NATIONAL VIEW
/ Armed Conflict Risk
The figures show our past and present forecast for the risk that a country suffers an outbreak of armed conflict within the next twelve months, i.e. that the country goes from no fatalities to over 0.5 fatalities per one million inhabitants within a time horizon of twelve months.
We provide two estimates based on two different forecast models. The BEST estimate is based on both past and present conflict dynamics and newspaper articles written on the respective country. The TEXT model only uses the newspaper articles to predict conflict and disregards conflict dynamics. For guidelines in using these estimates see the technical description
Every point in the figures is a past forecast which uses the information available at that time. For example, the forecast in August 2015 uses the information available in August 2015 to predict the likelihood of an outbreak of armed conflict 12 months ahead (in September 2015 to August 2016 in the example).
/ Topics
Our forecast uses millions of newspaper articles to make the conflict forecast. In our analysis of the content of the newspaper articles we rely on a so-called topic model which summarizes the millions of articles and words into topics using unsupervised machine learning. The topic model allows us to calculate 15 topic shares for each country/month which we display in the bubbles to the right.
The model assumes that if journalists write about a topic, say politics, they will use a different vocabulary than if they write about another topic, say conflict. The word “congress”, for example, is more likely to be used in an article about politics than in an article about conflict. Inside the bubbles to the right you can see a small part of this topic-specific vocabulary (the top thirty terms in each topic).
Our forecast method re-estimates the topic model every month and so the content of the topics changes across times. You can play around with the time bar to see the flow of topic shares and topic content over time.
Importance of topic/terms
Decreasing risk
Neutral
Increasing risk
No data available yet.