Báo cáo khoa học: "Ensemble Methods for Unsupervised WSD" doc

... Association for Computational Linguistics Ensemble Methods for Unsupervised WSD Samuel Brody School of Informatics University of Edinburgh s.brody@sms.ed.ac.uk Roberto Navigli Dipartimento di Informatica Universit ` a ... about 27 person-years of human annotation effort. This paper focuses on unsupervised methods which we argue are useful for broad coverage sense disambiguation....

Ngày tải lên: 08/03/2014, 02:21

... word recognition mechanisms for these dictionaries. One of the more important issues in word recognition for all morphologically complex languages involves mechanisms for dealing with affixes. ... the other hand, to reject ill-formed words. On the other hand, we want to use our existing word-recognition and analysis programs as tools for gathering further infor- mation about Eng...

Ngày tải lên: 24/03/2014, 02:20

... Feature set for CRF. responding feature from the Plain feature group that also includes the lexical form of the predicate is most likely a sparse feature. For the opinion holder me in (10), for example, ... LEX-PRED. 6 For this learning method, we use CRF++. 7 We choose a configuration that provides good perfor- mance on our source domain (i.e. ETHICS). 8 For semantic role labeling w...

Ngày tải lên: 22/02/2014, 02:20

... recall for disyllabic words is low. The results for the trigram model are listed in Ta- ble 5. Candidates are restricted to the eight POS cat- egories listed in Table 2 for this model. Precision for the ... information and the like- lihood for a character to appear in a par- ticular position of words of a particular length and POS category. By combining models that use different sou...

Ngày tải lên: 08/03/2014, 04:22

... calcu- lated in an explicit form as: function PKI classify (X) r = 0 # an array, initialized as 0 foreach i ∈ X foreach j ∈ h(i) r j = r j + 1 end end r esult = 0 foreach j ∈ SV r esult = result ... feature combinations are crucial to improving performance, they are heuris- tically selected. Kernel methods change this situation. The merit of the kernel methods is that effective feature co...

Ngày tải lên: 08/03/2014, 04:22

... options for the German word Aktionsplan Aktionsplan Empirical Methods for Compound Splitting Philipp Koehn Information Sciences Institute Department of Computer Science University of Southern California koehn@isi.edu Kevin ... This poses challenges for a number of NLP applications such as machine translation, speech recognition, text classification, information extraction, or inform...

Ngày tải lên: 08/03/2014, 21:20

... AFNLP Minimized Models for Unsupervised Part-of-Speech Tagging Sujith Ravi and Kevin Knight University of Southern California Information Sciences Institute Marina del Rey, California 90292 {sravi,knight}@isi.edu Abstract We ... that our approach performs better than existing state-of-the-art systems in both settings. 1 Introduction In recent years, we have seen increased interest in usin...

Ngày tải lên: 17/03/2014, 01:20

... occur in a variety of word forms. Each word form carries two pieces of information: 1) Lexical content and 2) Morphosyntactic properties. For example, the English word form gave ex- presses the ... citation form, or inflection class of surface word forms in a language. Striv- ing to bypass the time consuming, labor intensive task of constructing a morphological analyzer by hand, uns...

Ngày tải lên: 17/03/2014, 06:20

... base forms, as opposed to attachment information. It is therefore less resource-intensive and more portable than pre- vious corpus-based algorithm proposed for this task. We present results for ... Abstract We present several unsupervised statistical models for the prepositional phrase attachment task that approach the accuracy of the best su- pervised methods for this ta...

Ngày tải lên: 17/03/2014, 07:20

... Meeting of the Association for Computational Linguistics, pages 430–439, Portland, Oregon, June 19-24, 2011. c 2011 Association for Computational Linguistics An Algorithm for Unsupervised Transliteration ... of data for initial supervised training, which means that they are semi-supervised systems. In contrast, our system is fully unsupervised. We achieve an F- measure of up to 92...

Ngày tải lên: 23/03/2014, 16:20