Indexing scientific literature through semantic expansion – DELICES

Scientific digital libraries play a key role in the development and dissemination of the scientific literature. However, navigating and searching in these libraries still remains a laborious and time-consuming activity, owing to a shallow and inefficient indexing of scientific articles. This results in fragmentary and difficult access to the scientific literature. The goal of the project DELICES is to address this pitfall by exploiting semantic relations between scientific articles to improve and enhance indexing. To this end, DELICES relies on the latest advances in word embeddings and semantic comparison to both increase the relevance of extracted keyphrases and extend indexing to new terms from articles that are semantically similar.