Extractive summarization using complex networks and syntactic dependency

The realization that statistical physics methods can be applied to analyze written texts represented as complex networks has led to several developments in natural language processing, including automatic summarization and evaluation of machine translation. Most importantly, so far only a few metrics of complex networks have been used and therefore there is ample opportunity to enhance the statistics-based methods as new measures of network topology and dynamics are created. In this paper, we employ for the first time the metrics betweenness, vulnerability and diversity to analyze written texts in Brazilian Portuguese. Using strategies based on diversity metrics, a better performance in automatic summarization is achieved in comparison to previous work employing complex networks. With an optimized method the Rouge score (an automatic evaluation method used in summarization) was 0.5089, which is the best value ever achieved for an extractive summarizer with statistical methods based on complex networks for Brazilian Portuguese. Furthermore, the diversity metric can detect keywords with high precision, which is why we believe it is suitable to produce good summaries. It is also shown that incorporating linguistic knowledge through a syntactic parser does enhance the performance of the automatic summarizers, as expected, but the increase in the Rouge score is only minor. These results reinforce the suitability of complex network methods for improving automatic summarizers in particular, and treating text in general. (C) 2011 Elsevier B.V. All rights reserved.

The place of (explicit) grammar within the methods and approaches of foreign language teaching: in the north, in the centre or on the margins?

This article analyzes the role that has been attributed to grammar throughout the history of foreign language teaching, with special emphasis on methods and approaches of the twentieth century. In order to support our argument, we discuss the notion of grammar by proposing a conceptual continuum that includes the main meanings of the term which are relevant to our research. We address as well the issue of "pedagogical grammar" and consider the position of grammar in the different approaches of the "era of the methods" and the current "post-method condition" in the field of language teaching and learning. The findings presented at the end of the text consist of recognizing the central role that grammar has played throughout the history of the methods and approaches, where grammar has always been present by the definition of the contents' progression. The rationale that we propose for this is the recognition of the fact that the dissociation between what is said and how it is said can not be more than theoretical and, thus, artificial.

Mini-Mental State Examination sentence writing among community-dwelling elderly adults in Brazil: text fluency and grammar complexity

Background: In normal aging, the decrease in the syntactic complexity of written production is usually associated with cognitive deficits. This study was aimed to analyze the quality of older adults' textual production indicated by verbal fluency (number of words) and grammatical complexity (number of ideas) in relation to gender, age, schooling, and cognitive status. Methods: From a probabilistic sample of community-dwelling people aged 65 years and above (n = 900), 577 were selected on basis of their responses to the Mini-Mental State Examination (MMSE) sentence writing, which were submitted to content analysis; 323 were excluded as they left the item blank or performed illegible or not meaningful responses. Education adjusted cut-off scores for the MMSE were used to classify the participants as cognitively impaired or unimpaired. Total and subdomain MMSE scores were computed. Results: 40.56% of participants whose answers to the MMSE sentence were excluded from the analyses had cognitive impairment compared to 13.86% among those whose answers were included. The excluded participants were older and less educated. Women and those older than 80 years had the lowest scores in the MMSE. There was no statistically significant relationship between gender, age, schooling, and textual performance. There was a modest but significant correlation between number of words written and the scores in the Language subdomain. Conclusions: Results suggest the strong influence of schooling and age over MMSE sentence performance. Failing to write a sentence may suggest cognitive impairment, yet, instructions for the MMSE sentence, i.e. to produce a simple sentence, may limit its clinical interpretation.

Incorporating label dependency into the binary relevance framework for multi-label classification

In multi-label classification, examples can be associated with multiple labels simultaneously. The task of learning from multi-label data can be addressed by methods that transform the multi-label classification problem into several single-label classification problems. The binary relevance approach is one of these methods, where the multi-label learning task is decomposed into several independent binary classification problems, one for each label in the set of labels, and the final labels for each example are determined by aggregating the predictions from all binary classifiers. However, this approach fails to consider any dependency among the labels. Aiming to accurately predict label combinations, in this paper we propose a simple approach that enables the binary classifiers to discover existing label dependency by themselves. An experimental study using decision trees, a kernel method as well as Naive Bayes as base-learning techniques shows the potential of the proposed approach to improve the multi-label classification performance.