Báo cáo khoa học: "Computing and Evaluating Syntactic Complexity Features for Automated Scoring of Spontaneous Non-Native Speech" pot

... 722–731,Portland, Oregon, June 19-24, 2011.c2011 Association for Computational Linguistics Computing and Evaluating Syntactic Complexity Features for Automated Scoring of Spontaneous Non-Native ... extracting and evaluating features related to syntactic complexity of spontaneous spoken responses as part of an effort to expand the current feature set of an automated speech scoring system ... model of language ability, and therefore is often used as an index of language proficiency and development status of L2 learners. Various studies have proposed and inves-tigated measures of syntactic...

Báo cáo khoa học: "Diversify and Combine: Improving Word Alignment for Machine Translation on Low-Resource Languages" docx

... eventually thetranslation performance by producing and combining complementary word align-ments for low-resource languages. Instead of focusing on the improvement of a singleset of word alignments, ... the test set for the systems built on eachset of word alignment in Table 1. Similar to theobservation from Table 1, c2outperforms c0 and c1, and B + V + S + P with c2outperformsB + V ... align-ment quality and the translation performance. Wegenerated multiple sets of diversiﬁed alignmentsbased on linguistics, morphology, and heuris-tics, and demonstrated the effectiveness of com-bination...

Báo cáo khoa học: Upstream and intronic regulatory sequences interact in the activation of the glutamine synthetase promoter pot

... gastricantrum. Examples of the second group are adipocytes and muscle cells (for a review, see [1]).Because of these interorgan differences in distribution and cellular concentration of GS, and because ... degree of activation of the GS promoter and that these interactions differ quantitatively between cellsfrom hepatic and muscular origin.Materials and methodsSequence of the ﬁrst intron of GSThe ... concentration of foetal bovine serumwas reduced to 1% to induce the formation of myotubes.Harvest of the cells was carried out 48 h after transfectionin the case of FTO-2B cells, and 72 h in the case of...

Báo cáo khoa học: " Robust and Flexible Mixed-Initiative Dialogue for Telephone Services" pot

... Shows means results for each Group in high ASR situations and for both in low ASR. Table 2 shows mean results for each Group A and B of users for High ASR performance, and for all users in Low ... designed for the telephone line and for a wide range of potential users. Therefore our evalua- tions will be done taking into account different lev- els of speech recognition performance and user ... obtain different levels of recognition per- formance for a vocabulary of 1170 words: 96.4% word recognition rate for high performance and 80% for low performance. A pre-defined single fixed mixed-initiative...

Báo cáo khoa học: Trehalose and anhydrobiosis in tardigrades – evidence for divergence in responses to dehydration ppt

... quantiﬁcation of trehalose because of the highsensitivity and accuracy of determination. However,due to the sensitivity of the analysis technique for sam-ple size and the small amount of protein ... any species of Heterotardigrada so far investigated. Theseresults expand our current understanding of anhydrobiosis in tardigrades and, for the ﬁrst time, demonstrate the accumulation of trehalose ... macromolecules and cellu-lar structures in place of water during dehydration and thus preserves native structures. In addition, the vitriﬁ-cation hypothesis proposes the formation of amor-phous...

Báo cáo khoa học: Mycobacterium tuberculosis possesses a functional enzyme for the synthesis of vitamin C, L-gulono-1,4-lactone dehydrogenase doc

... after long (3 h) (lanes 2 and 3) and short (1 h) (lanes4 and 5) periods of induction with IPTG. Fractions obtained with along period of induction before (lane 2) and after (lane 3) enterokin-ase ... an optical density of 0.8 at600 nm in 200 mL of culture volume, and then IPTG wasadded to a ﬁnal concentration of 0.1 mm for induction, and the fusion protein was produced for 1 h or 3 h, as ... concen-tration, the 70 kDa band disappeared, and two newbands of 45 and 26 kDa were seen on SDS ⁄ PAGEthat corresponded to the mycobacterial Rv1771 pro-tein and the freed GST tag, respectively...

Báo cáo khoa học: "Joint Learning Improves Semantic Role Labeling Kristina Toutanova Dept of Computer Science Stanford " pot

... These features are a subset of features used in previouswork. The standard features at the top of the tablewere deﬁned by (Gildea and Jurafsky, 2002), and the rest are other useful lexical and ... between the la-bel of a node and input features of other argumentnodes. The features are speciﬁed by instantiation of templates and the value of a feature is the number of times a particular ... Features: Another very effective class of fea-tures we deﬁned are features that look at the label of a single argument node and internal features of otherargument nodes. The idea of these features...

Báo cáo khoa học: " Extending Latent Semantic Analysis with features for dialogue act classiﬁcation" pot

... meaning of a word as a kind of average of the meanings of all the passages in which it ap-pears, and the meaning of a passage as a kind of average of the meaning of all the words it contains(Landauer ... sparser: for MapTask, the number of rows increases from 1,835 for plain LSA to 2,324 for FLSA.These negative results on adding syntactic infor-mation to LSA may just reinforce one of the claims of ... respectively for CallHome and for MapTask. Finally, for both LSA and FLSA, Table 4includes, in parenthesis, the dimension k of the re-duced semantic space. For each LSA method and corpus, we...

Báo cáo khoa học: "A Hierarchical Pitman-Yor Process HMM for Unsupervised Part of Speech Induction" doc

... its base distribition auniform distribution over the set of tags, while thepriors for Bj and Tijback off by discarding an item of context. This allows the modelling of trigramtag sequences, ... sentence (ditto for a ﬁnal end of sentence marker) and the uniform base distributionranges over the set of characters. We expect thatthe HMM+LM model will outperform the uniformHMM as it can ... CohnDepartment of Computer ScienceUniversity of ShefﬁeldT.Cohn@dcs.shef.ac.ukAbstractIn this work we address the problem of unsupervised part -of- speech inductionby bringing together several strands of research...

Tài liệu Báo cáo khoa học: "A Uniﬁed Syntactic Model for Parsing Fluent and Disﬂuent Speech∗" ppt

... dual of a left-corner transform. It transforms all right recursivesequences in each tree into left recursive sequences of symbols of the form A1/A2, denoting an incom-plete instance of category ... saves more potential exam-ples of ﬂuent speech rules, and therefore potentiallymakes better use of limited data.3 EvaluationThe evaluation of this system was performed onthe Switchboard corpus, ... speech repairs at the word level and above.One salient aspect of structure is the fact that thereis often a good deal of overlap in words betweenthe reparandum and the alteration, as speakers...

Xem thêm