Decoding Chaos

Our Project Proposal (Revised)

Published

March 10, 2024

Modified

March 31, 2024

Note: This is a revised proposal from initial project proposal to incorporate Professor Kam’s advice and suggestions to adopt ACLED dataset.

Authors: Imran, Suan Ern, Zachary Wong

Motivation

“In contrast to the 1970s, much of today’s political violence is aimed at people instead of property.” - Reuters (2023)

Armed conflicts due to political violence and coordinated attacks targeting innocent civilians, have been on the rise globally. According to International Crisis Group, there was a fresh wave of major combats in the recent years whereby one of the conflict (which was close to our nation - Singapore) was prompted by Myanmar army’s 2021 power grab. The International Crisis Group has listed Myanmar (one of the countries in South-East Asia) as the top 10 conflicts to watch in 2024. This threatens the public at both physical and psychological levels, and raised concerns among the Defence and Security sectors on the urgency to develop effective counter terrorism measures and strategies.

The team aims to use open-source data from armed conflict events (Armed Conflict Location & Event Data Project (ACLED)) to provide easy-to-use and insightful visualisation tools that can be especially suitable to help (1) discover armed conflicts trends and (2) conceptualise armed conflict spaces.

Objectives

The objective of this project is to:

  1. Examine open-source data on armed conflict incidents (Armed Conflict Location & Event Data Project (ACLED)) and derive useful insights on the factors and conditions that may affect armed conflict incidents,

  2. Conceptualise armed conflict spaces in Myanmar, and

  3. Build a web-enabled interactive visual analytics application by using R Shiny based on the dataset.

Data

The project will examine the dataset from Armed Conflict Location & Event Data Project (ACLED), specifically Myanmar country, between Year 2010 and Year 2023. The dataset consists of 55,574 observations and 35 variables. Each row details the armed conflict event on the type, agents, location, date and other characteristics of conflict events (such as political violence, demonstration) in Myanmar.

The flowchart diagram below provides an overview of the key variables used in this project.

flowchart TD
  A(Key Variables Used \n event_id_cnty)
  A --> B(Time Period)
  A --> C(Characteristic of Incident)
  A --> D(Location)
  
  B --> E(year)
  B --> F(date)

  C --> G(event_type)
  C --> H(sub_event_type)
  C --> J(fatalities)
  C -.-> K(New Variables)
  K -.-> L(total incidents)
  K -.-> M(total fatalities)
  
  D --> N(country)
  D --> O(longitude)
  D --> P(latitude)
  D --> Q(admin1)
  D --> R(admin2)
  D -.-> S(New Variables)
  S -.-> T(geometry points)
  S -.-> U(shapeID)

Methodology & Analytical Approach

For this project, the following approach will be taken by the team:

  1. Data Preparation: Clean the dataset by removing irrelevant variables and filtering variables.

  2. Aspatial Analysis: Analyse the dataset to understand the aspatial distribution of variables, and identify patterns between variables at different Administration Region Levels.

    1. Distribution Analysis (proportional symbol map, choropleth map, line chart, density-ridge plot)

  3. Geospatial Data Analysis: Analyse the dataset to conceptualise armed conflict spaces within Myanmar.

    1. Clustering & Outlier Analysis

    2. Hot & Cold Zone Analysis

    3. Emerging hotspot analysis

  4. Confirmatory Data Analysis via Statistical Methods: Analyse the dataset using statistical testing tools to test hypotheses, evaluate the findings and arguments and make statistical observations under uncertainty.

    1. Oneway ANOVA Test

    2. Visualising Categorical Data (vcd)

Prototype

The team has decided to split the main prototype modules as follows:

Please refer to Prototype Tab (or respective modules) for more.

Storyboard

The team has conceptualised and developed a prototype to visually display the layouts of the R Shiny Application along with its UI and Server components.

Please refer to Storyboard page for more.

R Packages

The team intends to use the following R packages to run the R Shiny Application:

  • tidyverse: a family of modern R packages specially designed to support data science, analysis and communication task including creating static statistical graphs.

  • knitr: an report generation tool.

  • dplyr: an R tool for working with data frame (e.g. objects).

  • plotly: R library for plotting interactive statistical graphs.

  • DT: an R interface to the JavaScript library DataTables that create interactive table on html page.

  • lubridate: an R package that facilitates to use of dates and time elements.

  • ggforce: an extension of ggplot2 to provide visual data analysis with newer stats and geoms.

  • highcharter: a wrapper that contains ‘highcharts’ library for plotting of R objects.

  • sf: an R package that supports the importing, managing, and processing of geospatial data.

  • spdep: an R package that supports the handling of geospatial data.

  • tmap: an R package that provides layer-based creation of thematic maps.

  • leaflet: an R package that allows the creation and customisation of interactive maps.

  • ggstatsplot: an extension of ggplot2 package for statistical visualisation and graphics creation.

  • ggdist: an R package that support visualisation of distribution and uncertainty.

Back to top