Tài liệu Báo cáo khoa học: "Unsupervised Semantic Role Induction with Global Role Ordering" doc
... generative model for unsupervised semantic role induction, which integrates local role assignment deci- sions and a global role ordering decision in a unified model. The role sequence is divided into ... 8-14 July 2012. c 2012 Association for Computational Linguistics Unsupervised Semantic Role Induction with Global Role Ordering Nikhil Garg University of Geneva Swit...
Ngày tải lên: 19/02/2014, 19:20
... sufficient to discriminate between forward entailment and semantic equivalence. To cope with these issues, we explore the contribution of syntactic and semantic features as a complement to lexical ones ... based on pure lexical match by means of “generalized” phrase tables annotated with shallow semantic labels. SPTs, with entries in the form “[LABEL] word 1 word n [LABEL]”, are u...
Ngày tải lên: 19/02/2014, 19:20
... our model is as follows: • Draw the document-level topic proportions β (doc) ∼ GEM(α (doc) ). • Choose the document-level language model φ (doc) i ∼ Dir(γ (doc) ) for i ∈ {1, 2, . . .}. • Draw ... and each sentence n: – Draw type t (k) n ∼ Unif (Doc, P art). – If (t (k) n = Doc) ; draw topic z (k) n ∼ β (doc) ; gen- erate words x (k) n ∼ Mult(φ (doc) z (k) n ) – Otherwise; draw t...
Ngày tải lên: 20/02/2014, 04:20
Tài liệu Báo cáo khoa học: "Unsupervised Search for The Optimal Segmentation for Statistical Machine Translation" doc
... cor- pus contained 19,972 sentences with average sen- tence length 5.6 and 7.7 words for Turkish and English, respectively. The test corpus consisted of 1,512 sentences with 16 reference translations. We ... the two languages by starting with a fine-grained seg- mentation of the Arabic side of the corpus and then merging or deleting Arabic morphemes us- ing alignments with a part-of-sp...
Ngày tải lên: 20/02/2014, 04:20
Tài liệu Báo cáo khoa học: "Unsupervised Translation Induction for Chinese Abbreviations using Monolingual Corpora" ppt
... relations. With the count table Count, we can calculate the relative frequency and get the following probability, P (f ull|abbr) = Count[abbr, f ull] Count[abbr, ∗] (1) 3.4 Translation Induction ... full-form, we can replace the full- form Chinese with its abbreviation to generate trans- lation entries for the abbreviation. Moreover, to deal with the case that an abbreviation may ha...
Ngày tải lên: 20/02/2014, 09:20
Tài liệu Báo cáo khoa học: "Unsupervised Topic Modelling for Multi-Party Spoken Discourse" ppt
... combined LC- Seg with supervised learning over discourse fea- tures (P k = .23); but we expect that a similar ap- proach would be possible here, combining our seg- mentation probabilities with other ... approach al- lows topic mixtures, it requires supervision with hand-labelled topics. In our experiments we therefore compared our results with those obtained by a similar but simpler 1...
Ngày tải lên: 20/02/2014, 11:21
Tài liệu Báo cáo khoa học: "Ontologizing Semantic Relations" pdf
... lexical co-occurrences with the concept in a corpus. 793 tion. This knowledge about mutual selectional preference (the preferred semantic class that fills a certain relation role, as x or y) can ... precision sense-tagged corpora, methods are re- quired to ontologize semantic resources without fully disambiguating text. 3 Ontologizing Semantic Relations Given an instance (x,...
Ngày tải lên: 20/02/2014, 12:20
Tài liệu Báo cáo khoa học: "N Semantic Classes are Harder than Two" doc
... automatically clas- sify semantically related phrases into 10 classes. Classification robustness is im- proved by training with multiple sources of evidence, including within-document cooccurrence, ... related semantic classes • Demonstration that dependency parser paths are inadequate for semantic classification into 7 WordNet classes on TREC news corpora • A benchmark of 10-class seman...
Ngày tải lên: 20/02/2014, 12:20
Tài liệu Báo cáo khoa học: "Unsupervised Relation Disambiguation Using Spectral Clustering" ppt
... Precision/Recall/F-measure compared with other clustering methods. 4 Conclusion and Future work In this paper, we approach unsupervised relation ex- traction problem by using spectral-based clustering technique with diverse ... this table we can find that with the con- text window size setting, 2, the algorithm achieves the best performance of 43.5%/49.4%/46.3% in Precision/Recall/F-meas...
Ngày tải lên: 20/02/2014, 12:20
Tài liệu Báo cáo khoa học: "Unsupervised Segmentation of Chinese Text by Use of Branching Entropy" pdf
Ngày tải lên: 20/02/2014, 12:20