Termine

Aktuelles

UNIGIS Abschlussarbeiten


Der krönende Abschluss eines UNIGIS MSc Studiums ist sicherlich die Master Thesis. Mit ihr belegen unsere MSc-AbsolventInnen, dass sie den akademischen Grad "Master of Science (Geographical Information Science & Systems)" zu Recht führen.  Im UNIGIS professional Studiengang muss keine Abschlussarbeit verfasst werden. Dennoch nehmen einige Studierende die Möglichkeit war, ein Geoinformatikprojekt durchzuführen und entsprechend zu dokumentieren.

Sie sind auf der Suche nach aktueller Literatur zu Geoinformatik-Themen?
Hier finden sie die mitunter preisgekrönten Abschlussarbeiten unserer AbsolventInnen!

Johannes Reiter [04-2017]:

VGI in Disaster Management – Fusing Remote Sensing Data with User-Generated Data for Improved Flood Management

Diese Arbeit ist online verfügbar: Download


Various disasters like the severe German flood in 2003 keep demonstrating the vulnerability of human civilisation and infrastructure as well as the critical need of following up the information gap to offer spatial analyses to local decision-makers and international humanitarian mission. In the last few years Social Media services like Facebook and Twitter and their millions of followers have received high attention by research groups in relation to their situational awareness of occurring disasters. Local individuals not only consume but also produce valuable informations about disasters. However, the so-called Big Data, generated via these networks, is very challenging to analyse and to compute in a fair amount of time. While other research studies in the field of Disaster Management mainly concentrated on time-consuming keyword-searches, finding the relevant information, this thesis focused on similarity assessments in the form of a semantic probability-based topic model called Latent Dirichlet Allocation (LDA) (Blei et al., 2003). This unsupervised machine learning model identifies latent topics by clustering co-occurring words from a collection of Tweets. Furthermore, an analysis framework is presented providing the methods of extracting, organising, filtering and analysing the Tweets in near real-time within a developed application. Together with spatiotemporal filtering via the remote sensing data from the Center for Satellite Based Crisis Information (DLR Oberpfaffenhofen) the attempt is made to classify the flood related Tweets with the LDA algorithm. All calculations were assessed by a confusion matrix and further statistical analysis methods. The results of this thesis show that Social Media messages not only could be used for additional information sources on the crisis event itself, but also that LDA provides a stable overview in a fraction of time compared to manual or keyword-based filtering methods.


Zurck weitere Arbeiten ...