Tài liệu Báo cáo khoa học: "Learning Word Senses With Feature Selection and Order Identification Capabilities" pdf

... Learning Word Senses With Feature Selection and Order Identification Capabilities Zheng-Yu Niu, Dong-Hong Ji Institute for Infocomm ... of second order context vectors: to select better feature words in contexts to construct better second order context vectors enabling better feature selection. Since the sense associated with a word s ... counting second order co...

Ngày tải lên: 20/02/2014, 16:20

... Chinese Word Segmentation without Using Lexicon and Hand-crafted Training Data Sun Maosong, Shen Dayang*, Benjamin K Tsou** State Key Laboratory of Intelligent Technology and Systems, ... Chinese word segmentation developed so far, both statistical and rule-based, exploited two kinds of important resources, i.e., lexicon and hand-crafted linguistic resources(manually segmen...

Ngày tải lên: 20/02/2014, 18:20

... of salient words ag- gregated using synonymy, similarity, or subtrees of a thesaurus. However, salient word selection and aggregation is non-obvious and furthermore it falls into word sense disambiguation, ... de- fined with TARGET, thus this frequent token is also included in F . We use the set of frequent words F to generalize words to word classes”. We define a word class as...

Ngày tải lên: 20/02/2014, 04:20

... par- ticular word occurs. The hyper-parameters of the model are the regularization weights (λ and ν), and the word vector dimensionality β. Maximizing the objective function with respect to R, b, ψ, and ... results on a standard dataset, and introduce a new dataset for the task. In both tasks we com- pare our model’s word representations with several bag of words weighting m...

Ngày tải lên: 20/02/2014, 04:20

... # of features, and metrics used. Our MT models are trained with standard phrase-based Moses software (Koehn and others, 2007), with IBM M4 alignments, 4gram SRILM, lexical ordering for PubMed and ... metrics using machine learning for better correlation with human judgments (Liu and Gildea, 2007; Albrecht and Hwa, 2007; Gimnez and M ` arquez, 2008) and may give insights fo...

Ngày tải lên: 19/02/2014, 19:20

... for clustering word instances, which is used in the multi-prototype ver- sion of our model that accounts for words with mul- tiple senses. We evaluate our new model on the standard WordSim-353 (Finkelstein ... only local context and one represen- tation per word. This is problematic because words are often polysemous and global con- text can also provide useful information for lear...

Ngày tải lên: 19/02/2014, 19:20

... many-to-one word alignments, where each source word is aligned with zero or one target words, and therefore each target word can be aligned with many source words. Each source word is labelled with ... include indicator features for an exact string match, both with and without vowels, and the edit-distance between the source and target words as a real- valued feature....

Ngày tải lên: 20/02/2014, 11:21

... classes and randomly drew 100 documents for each class. We used unigrams and bigrams as features and ran our method for various hash sizes m (Figure 1). The number of unigrams is 33,017 and bigrams ... parameters. Let us explain in detail. In NLP, features can be often expediently expressed with strings. For in- stance, a feature ‘the current word ends with -ing’ can be expres...

Ngày tải lên: 20/02/2014, 05:20

... coherence. Hybrid word/ sub -word recognizers can produce a sequence of sub -word units in place of OOV words. Ideally, the recognizer outputs a complete word for in-vocabulary (IV) utterances, and sub -word ... hybrid system’s lexicon has 83K words and 5K or 10K sub-words. Note that the word vocabulary is com- mon to both systems and only the sub-words are se- lected using eith...

Ngày tải lên: 20/02/2014, 04:20

... tagging and parsing, and measures of selectional preference and argument structure as complementary features for the classi- fier. Finally, our task-based evaluation, verb clustering with Levin ... clusters are then compared to the gold standard clusters with the purity-based F-Score from Sun and Korhonen (2009) and the more familiar Adjusted Rand Index (Hubert and Arabie, 1985...

Ngày tải lên: 19/02/2014, 19:20