automatic verb classification incorporating selectional preferences

Ngày tải lên: 08/03/2014, 01:20

... to classify more verbs into a larger number of classes. In our automatic verb classification, we aim for a larger scale experiment. We select our experimental verb classes and verbs as follows: ... (2002). Automatic verb classification using a general feature space. Master’s thesis, University of Toronto. Joanis, E., Stevenson, S., and James, D. (2007). A general feature space for automatic verb ... on a small set of verbs or a small number of verb classes. For example, Schulte im Walde (2000) uses 153 verbs in 30 classes, and Joanis et al. (2007) takes on 835 verbs and 15 verb classes. Since...

Ngày tải lên: 20/02/2014, 09:20

Ngày tải lên: 23/03/2014, 14:20

Ngày tải lên: 30/03/2014, 23:20

Ngày tải lên: 22/06/2014, 23:20

Ngày tải lên: 23/03/2014, 17:20

Ngày tải lên: 31/03/2014, 01:20

... token at the head of the buffer, and pop the stack. 2.1 Classification Transition-based dependency parsing reduces parsing to consecutive multiclass classification. From each configuration one amongst ... in the MaltParser is to use a 2nd- degree polynomial kernel with the SVM. 3 Confidence-weighted classification Dredze et al. (2008) introduce confidence- weighted linear classifiers which are online- classifiers ... On the other hand if it has never been updated before the estimation is prob- ably very bad. CW classification deals with this by having a confidence-parameter for each weight, modeled by a Gaussian...

Ngày tải lên: 20/02/2014, 04:20

... algorithms for sentiment classification, SCL and SFA. Each set of bars represent a cross-domain sentiment classifica- tion task. The thick horizontal lines are in-domain sentiment classification accuracies. ... 19-24, 2011. c 2011 Association for Computational Linguistics Automatically Extracting Polarity-Bearing Topics for Cross-Domain Sentiment Classification Yulan He Chenghua Lin † Harith Alani Knowledge ... train- ing. Figure ?? shows the classification results on the five different domains by varying the number of top- ics from 1 to 200. It can be observed that the best classification accuracy is obtained...

Ngày tải lên: 20/02/2014, 04:20

... be effec- tive in text classification tasks (Joachims, 1998). We then apply a simple word filter based on POS tags to select content words (nouns, verbs, adjectives, and adverbs). In particular, ... sentiment classification using multi- ple source domains. Experimental results using a benchmark dataset for cross-domain sentiment clas- sification show that our proposed method can im- prove classification ... as the classification algo- rithm (Ng, 2004), which produces a sparse model in which most irrelevant features are assigned a zero weight. This enables us to select useful features for classification...

Ngày tải lên: 20/02/2014, 04:20

... learning of hierar- chical multilabel classification models. In Journal of Machine Learning Research. G. Salton, A. Wong, and CS Yang. 1975. A vector space model for automatic indexing. Communications ... hierarchical text classification with latent concepts. Experimental results show that the performance of our algorithm is com- petitive with the recently proposed hierarchi- cal classification algorithms. 1 ... criterions, e.g. “Entertainment”, “Sport- s” and “Education” in news classification, “Junk E- mail” and “Ordinary Email” in email classification. In the literature, many algorithms (Sebastiani, 2002; Yang...

Ngày tải lên: 20/02/2014, 05:20

... between verbs. The main idea is that asymmetric entailment relations between verbs can be analysed in the context of class-level and word-level selectional preferences (Resnik, 1993). Selectional preferences ... that selectional preferences like {player} win may be read as suggesting the entailment relation win(x) → play(x). Selectional preferences have been often used to infer semantic relations among verbs ... win. 3 Verb entailment: a classification The focus of our study is on verb entailment. A brief review of the WordNet (Miller, 1995) verb hierarchy (one of the main existing resources on verb entailment...

Ngày tải lên: 20/02/2014, 12:20

... of inference. The order is learned automatically, and partial output is in turn used to train the local clas- sifier. Therefore, the order of inference and the lo- cal classification are dynamically ... although we do not know which is the most desirable. In this approach, we can eas- ily collect the automatically generated negative sam- ples, and use them in learning. These negative sam- ples are ... search was employed. Our guided learning al- gorithm provides more flexibility in search with an automatically learned order. In addition, our treat- ment of the score of action and the score of...

