Abstract | Our results show that summaries biased by dependency pattern models lead to significantly higher ROUGE scores than both n-gram language models reported in previous work and also Wikipedia baseline summaries. |
Discussion and Conclusion | Our evaluations show that such an approach yields summaries which score more highly than an approach which uses a simpler representation of an object type model in the form of a n-gram language model. |
Introduction | a corpus of descriptions of churches, a corpus of bridge descriptions, and so on) and reported results showing that incorporating such n-gram language models as a feature in a feature-based extractive summarizer improves the quality of automatically generated summaries. |
Introduction | The main weakness of n-gram language models is that they only capture very local information aboutshofitennsequencesandcannotnnxkfllong distance dependencies between terms. |
Introduction | If this information is expressed as in the first line of Table l, n-gram language models are likely to |
Representing conceptual models 2.1 Object type corpora | We derive n-gram language and dependency pattern models using object type corpora made available to us by Aker and Gaizauskas. |
Representing conceptual models 2.1 Object type corpora | 2.2 N-gram language models |
Representing conceptual models 2.1 Object type corpora | they calculate the probability that a sentence is generated based on a n-gram language model. |
Summarizer | o LMSim3 : The similarity of a sentence S to an n-gram language model LM (the probability that the sentence S is generated by LM). |
Background | The computation of t/I(e,H(v)) is based on a linear combination of a set of n-gram consensuses-based features. |
Background | For each order of n-gram, h; (e,H(v)) and h”— (e,H(v)) are defined to measure the n-gram agreement and disagreement between e and other translation candidates in H(v), respectively. |
Background | If p orders of n-gram are used in computing t//(e,H(v)) , the total number of features in the system combination will be T +2>< p (T model-score-based features defined in Equation 8 and 2x p consensus-based features defined in Equation 9). |
How Frequent is Unseen Data? | The third line across the bottom of the figure is the number of unseen pairs using Google n-gram data as proxy argument counts. |
How Frequent is Unseen Data? | Creating argument counts from n-gram counts is described in detail below in section 5.2. |
Models | Using the Google n-gram corpus, we recorded all verb-noun co-occurrences, defined by appearing in any order in the same n-gram , up to and including 5-grams. |
Models | For instance, the test pair (throwsubject, ball) is considered seen if there exists an n-gram such that throw and ball are both included. |
Models | C (vd, n) as the number of times 2) and n (ignoring d) appear in the same n-gram . |
Results | The Google n-gram backoff model is almost as good as backing off to the Erk smoothing model. |
Distributed representations | For each training update, we read an n-gram x = (W1, . |
Distributed representations | We also create a corrupted or noise n-gram J? |
Distributed representations | 0 We corrupt the last word of each n-gram . |
Introduction | Aside from including dependency and n-gram relations in the scoring, we also apply and evaluate SemPOS for English. |
Problems of BLEU | Table 1 estimates the overall magnitude of this issue: For 1-grams to 4-grams in 1640 instances (different MT outputs and different annotators) of 200 sentences with manually flagged errors4, we count how often the n-gram is confirmed by the reference and how often it contains an error flag. |
Problems of BLEU | Fortunately, there are relatively few false positives in n-gram based metrics: 6.3% of unigrams and far fewer higher n-grams. |
Automated Classification | Web 1T N-gram Features |
Automated Classification | Table 3 describes the extracted n-gram features. |
Automated Classification | The influence of the Web lT n-gram features was somewhat mixed. |