Spatio-Temporal Machine Learning Analysis of Social Media Data and Refugee Movement Statistics

Abstract

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. 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. Therefore, we propose an approach utilising machine learning methods and publicly available data to provide more information about refugee movements. 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. The results include spatial patterns and factual information about collective refugee movements extracted from social media data that match actual movement patterns. 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. We demonstrate that the approach proposed in this article benefits refugee management and vastly improves the status quo.

Drowning in the Information Flood: Machine-Learning-Based Relevance Classification of Flood-Related Tweets for Disaster Management

“…We retrieved Tweets using both the REST and the streaming API of Twitter, via which georeferenced data can be accessed. In our data collection approach, we followed [18,43].…”

“…One possibility is to extract explicitly georeferenced Tweets, most of which become geocodable via a "place" tag set by the user for the respective Tweet. Numerous studies have already shown that significant statements on geo-social phenomena can be derived from this selection [17][18][19][20][21]. A major difficulty, however, is reducing the large, rather unstructured amount of data for each use case, i.e., to consider only relevant Tweets.…”

Sinirlarin Ötesi̇ndeki̇ Di̇ji̇tal İzler: Göç Kri̇zi̇ Üzeri̇ne Si̇stemati̇k Bi̇r Derleme

“…In a study in which immigration movements from the Near East towards Central Europe between 2015 and 2016 were examined, the hot spots were mostly detected either on the routes of the refugees and immigrants towards Central Europe or the Western border of Turkey where most refugees and immigrants started their journey (Havas et al, 2021). Analyses revealed that refugees and immigrants with moderate and high socioeconomic status used a wide area, they travelled long distances, and those with low socioeconomic status seemed to be trapped in small neighbourhoods (Kilic et al, 2019).…”

Clustering-Based Joint Topic-Sentiment Modeling of Social Media Data: A Neural Networks Approach

“…To extract meaningful information from the myriad of posts at hand, computational methods such as topic modeling [3,4] and sentiment analysis [1,2,5] have been researched and adapted specifically for short-form textual social media data. In practice, this allows for the analysis and monitoring of a variety of real-world phenomena such as earthquakes [6,7], floodings [8,9], refugee movements [10] or disease outbreaks [11][12][13] through social media data.…”