Material Detail

This video was recorded at Workshop on Applications of Pattern Analysis (WAPA), Cumberland Lodge 2010. The Europe Media Monitor (EMM) is the text gathering and analysis engine underlying a number of European media monitoring and other information analysis applications (e.g. EMM (http://emm.newsbrief.eu), MediSys (http://medisys.newsbrief.eu) ) that are serving EU policies especially those concerned with crisis management. The EMM engine consists of a growing number of text analysis and information processing modules currently performing the following tasks: language detection, known entity extraction, geo-tagging, sentiment analysis, categorization, duplicate detection, clustering, event detection and indexing. The system has a number of information aggregation modules to present the analysis results per category, per country, as a story etc. In the case of the EMM NewsBrief, the system harvests and analyses around 100.000 news articles per day in 40 languages from around 5500 RSS feeds and HTML pages, and categorizes these in approximately 1000 different categories defined by 35000 different keywords and keyword combinations. The system is developed and operated by the European Commission's Joint Research Centre (JRC). The presentation will focus on the history, development and architecture of the system.

Disciplines:

• Science and Technology  / Computer Science