Last updated: 2021-06-25

Checks: 7 0

Knit directory: thesis-predicting-conflict/analysis/

This reproducible R Markdown analysis was created with workflowr (version 1.6.2). The Checks tab describes the reproducibility checks that were applied when the results were created. The Past versions tab lists the development history.

Great! Since the R Markdown file has been committed to the Git repository, you know the exact version of the code that produced these results.

Great job! The global environment was empty. Objects defined in the global environment can affect the analysis in your R Markdown file in unknown ways. For reproduciblity it’s best to always run the code in an empty environment.

The command set.seed(20210321) was run prior to running the code in the R Markdown file. Setting a seed ensures that any results that rely on randomness, e.g. subsampling or permutations, are reproducible.

Great job! Recording the operating system, R version, and package versions is critical for reproducibility.

Nice! There were no cached chunks for this analysis, so you can be confident that you successfully produced the results during this run.

Great job! Using relative paths to the files within your workflowr project makes it easier to run your code on other machines.

Great! You are using Git for version control. Tracking code development and connecting the code version to the results is critical for reproducibility.

The results in this page were generated with repository version b1528ac. See the Past versions tab to see a history of the changes made to the R Markdown and HTML files.

Note that you need to be careful to ensure that all relevant files for the analysis have been committed to Git prior to generating the results (you can use wflow_publish or wflow_git_commit). workflowr only checks the R Markdown file, but you know if there are other scripts or data files that it depends on. Below is the status of the Git repository when the results were generated:

Ignored files:
    Ignored:    .Rproj.user/
    Ignored:    data/DB/
    Ignored:    data/raster/
    Ignored:    data/raw/
    Ignored:    data/vector/
    Ignored:    docker_command.txt
    Ignored:    output/acc/
    Ignored:    output/bayes/
    Ignored:    output/ffs/
    Ignored:    output/models/
    Ignored:    output/plots/
    Ignored:    output/test-results/
    Ignored:    renv/library/
    Ignored:    renv/staging/
    Ignored:    report/presentation/

Note that any generated files, e.g. HTML, png, CSS, etc., are not included in this status report because it is ok for generated content to have uncommitted changes.

These are the previous versions of the repository in which changes were made to the R Markdown (analysis/index.Rmd) and HTML (docs/index.html) files. If you’ve configured a remote Git repository (see ?wflow_git_remote), click on the hyperlinks in the table below to view the files as they were in that past version.

File Version Author Date Message
Rmd b1528ac Darius Görgen 2021-06-25 wflow_publish(“analysis/index.Rmd”)
html c7a27c2 Darius Görgen 2021-05-12 Build site.
Rmd 7fd4ff2 Darius Görgen 2021-05-12 workflowr::wflow_publish(files = list.files(“.”, “Rmd$”))
Rmd 3cc16e8 Darius Görgen 2021-05-06 wflow_publish(files = rmds, dry_run = F)
Rmd 641d928 Darius Görgen 2021-05-06 wflow_publish(files = rmds, dry_run = F)
Rmd de8ce5a Darius Görgen 2021-04-05 add content
html de8ce5a Darius Görgen 2021-04-05 add content

1 Predicting violent conflict in Africa - Leveraging open geodata and deep learning for spatiotemporal event detection

1.1 Abstract

Violent conflicts endanger human lives, the social cohesion of societies and the natural environment. While the number of intensive international conflicts has remained on a low level during the 21st century, civil wars are on the rise. Since the 1990s, research engages in predicting the outbreak of violence. However, findings on the role the natural environment plays in the emergence of violence remain mostly inconclusive. In order to contribute to the discussion this thesis sets out to compare the predictive performance of deep learning models using data from the Uppsala Conflict Data Program (UCDP) on civil conflict between 2001 to 2019. The data is simultaneously aggregated on administrative districts and sub-basin watersheds and combined with socio-economic and environmental predictors. The hyperparameters of CNN-LSTM architectures are optimized employing a Bayesian Optimization strategy. The results in terms of F2-score suggest significant improvements for aggregating predictors on sub-basin watersheds (+7.16,p=3.4e-11) as well as integrating environmental predictors (+3.98,p=5.9e-05) for a combined conflict class. For other conflict classes, the results tend to the same direction but are not significant. Through the comparison to existing conflict prediction tools, the thesis exposes the sensitivity of prediction models to spatial scale and units of aggregation. It is argued that in order to fulfill the requirements of effective conflict prevention efforts, prediction research will have to fully integrate modern deep learning frameworks and constant data streams on different earth processes in the future.

1.2 Graphical Abstract

1.3 Disclaimer

This thesis was submitted to the Department of Geography, University of Marburg, in partial fulfillment of the requirements for the degree of M.Sc. Phyisical Geography. It was written by customizing the {huwiwidown} template of the Berlin School of Buisness and Economics, HU Berlin. The online version of the thesis just wraps the original R Markdown files which were written to produce LaTex and builds a workflowr project out of it. By the conversion to html some outputs my render not totally as expected. However, you can download the original pdf version from here. All code available in this repository is licensed under GPL-3.

1.3.1 Assets

  • An online version of the thesis powered by {workflowr} can be found here

  • The original version of the thesis as a pdf can be found here, an occasionally updated version revising minor errors is found here

  • Slides for the final presentation are powered by {xaringan} and are found here

R version 4.1.0 (2021-05-18)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Ubuntu 20.04.2 LTS

Matrix products: default
BLAS:   /usr/lib/x86_64-linux-gnu/blas/
LAPACK: /usr/lib/x86_64-linux-gnu/lapack/

 [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C              
 [3] LC_TIME=de_DE.UTF-8        LC_COLLATE=en_US.UTF-8    
 [7] LC_PAPER=de_DE.UTF-8       LC_NAME=C                 
 [9] LC_ADDRESS=C               LC_TELEPHONE=C            

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
[1] stringr_1.4.0  metathis_1.0.3

loaded via a namespace (and not attached):
 [1] bookdown_0.22     workflowr_1.6.2   Rcpp_1.0.6        rprojroot_2.0.2  
 [5] digest_0.6.27     later_1.2.0       R6_2.5.0          git2r_0.28.0     
 [9] magrittr_2.0.1    evaluate_0.14     stringi_1.6.2     rlang_0.4.11     
[13] fs_1.5.0          promises_1.2.0.1  whisker_0.4       rmarkdown_2.9    
[17] tools_4.1.0       glue_1.4.2        purrr_0.3.4       httpuv_1.6.1     
[21] xfun_0.24         yaml_2.2.1        compiler_4.1.0    htmltools_0.5.1.1
[25] knitr_1.33