Báo cáo khoa học: "Graph Ranking for Sentiment Transfer" doc

... graph -ranking algorithm for sentiment transfer by integrating the sentiment orientations of the documents, which could be considered as a sen- timent-transfer version of the graph -ranking ... old-domain documents as well as the “pseudo” labels of new-domain documents. Experimental results show that proposed algorithm could improve the per- formance of baseline methods dramatic...

Ngày tải lên: 31/03/2014, 00:20

... S 1 , which contains all of the information that is in the input sentence. The part of the information that is implicit in the sentence (tense, voice, and so forth) is made explicit in S 1 . ... certain German verb form into English, it is necessary to un- derstand the German verb form as part of a complex of features of German structure in- cluding possibly other verb forms within t...

Ngày tải lên: 19/02/2014, 19:20

... unlike formal sign language – rarely fol- lows any predefined form or grammar, and may vary dramatically by speaker. A key implementational challenge is automati- cally extracting gestural information ... gestures, in which the form of motion is the principle communicative fea- ture (McNeill, 1992). Non-textual features for topic segmentation Re- search on non-textual features for topic se...

Ngày tải lên: 20/02/2014, 09:20

... June 2008. c 2008 Association for Computational Linguistics Self-Training for Biomedical Parsing David McClosky and Eugene Charniak Brown Laboratory for Linguistic Information Processing (BLLIP) Brown ... al., 2006a; McClosky et al., 2006b). One possible use for this technique is for parser adaptation — initially training the parser on one type of data for which hand-labeled tree...

Ngày tải lên: 20/02/2014, 09:20

... attentional information to produce coherent text (McKeown, 1985, McCoy, 1985, Paris, 1988b), omitting intentional informa- tion, or conflating intentional and rhetorical information (Hovy, ... ers. This is necessary for linguis- tic reasons (e.g., to generate the appropriate clausal connectives in multi-sentential responses) and for responding to requests for elabora- tion...

Ngày tải lên: 21/02/2014, 20:20

... are required for the word forms of a given language. A part of such a ma- chine is shown in Figure 1. The sequences which this will generate are those required for the Future forms both of ... syntactic and semantic justification. The various forms discussed in this paper are customarily displayed in paradigms; for example, see Reynolds [1962] for the paradigms of MANDARE, a v...

Ngày tải lên: 07/03/2014, 18:20

Ngày tải lên: 08/03/2014, 01:20

... 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 ... senses for w. For each sense s i and for each neighbor n j , the algorithm selects the neighbor’s sense which has the highest WordNet similarity score (wnss) with...

Ngày tải lên: 08/03/2014, 02:21

... paper, I will argue that a hybrid representation formalism is best suited for the knowledge I need to represent. 2 A hybrid action representa- tion formalism As I have argued elsewhere based ... Action representation for NL instructions Barbara Di Eugenio* Department of Computer and Information Science University of Pennsylvania Philadelphia, ... figures, we should start lookin...

Ngày tải lên: 17/03/2014, 08:20

Ngày tải lên: 18/03/2014, 02:20