Word sense disambiguation scaling up, domain adaptation and application to machine translation

... for Word Sense Disambiguation 19 2.2 Domain Adaptation for Word Sense Disambiguation 23 2.3 Word Sense Disambiguation for Machine Translation 24 Our Word Sense Disambiguation ... WORD SENSE DISAMBIGUATION: SCALING UP, DOMAIN ADAPTATION, AND APPLICATION TO MACHINE TRANSLATION CHAN YEE SENG (B.Computing (Hons.), NUS) A THESIS SUBMITTED FOR...

Ngày tải lên: 11/09/2015, 16:08

... equivalent to several machine- language instructions, and they refer to m e m o r y locations by names called variables." One point is N M P analysis method by recursive calling for case frame analysis ... obligatory prepositional phrase name if the verb is intransitive Furthermore, the control moves to the next case slot to fill i t , i f the case frame has more slots,...

Ngày tải lên: 31/03/2014, 17:20

... 4.2 In -Domain and Out-of -Domain Evaluation 47 4.2.1 47 Training and Evaluating on OntoNotes ii 4.2.2 4.3 Using Out-of -Domain Training Data 49 Concatenating ... attempt to use the existing training data of one word as the training data for other words Kohomban and Lee (2005) tried to use training examples of words different from the actual word to be cla...

Ngày tải lên: 09/09/2015, 10:06

... class membership estimates, for the instances in the target domain These probabilities were then used by the machine learning methods to estimate the sense priors of each word in the target domain ... of predominant sense is often indicative of a change in domain, as different corpora drawn from different domains usually give different predominant senses For example,...

Ngày tải lên: 08/03/2014, 02:21

... active learning for domain adaptation, followed by count-merging Next, we describe an EMbased algorithm to estimate the sense priors in the new domain Performance of domain adaptation using active ... adaptation examples added (%) Figure 3: Using true predominant sense for the nouns the sense priors in WSJ Using this new set of training examples, we perform domain adapta...

Ngày tải lên: 08/03/2014, 02:21

... Upper and Lower Bounds 5.1 Lower Bounds W e could be in a better position to address the question o f the relative difficulty of interest if we could establish a rough estimate of the upper and ... lower bounds on the level of performance that can be expected W e will estimate the lower bound by evaluating the performance of a straw man system,...

Ngày tải lên: 17/03/2014, 08:20

... neighbour for the chosen word and then assigns a sense based on the word, its neighbour and the topic Boyd-Graber and Blei (2007) test their method on WSD and information retrieval tasks and find ... corrected any bad tags and lemmas for the target instances.4 Sense Paraphrases For word sense disambiguation tasks, the paraphrases of the sense keys are represent...

Ngày tải lên: 23/03/2014, 16:20

... 2006 An Empirical Study of the Behavior of Active Learning for Word Sense Disambiguation, Proc of the main conference on Human Language Technology Conference of the North American Chapter of ACL, ... A combination of active learning and semi-supervised learning starting with positive and unlabeled examples Proposed Algorithm At the beginni...

Ngày tải lên: 23/03/2014, 17:20

... of the text in which the word is located is a crucial information for WSD For example the (domain) polysemy among the C OM PUTER S CIENCE and the M EDICINE senses of the word virus can be solved ... abstraction and term similarity for word sense disambiguation: Irst at senseval-3 In Proc of SENSEVAL-3 Third International Workshop on Evaluation of Systems for the Semantic Analy...

Ngày tải lên: 23/03/2014, 19:20

... as selected from among the class of decomposable models Decomposable models are a subclass of log-linear models and, as such, can be used to characterize and study the structure of data ([2]), ... for formulating decomposable models and Section describes the details of the case study performed to test the approach The results of the disambiguation case study are discussed...

Ngày tải lên: 23/03/2014, 20:21

... a domain specific sense) in the sense tagged corpus and there is a large amount of untagged corpora that contain instances for both general senses and the missed sense, then a sense tagger built ... predefined sense inventories for target words The information for semi-supervised sense disambiguation is usually obtained from bilingual corpora (e.g parallel corpora...

Ngày tải lên: 12/09/2015, 11:05

... instances in the five supported languages Online machine translations tools have already been used before to create artificial parallel corpora that were used for NLP tasks such as for instance Named ... features related to the focus word itself being the word form of the focus word, the lemma, Part-of-Speech and chunk information • local context features related to a window...

Ngày tải lên: 20/02/2014, 05:20

... more specific the sense is The fifth level often contains a few words or only one word, which is called an atom word group, an atom class or an atom node The words in the same atom node hold the ... artificial ambiguous word need to simulate the function of the real ambiguous word, and to acquire semantic knowledge as the real ambiguous word does Thus, we call it an equivalent...

Ngày tải lên: 20/02/2014, 12:20

... Mihalcea and E Faruque 2004 SenseLearner: Minimally supervised word sense disambiguation for all words in open text In Proceedings of ACL/SIGLEX Senseval-3, Barcelona, Spain, July R Mihalcea and ... an efficient algorithm for minimally supervised word- sense disambiguation that attempts to disambiguate all content words in a text using WordNet senses The results ob...

Ngày tải lên: 20/02/2014, 15:20

... the Senseval-2 sense classes for the target word “art”, from WordNet 1.7 (Fellbaum 1998) Class Sense the creation of beautiful or significant things a superior skill large, a condition where kernel ... transformed via the trained PCA and KPCA parameters The PCA- based and KPCA-based sense class predictions disagree t 9 Observed vectors (x1 , x2 , x3 ) t t t (1, 0, 1) (1, 0, 1)...

Ngày tải lên: 20/02/2014, 16:20