Tulajdonnevek felismerése az Európai Média Monitor magyar moduljának fejlesztésében

Júlia Pajzs

doi:10.29178/NevtErt.2015.17

Júlia Pajzs MTA Nyelvtudományi Intézet

DOI: https://doi.org/10.29178/NevtErt.2015.17

Kulcsszavak: Európai Média Monitor, hírportál, nyelvtechnológia, névfelismerés

Absztrakt

Recognition of proper names in the development of the Hungarian module of the European Media Monitor

The European Media Monitor (http://emm.newsbrief.eu) developed by the European Joint Research Centre collects and categorizes news items automatically from several thousands of news sites all over the world, 24 hours a day, updating every 10 minutes, using the tools and techniques of language technology. The full range of the service has been fitted for use in Hungarian. News items of international relevance are available in all languages involved in the project. This paper presents the attempts and achievements concerning the recognition of Hungarian proper names and the treatment of their suffixed forms within the European Media Monitor system.