Acquisition of Hyponymy Relations from Wikipedia | (2008) but LFl—LF5 and SFl-SFQ are the same as their feature set . |
Acquisition of Hyponymy Relations from Wikipedia | Let us provide an overview of the feature sets used in Sumida et al. |
Acquisition of Hyponymy Relations from Wikipedia | These are the feature sets used in Sumida et al. |
Motivation | Since the learning settings ( feature sets , feature values, training data, corpora, and so on) are usually different in two languages, the reliable part in one language may be overlapped by an unreliable part in another language. |
Experimental Comparison with Unsupervised Learning | With this feature set , the CRF model is less expressive than DMV. |
Experimental Comparison with Unsupervised Learning | The CRF cannot consider valency even with the full feature set , but this is balanced by the ability to use distance. |
Experimental Comparison with Unsupervised Learning | First we note that GE training using the full feature set substantially outperforms the restricted feature set , despite the fact that the same set of constraints is used for both experiments. |
Machine learning-based cache model | Therefore, the intra-sentential and inter-sentential zero-anaphora resolution models are separately trained by exploiting different feature sets as shown in Table 2. |
Machine learning-based cache model | Table 1: Feature set used in the cache models |
Machine learning-based cache model | The feature set used in the cache model is shown in Table l. The ‘CASE_MARKER’ feature roughly captures the salience of the local transition dealt with in Centering Theory, and is also intended to capture the global foci of a text coupled with the BEGINNING feature. |
Conclusion and Outlook | Conceputalizing MT evaluation as an entailment problem motivates the use of a rich feature set that covers, unlike almost all earlier metrics, a wide range of linguistic levels, including lexical, syntactic, and compositional phenomena. |
Expt. 2: Predicting Pairwise Preferences | Feature set Consis- System-level tency (%) correlation (p) |
Introduction | (2005)), and thus predict the quality of MT hypotheses with a rich RTE feature set . |
Regression-based MT Quality Prediction | (2007) train binary classifiers on a feature set formed by a number of MT metrics. |
Dependency Parsing with HPSG | Therefore, we extend this feature set by adding four more feature categories, which are similar to the original ones, but the dependency relation was replaced by the dependency backbone of the HP S G outputs. |
Dependency Parsing with HPSG | The extended feature set is shown in Table 1. |
Dependency Parsing with HPSG | The extended feature set is shown in Table 2 (the new features are listed separately). |
Dependency Parsing: Baseline | With notations defined in Table l, a feature set as shown in Table 2 is adopted. |
Dependency Parsing: Baseline | We used a large scale feature selection approach as in (Zhao et al., 2009) to obtain the feature set in Table 2. |
Evaluation Results | The results with different feature sets are in Table 4. |
Evaluation Results | Table 4: The results with different feature sets features with p without p |
Dependency parsing for machine translation | The three feature sets that were used in our experiments are shown in Table 2. |
Dependency parsing for machine translation | It is quite similar to the McDonald (2005a) feature set , except that it does not include the set of all POS tags that appear between each candidate head-modifier pair (i , j). |
Dependency parsing for machine translation | The primary difference between our feature sets and the ones of McDonald et a1. |
Corpus Details | However, stopwords were retained in the feature set as various sociolinguistic studies have shown that use of some of the stopwords, for instance, pronouns and determin-ers, are correlated with age and gender. |
Corpus Details | Also, only the ngrams with frequency greater than 5 were retained in the feature set following Boulis and Ostendorf (2005). |
Related Work | Another relevant line of work has been on the blog domain, using a bag of words feature set to discriminate age and gender (Schler et al., 2006; Burger and Henderson, 2006; Nowson and Oberlander, 2006). |
Conclusions | Future research directions include developing rich feature sets and using corpus level or external information. |
Experiments | Since different feature sets , NLP tools, etc are used in different benchmarked systems, we are also interested in comparing the proposed algorithm with different soft relational clustering variants. |
Experiments | With the same feature sets and distance function, KARC-S outperforms FRC in F score by about 5%. |
Discussion | The held-out results in Figure 2 suggest that the combination of syntactic and lexical features provides better performance than either feature set on its own. |
Evaluation | At most recall levels, the combination of syntactic and lexical features offers a substantial improvement in precision over either of these feature sets on its own. |
Evaluation | No feature set strongly outperforms any of the others across all relations. |