STICS has been developed at the Max Planck Institute for Informatics and already presented as a Demo at SIGIR 2014. By extending the Google slogan of "things, not strings" to support also entity categories, STICS provides powerful functionality for querying and analyzing news and other text corpora in terms of entities, semantic classes, and text phrases. STICS is based on state-of-the-art methods for named entity recognition and disambiguation (AIDA [1]), linking them to knowledge bases like YAGO and the Wikipedia category system. The online service currently has indexed 1,000,000 news articles since June 2013, with more than 22 million entity occurrences of 300,000 distinct entities. This has been added to the tools section of Research Resources Subject Tracer™ Information Blog. This will be added to Academic and Scholar Search Engines and Sources white paper.
posted by Marcus Zillman |
4:56 AM
