A Semantic Web Approach for Visualization-Based News Analytics
In order to understand news, dependency patterns between objects in (economic) news items have to be detected. We propose a framework which makes it possible to discover these patterns, and support the observations with statistical analysis. Based on these patterns, alerts can be generated based on emerging news. These alerts can then be used to manage (equity) portfolios. We test our framework based on historical data. The tests show statistically significant results supporting the idea that it is possible to discover such dependency patterns between objects in news items.