage understanding: an integrated approach. In: Proceedings of the 13th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2007, 903-912

[27] Truyen T T, Phung D Q, Bui H H, Venkatesh S. Hierarchical semimarkov conditional random fields for recursive sequential data. In: Proceedings of the 22nd Annual Conference on Neural Information Processing Systems. 2008

[28] Zhu J. Nie Z Q, Zhang B, Wen J R. Dynamic hierarchical Markov random fields for integrated web data extraction. Journal of Machine Learning Research, 2008, 9: 1583-1614

[29] Nie Z Q, Zhang Y Z, Wen J R, Ma W Y. Object-level ranking: bringing order to web objects. In: Proceedings of WWWConference. 2005, 567-574

[30] Yang X Y, Liu J. Maximum entropy random fields for texture analysis. Pattern Recognition Letters, 2002, 23(1): 93-101

[31] Salton G, Wong A, Yang C S. A vector space model for automatic indexing. Communication of the ACM, 1975, 18(5): 613-620

[32] Cai D, Yu S P, Wen J R, Ma W Y. VIPS: a visionbased page segmentation algorithm. Microsoft Technical Report, MSR-TR-2003-79, 2003

RIGHTS & PERMISSIONS

Higher Education Press and Springer-Verlag Berlin Heidelberg