This paper addresses the task of automatic extraction of definitions by thoroughly exploring an approach that solely relies on machine learning techniques, and by focusing on the issue of the imbalance of relevant datasets. We obtained a breakthrough in terms of the automatic extraction of definitions, by extensively and systematically experimenting with different sampling techniques and their combination, as well as a range of different types of classifiers. Performance consistently scored in the range of 0.95-0.99 of area under the receiver operating characteristics, with a notorious improvement between 17 and 22 percentage points regarding the baseline of 0.73-0.77, for datasets with different rates of imbalance. Thus, the present paper also represents a contribution to the seminal work in natural language processing that points toward the importance of exploring the research path of applying sampling techniques to mitigate the bias induced by highly imbalanced datasets, and thus greatly improving the performance of a large range of tools that rely on them.

Abstract. In this paper we present a rule-based system for automatic extraction of definitions from Portuguese texts. As input, this system takes text that is previously annotated with morpho-syntactic information, namely on POS and inflection features. It handles three types of definitions, whose connector between definiendum and definiens is the so-called copula verb "to be", a verb other that one, or punctuation marks. The primary goal of this system is to act as a tool for supporting glossary construction in e-learning management systems. It was tested using a collection of texts that can be taken as learning objects, in three different domains: information society, computer science for non experts, and e-learning. For each one of these domains and for each type of definition typology, evaluation results are presented. On average, we obtain 14% for precision, 86% for recall and 0.33 for F 2 score.

This paper presents the description of 12 systems submitted to the WMT16 IT-task, covering six different languages, namely Basque, Bulgarian, Dutch, Czech, Portuguese and Spanish. All these systems were developed under the scope of the QTLeap project, presenting a common strategy. For each language two different systems were submitted, namely a phrasebased MT system built using Moses, and a system exploiting deep language engineering approaches, that in all the languages but Bulgarian was implemented using TectoMT. For 4 of the 6 languages, the TectoMT-based system performs better than the Moses-based one.

Content Management Systems (CMSs) are known for their ability for storing data, both structured and non-structured data. However they are not able to associate meaning and context to the stored information. Furthermore, these systems do not meet the needs and expectations of their users, because as the size of data increases, the system loses its capacity of retrieving meaningful results. In order to overcome this issue, we propose a method to implement data contextualization on a CMS. The proposed method consists of enriching the data with semantic information, allowing a more accurate retrieval of results. The implementation of this approach was validated by applying this contextualization method to a currently used CMS with real information. With this improved CMS, it is expected that the users will be able to retrieve data related to their initial search.