The aim of the xLiMe project is to extract information from various media platforms and in different languages in order to produce a cross-lingual and cross-channel knowledge and data base.
This process runs in near real time in order to obtain constant updates and a comprehensive overview of information distribution across the media, e.g. from a European community in Catalonia to worldwide distribution as English-language content. The solution to currently open research problems is being developed through a combination of speech recognition, natural language processing, machine learning and semantic technologies:
- extracting automated, machine-readable information and knowledge (entities, sentiment, events & opinions) from multilingual, multichannel social media content, integration into the cross-lingual, cross-channel knowledge and data base,
- automated search queries for information via structured and unstructured queries in near real time,
- monitoring the source, need and distribution,
- Analysis of interaction between medial exposure and the behavioural pattern.