Discriminating antonyms and synonyms is an important NLP task that has the difficulty that both, antonyms and synonyms, contains similar distributional information. Consequently, pairs of antonyms and synonyms may have similar word vectors. We present an approach to unravel antonymy and synonymy from word vectors based on a siamese network inspired approach. The model consists of a two-phase training of the same base network: a pre-training phase according to a siamese model supervised by synonyms and a training phase on antonyms through a siamese-like model that supports the antitransitivity present in antonymy. The approach makes use of the claim that the antonyms in common of a word tend to be synonyms. We show that our approach outperforms distributional and patternbased approaches, relaying on a simple feed forward network as base network of the training phases.

We describe the TEMANTEX annotation scheme for temporal expressions and other lexical indicators of temporality and we analyze a first annotation experience. TEMANTEX is mainly a revision of the markup language TIMEX3, but with some additions and a different treatment for relative expressions. Our alternative proposal is justified for two reasons. First, our system aims to cover other temporality-related lexical elements by defining annotations for what we call temporal indicators, which do not have an equivalent in the TimeML system. Second, regarding temporal expressions, our scheme has relevant differences that improve the annotation process and the interpretation potential. A first task of corpus annotation on a set of 2.300 words, comprising 33 temporal expressions and 35 temporal indicators, showed encouraging results.

No abstract

No abstract

Nowadays, many approaches for Sentiment Analysis (SA) rely on affective lexicons to identify emotions transmitted in opinions. However, most of these lexicons do not consider that a word can express different sentiments in different predication domains, introducing errors in the sentiment inference. Due to this problem, we present a model based on a context-graph which can be used for building domain specic sentiment lexicons(DL: Dynamic Lexicons) by propagating the valence of a few seed words. For different corpora, we compare the results of a simple rule-based sentiment classier using the corresponding DL, with the results obtained using a general affective lexicon. For most corpora containing specic domain opinions, the DL reaches better results than the general lexicon.