Discussion | As expected, the structural methods on either skewed or flattened hierarchies are not significantly better than the flat SVM . |
Discussion | For the flattened hierarchy of 15 leaf genres the maximal accuracy is 54.2% vs. 52.4% for the flat SVM (Figure 3), a nonsignificant improvement. |
Experiments | As a baseline we use the accuracy achieved by a standard "flat" SVM. |
Experiments | A standard flat SVM achieves an accuracy of 64.4% whereas the best structural SVM based on Lin’s information content distance measure (IC-lin-word-bnc) achieves 68.8% accuracy, significantly better at the 1% level. |
Experiments | Table 1 summarizes the best performing measures that all outperform the flat SVM at the 1% level. |
Genre Distance Measures | The structural SVM (Section 2) requires a distance measure h between two genres. |
Structural SVMs | To strengthen the constraints, the zero value on the right hand side of the inequality for the flat SVM can be replaced by a positive value, corresponding to a distance measure h(yi, m) between two genre classes, leading to the following constraint: |
Evaluation | We compared the Gibbs sampling compressor (GS) against a version of maximum a posteriori EM (with Dirichlet parameter greater than 1) and a discriminative STSG based on SVM training (Cohn and Lapata, 2008) ( SVM ). |
Evaluation | EM is a natural benchmark, while SVM is also appropriate since it can be taken as the state of the art for our task.4 |
Evaluation | Nonetheless, because the comparison system is a generalization of the extractive SVM compressor of Cohn and Lapata (2007), we do not expect that the results would differ qualitatively. |
Introduction | We achieve substantial improvements against a number of baselines including EM, support vector machine ( SVM ) based discriminative training, and variational Bayes (VB). |
Unsupervised Mining of Personal and Impersonal Views | We apply both support vector machine ( SVM ) and Maximum Entropy (ME) algorithms with the help of the SVM-light4 and Mallet5 tools. |
Unsupervised Mining of Personal and Impersonal Views | We find that ME performs slightly better than SVM on the average. |
Unsupervised Mining of Personal and Impersonal Views | Transductive SVM , which seeks the largest separation between labeled and unlabeled data through regularization (Joachims, 1999). |
Experience Detection | While we tested several classifiers, we chose to use two different classifiers based on SVM and Logistic Regression for the final experimental results because they showed the best performance. |
Experience Detection | Logistic Feature Regression SVM |
Experience Detection | Logistic Feature Regression SVM |
Lexicon Construction | ME SVM Prec. |
Experiments and Results | We employ an SVM coreference resolver trained and tested on ACE 2005 with 79.5% Precision, 66.7% Recall and 72.5% F1 to label coreference mentions of the same named entity in an article. |
Incorporating Structural Syntactic Information | And thus an SVM classifier can be learned and then used for recognition. |
Introduction | Section 4 introduces the frame work for discourse recognition, as well as the baseline feature space and the SVM classifier. |
The Recognition Framework | The classifier learned by SVM is: |
The Recognition Framework | One advantage of SVM is that we can use tree kernel approach to capture syntactic parse tree information in a particular high-dimension space. |
Experimental Evaluation and Discussion | We set the degree of the kernels to 3 since cubic kernels with SVM have proved effective for Japanese dependency parsing (Kudo and Matsumoto, 2000; Kudo and Matsumoto, 2002). |
Experimental Evaluation and Discussion | Stopping Criteria It is known that increment rate of the number of support vectors in SVM indicates saturation of accuracy improvement during iterations of active learning (Schohn and Cohn, 2000). |
Experimental Evaluation and Discussion | It is interesting to examine whether the observation for SVM is also useful for support vectors7 of the averaged perceptron. |
Substructure Spaces for BTKs | In the 1st phase, a kernel based classifier, SVM in our study, is employed to classify each candidate subtree pair as aligned or unaligned. |
Substructure Spaces for BTKs | Since SVM is a large margin based discriminative classifier rather than a probabilistic model, we introduce a sigmoid function to convert the distance against the hyperplane to a posterior alignment probability as follows: |
Substructure Spaces for BTKs | We use SVM with binary classes as the classifier. |
Experimental Setup | We learned the feature weights with a linear SVM , using the software SVM-OOPS (Woodsend and Gondzio, 2009). |
Experimental Setup | For each phrase, features were extracted and salience scores calculated from the feature weights determined through SVM training. |
Experimental Setup | The distance from the SVM hyperplane represents the salience score. |
Comparing the two Datasets | SMO is an implementation of Support Vector Machines ( SVM ), rules.JRip is the RIPPER algorithm, and bayes .NaiveBayes is a Naive Bayes classifier. |
Comparing the two Datasets | In task C, the SVM algorithm was also the best performing algorithm among those that were also tried on the English data, but decision trees produced even better results here. |
Comparing the two Datasets | The results are: in task A the lazy.KStar classifier scored 58.6%, and the SVM classifier scored 75.5% in task B and 59.4% in task C, with trees . |
Empirical Evaluation | SVM ) should be separately created with regards to distinct features. |
Empirical Evaluation | We utilised SVanl‘;8 as an implementation of the Ranking SVM algorithm, in which the parameter c was set as 1.0 and the remaining parameters were set to their defaults. |
Reference Resolution using Extra-linguistic Information | Although the work by Denis and Baldridge (2008) uses Maximum Entropy to create their ranking-based model, we adopt the Ranking SVM algorithm (J oachims, 2002), which learns a weight vector to rank candidates for a given partial ranking of each referent. |
Multilingual Subjectivity System | Previous studies have found that, among several ML-based approaches, the SVM classifier generally performs well in many subjectivity analysis tasks (Pang et al., 2002; Banea et al., 2008). |
Multilingual Subjectivity System | An SVM score (a margin or the distance from a learned decision boundary) with a positive value predicts the input as being subjective, and negative value as objective. |
Multilingual Subjectivity System | The second and the third approaches are carried out as follows: Corpus-based (T-CB): We translate the MPQA corpus into the target languages sentence by sentence using a web-based service.6 Using the same method for S-CB, we train an SVM model for each language with the translated training corpora. |
Experimental setup | We use a Ranking SVM (Si/Mug“ (Joachims, 2002)) to score summaries using our features. |
Experimental setup | The Ranking SVM seeks to minimize the number of discordant pairs (pairs in which the gold standard has :31 ranked strictly higher than :52, but the learner ranks x2 strictly higher than :01). |
Experimental setup | For system-level evaluation, we treat the real-valued output of the SVM ranker for each summary as the linguistic quality score. |