The UPV/EHUs IXA Group helps to improve the use of the colossal digital library Europeana
The IXA Group of the UPV/EHU-University of the Basque Country has helped to improve the use of the colossal digital library Europeana. The Library currently has 15 million content items in the field of culture in a range of formats (images, texts, audio and video sequences). The IXA Group has enriched all this content by means of language technologies in order to facilitate users’ searches and offer more comprehensive information. This work is part of the European project known as PATHS (Personal Access To cultural Heritage Spaces). During the next academic year these technologies, as well as others, will be covered at the UPV/EHU in the Master's Degree Course on Language Analysis and Processing. Automatic language processing is essential to be able to select information properly in any Internet search. As part of the European PATHS project, the UPV/EHU’s IXA Group has, by means of language technologies, analysed and added to the content of the huge Europeana digital library. This library is made up of books, films, paintings, newspapers, sound archives, maps, manuscripts and other kinds of archives. Today, it has 15 million digital items in a whole range of formats: images, texts, audio and video sequences. Some users may on occasions be confused by these vast numbers, because they may experience difficulties interpreting all the information they gather. So what the PATHS project does, among other things, is to offer guided tours to visit the cultural works contained in this library. “However, some of the cultural items (content) in these guided tours tend to have a limited description or provide little information,” explains Eneko Agirre, one of the IXA Group’s computer scientists, “and the user sometimes ends up wanting to know more.” The IXA group has enriched this cultural content by means of language technologies. So, similar content that might be of interest is suggested to the user; he/she is offered a list of similar items thanks to neat organisation by subject; for external resources, for example, Wikipedia links are added; and in addition, the user receives help so the he/she can interpret the information more easily. The IXA group researchers have used various techniques: machine learning, linguistic, morphological and syntactic analysis, text corpora, terminology extraction, text comprehension, disambiguation, etc. All these techniques will in fact be covered during the University Master’s Degree course on Language Analysis and Processing which the IXA Group will be running at the UPV/EHU-University of the Basque Country during the next academic year, 2013-2014.
Countries
Spain