In last decades because of the rising poverty and conflicts in developing regions such as Maghreb(Sub-Saharan Africa), Latin America and Middle East the immigration to Europe had been steadily increasing. In recent years, this tendency was accelerated by the revolutions and civil wars in Arab countries - so called Arab Spring and reached the crisis level (https://en.wikipedia.org/wiki/European_migrant_crisis).
In our report, we want to foucs on the topic of the refugee. To do so, first we start with exploring how the immigration in Europe changes by visulizing UNHCR's database. After getting a quantized results of the refugee's immigration, we want to get the oponions from different countries. we will look at the meadia's oponions and the origins of media by the year and the event related to refugee.
Further, it's also interesting we could classify the events. We could try to visualize the location that most related to those categories and the oponions related to those categories.
In general, we want to explore the refugee cirsis in European and how the oponions changed with the time and the different events happend.
How is the refugee's immigration changes in European countries with the time?
Which countries are the most important refugee destinations?
How are refugee movements associated with news?
How is the distribution of origin of media sources related to the refugee events?
What are the differences between Eastern and Western European countries' approach towards refugees?
How political state of country is related to its media coverage towards refugees?
How is the the media's opinion change with time?
Global Database of Events, Language, and Tone (GDELT) The dataset monitors and analyses news articles around the world from 1979 to present. It is important to mention that the dataset consists of two versions: the GDELT Knowledge Graph and the GDELT Event Database. We will use the simple sentiment algorithm of GDELT which is already precalculated and referred to 'tone' metadata on the dataset. Sentiment scores range from -100 to +100, where 0 is neutrality. However most of the news and events range from -10 to +10.
Wikipedia and UNHCR's database: After dealing with textual descriptions, we are planning extract the numerical data about the refugee crisis for the statistics from Wikipedia tables and UN Refugee Database(UNHCR).
Here is the Data Story.
Here is the Jupyter Notebook. It contains all the code of the analysis about what we have done and how we get there.
Mammad Hajili: Analysis on events and mention tables, combining the final code and writing the report.
Pavlo Karalupov: In charge of building the website for data story, the analysis on mention table and UNHCR's database.
Hongyu Luo: Analysis on GKG table and events tables, writing the report.