Word embeddings are used with success for a variety of tasks involving lexical semantic similarities between individual words.Using unsupervised methods and just cosine similarity, encouraging results were obtained for analogical similarities.In this paper, we explore the potential of pre-trained word embeddings to identify generic types of semantic relations in an unsupervised experiment.We propose a new relational similarity measure based on the combination of word2vec's CBOW input and output vectors which outperforms alternative vector representations, when used for unsupervised clustering on SemEval 2010 Relation Classification data.

The Légilocal project aims to help local authorities to improve the quality, interoperability and publication of French local administrative acts in the same way as Légifrance does at the state and EU level. The originality of the approach is to unify the management of contents and the interactions between actors on these contents.The Légilocal platform combines various tools (content management, networking, semantic annotation and search) and resources to assist clerks in the drafting and publication of local acts.