From scarcity to bounty: how Galateas can turn your scarce short queries into gold
Frédérique Segond, Eduard Barbu, Igor Barsanti, Bogomil Kovachev, Nikolaos Lagos, Marco Trevisan, Ed Vald · UvA-DARE (University of Amsterdam) · 2012
With the growth of digital libraries and the digital library federation in addition to partially unstructured collections of documents such as web sites, a large set of vendors are offering engines for retrieving content and metadata via search requests by the end user (queries). In most cases these queries are short unstructured fragments of text in different languages that are difficult to make sense of because of the lack of context. When attempting to perform automatic translation of these queries, using machine learning approaches, the problem becomes worse as aligned corpora are almost inexistent for such types of linguistic data. The GALATEAS European project concentrates on analyzing language-based information from transaction logs and facilitates the development of improved navigation and search technologies for multilingual content access, in order to offer digital content providers an innovative approach to understanding users' behaviour.