Ngày tải lên: 20/02/2014, 12:20

... the above ker- 776 4.2 Answer classification Question classification does not allow to fully ex- ploit the PAS potential since questions tend to be short and with few verbal predicates (i.e. the ... state-of-the-art accuracy on question classification. (b) PB predicative structures are not effective for question classification but show promising results for answer classification on a cor- pus of answers ... the question but could not be judged as valid answers 5 . Answer classification results To test the impact of our models on answer classification, we ran 5-fold cross-validation, with the constraint...

Ngày tải lên: 20/02/2014, 12:20

... (unsegmented) is encoded in a similar way, but does not contain the class labels B and N. Finally, we automatically assign probability of 0.5 for unseen events. 4.3 Predicting word boundary with a classifier The ... information and it treats character strings as context which provides infor- mation on the possible classification of character- breaks as word-breaks. We are confident that once a standard model of ... Association for Computational Linguistics Rethinking Chinese Word Segmentation: Tokenization, Character Classification, or Wordbreak Identification Chu-Ren Huang Institute of Linguistics Academia Sinica,Taiwan churen@gate.sinica.edu.tw Petr ˇ Simon Institute...

Ngày tải lên: 20/02/2014, 12:20

... resolution. In Proc. of NAACL, pages 55–62. R. Iida, K. Inui, H. Takamura, and Y. Matsumoto. 2003. Incorporating contextual cues in trainable models for coreference resolution. In Proc. of the EACL ... a training or test instance, and the clustering algorithm used to coordinate the coref- erence classification decisions. Selecting a corefer- ence system, then, is a matter of instantiating these elements ... induction system (Quinlan, 1993), the RIPPER rule learning algorithm (Cohen, 1995), and maximum entropy classification (Berger et al., 1996). The classifica- tion model induced by each of these learners...

Ngày tải lên: 20/02/2014, 15:20

... function of d. Models that take this form are commonplace in classification. 2.3 Turney’s Classifier as Naive Bayes Although Naive Bayes classification requires a la- beled corpus of documents, we ... slightly better than the average pair. Whereas mean performance ranges from 37.3% to 39.6%, misclassification rates for this pair of anchors ranges from 37.4% to 38.1%. 4.3 A Smaller Unlabeled ... 0.40 Num. of Labeled Documents Classif. Error Misclassification versus Sample Size Figure 5: Misclassification with Labeled Docu- ments. The solid curve represents a latent fac- tor model with...

Ngày tải lên: 20/02/2014, 16:20

... O-polysaccharide of Proteus mirabilis G1 (Eur. J. Biochem. 269) 1409 Structure of the O-polysaccharide and classification of Proteus mirabilis strain G1 in Proteus serogroup O3 Zygmunt Sidorczyk 1 , Krystyna...

Ngày tải lên: 21/02/2014, 15:20

... types based on manual examina- tion of 50 fluency edit misclassifications and 50 factual edit misclassifications. leads to a small decrease in classification accu- racy, namely 86.68% instead of 87.14% ... have better understanding of errors made by the classifier, 50 fluency edit misclassifications and 50 factual edit misclassifications are ran- domly selected and manually examined. The er- rors are ... to recognize a fluency edit. 3 modify adjectives or adverbs that do not change the meaning such as “entirely” and “various”. Factual edit misclassifications: the big major- ity, 35 instances (70%),...

Ngày tải lên: 22/02/2014, 03:20

... POS N-grams to generate test data for their selectional preferences model, but not to infer preferences. Zhou et al. (2011) identified selectional preferences of one word for another Reference ... 1999 Verb- object Verb- subject none Target, relative, and relation in a parsed corpus (parsed BNC) none EM-based clustering Ritter, 2010 Verb- subject Verb- object Subject -verb- object ... Schmid, H. 2008. Combining Em Training and the Mdl Principle for an Automatic Verb Classification Incorporating Selectional Preferences. In Proceedings of the 46th Annual Meeting of the Association...

Ngày tải lên: 22/02/2014, 03:20