![]() Additionally, we applied statistical models on the UNHCR dataset to forecast the number of incoming refugees per country. Next, we trained a Convolutional Neural Network (CNN) to identify tweets describing refugee movements, including various facts mentioned in the tweet text (refugee-related keywords, origin, destination, means of transport and number of refugees). Through this, we aimed to detect collective refugee movements and hot spots. For this research, we acquired tweets through spatial and temporal queries before filtering the dataset with keywords and performing a spatio-temporal analysis on this subset. Then, we lay the foundation of our analysis by providing exploratory analysis results on refugee-related Twitter data (tweets) to show that the collected Twitter dataset can be used to capture real-world events. The paper starts with reviewing related work. It demonstrated the need for accurate information that can be collected quickly and at any time throughout humanitarian crisis situations. This disorganisation was caused by the limited communication between those involved and the inability to collect information. Especially at the beginning of 2015, refugee groups of varying sizes arrived at the border at all hours of the day without prior notification. Besides these practical problems, public authorities and relief organisations had to deal with a lack of information about refugee movements, making the organisation of efficient and effective humanitarian aid virtually impossible. Local public authorities and relief organisations in these countries faced the extreme logistical challenge of providing for such a large number of arriving refugees. In this period, the refugee movements led to challenging situations in the affected countries as the movements along the Balkan route did not result from a coordinated decision among these countries but instead developed gradually. From 2015 until March 2016, 1.2 million people moved along this route. One of the primary paths to Europe was the informal “Balkan route” that spans from Turkey to Central Europe through Greece, North Macedonia, Albania, Kosovo, Montenegro, Serbia, Bosnia and Herzegovina, Slovenia, and Hungary. ![]() ![]() Ongoing conflicts in the Near East led to mass refugee movements to Central Europe in 20 when more than 2.5 million people applied for asylum in member states of the European Union. We demonstrate that the approach proposed in this article benefits refugee management and vastly improves the status quo. Furthermore, our approach enables us to forecast and simulate refugee movements to forecast an increase or decrease in the number of incoming refugees and to analyse potential future scenarios. The results include spatial patterns and factual information about collective refugee movements extracted from social media data that match actual movement patterns. The approach combines methods to analyse the textual, temporal and spatial features of social media data and the number of arriving refugees of historical refugee movement statistics to provide relevant and up to date information about refugee movements and expected numbers. Therefore, we propose an approach utilising machine learning methods and publicly available data to provide more information about refugee movements. At the time, public authorities and relief organisations struggled with the admission, transfer, care, and accommodation of refugees due to the information gap about ongoing refugee movements. In 2015, within the timespan of only a few months, more than a million people made their way from Turkey to Central Europe in the wake of the Syrian civil war.
0 Comments
Leave a Reply. |