engleski

XLike: Cross-lingual Knowledge Extraction

The goal of the XLike project is to develop technology to monitor and aggregate knowledge that is currently spread across mainstream and social media, and to enable cross-lingual services for publishers, media monitoring and business intelligence. Knowledge resources from Linked Open Data cloud will be used, with special focus paid to using general common sense knowledge base CycKB as “semantic Interlingua”. The use of Machine Translation will be oriented towards translation from a natural language as the source language into a formal language of semantic representation as source language. For languages where not enough required linguistic resources are available, we will use a probabilistic Interlingua representation trained from a parallel corpora and/or from comparable corpus derived from the Wikipedia, or use MT as a fallback option for translating the text from less resourced language to English and then process this translation.

knowledge extraction; machine translation; natural language; formal language

nije evidentirano