After the Rwandan Genocide of 1994, when the world stood by and watched the slaughter of 800,000 people, the United Nations stated that the world has an obligation to intervene and prevent ethnic cleansing. The UN said it would never allow that to happen again – but it has. In November 2016, the U.N. commission of human rights visited South Sudan, the world’s youngest country. They found a conflict marked by mass slaughter and what they described as a “warped environment” where the rape of women and girls has become normal. Since the civil war of South Sudan broke out in December 2013, more than 400,000 people have been killed. The situation in South Sudan reached an unprecedented level in 2017 as more than 2.1 million people have been forced to flee their homes. For people living in the country, life has became a total nightmare. The country gained its independence just eight years ago in a move that was supposed to bring peace to an area that had known only war. With the total numbers of displaced refugees rising, forced displacement is the most urgent issue that the United Nations Refugee Agency (UNCHR) must deal with as the current displacement situation reaches unprecedented levels and creates a grave humanitarian crisis.

As specified for this project, we will mainly use the provided datasets to discuss two objectives:
• How can machine learning assist in predicting human displacement in the country of South Sudan?
• What are the most influential factors that affect the displacement of persons of concern (POC)?

These instructions will get you a copy of the project up and running on your local machine for development and testing purposes. See deployment for notes on how to deploy the project on a live system.

1. Raw data folder include all origin data from the lecturer.
2. Files end with '_cleaned.csv' are data manipulated from the raw data given from the lecturer. For the manipulation process, check out .ipynb files.
3. PDFs are additional informations about the project, datasets.
4. All csv files has to be converted into arff with R (Check Covert_Data_To_Arff_R) in order to run in Weka.
5. Tableau Visualization files are included.
6. The main files used for model training is refugees_T_F_V_P_full_data.csv

1. Clone this repository
2. cd to the cloned directory in a terminal
3. Run ipython notebook in the directory

• iPython
• Jupyter Notebook

The datasets are provided by the United Nations Refugee Agency (UNCHR). More datasets can be find in: UNCHR Database

• Hau-Ben Benjamin Shih - Ben Contact [email protected]
• Yi-Hsiou Hsu - goodian12

See also the list of contributors who participated in this project.

• University of Twente
• UNCHR