This project represents a comprehensive, multi-dimensional Exploratory Data Analysis (EDA) of the Global Terrorism Index (GTI), compiled by the Institute for Economics & Peace (IEP). Encompassing 1,793 country-year records spanning 163 nations, this study investigates the shifting geopolitical landscape of global terrorism over an 11-year longitudinal period (2012–2022).
By processing 47,873 distinct incidents resulting in 95,993 fatalities, this repository serves as a robust analytical framework for national security analysts, geopolitical risk consultants, and humanitarian policy-makers aiming to understand tactical correlations, regional escalation matrices, and long-term conflict trajectories.
- Temporal Trajectory Mapping: Quantify the global escalation and subsequent de-escalation of terrorist incidents and fatalities from 2012 to 2022.
- Geospatial Concentration: Isolate the geographical epicenter of global terrorism, tracking the migration of conflict from the Middle East/South Asia to the African Sahel and Southeast Asia.
- Tactical Correlation Modeling: Utilize statistical matrices to determine the predictability of mass casualties (Fatalities vs. Injuries) versus specialized asymmetrical tactics (Hostage operations).
- Country-Specific Deep Dives: Conduct comparative time-series analyses of the most impacted nations (e.g., Iraq vs. Afghanistan) to evaluate the efficacy of domestic counter-terrorism countermeasures.
The primary dataset aggregates intelligence from Terrorism Tracker and open-source IEP databases. It represents 99.7% of the global population.
- Volume: 1,793 rows × 9 dimensions.
- Integrity: Audited for data hygiene yielding 0% missing/null values across the 11-year timeline.
- Key Features:
iso3c/Country: Geographical identifiers.Score: GTI impact metric (0 = No Impact, 10 = Highest Measurable Impact).Incidents,Fatalities,Injuries,Hostages: Tactical casualty metrics.
Analysis of the temporal data reveals a distinct bell curve of lethality peaking in the mid-2010s.
- Total Recorded Incidents: 47,873
- Total Global Fatalities: 95,993
- Deadliest Year Globally: 2015, recording an unprecedented 10,881 fatalities, followed by 2016 (10,372 fatalities) and 2013 (10,317 fatalities).
- Recent Trends: While total incidents peaked again in 2021 (5,443 incidents), lethality has trended downward (6,701 fatalities in 2022), suggesting a shift toward higher-frequency, lower-casualty asymmetric warfare.
A heavy concentration of incidents is isolated to a handful of sovereign states. Iraq and Afghanistan alone account for >32% of all recorded global attacks.
| Global Rank | Country | Total Incidents | Total Fatalities | Total Injuries | Hostages | Avg GTI Score |
|---|---|---|---|---|---|---|
| 1 | Iraq | 11,183 | 21,286 | 38,381 | 1,771 | 9.33 |
| 2 | Afghanistan | 4,443 | 12,726 | 21,263 | 4,101 | 8.95 |
| 3 | Pakistan | 3,358 | 7,692 | 14,060 | 969 | 8.50 |
| 4 | Somalia | 2,951 | 7,209 | 6,740 | 828 | 8.24 |
| 5 | India | 2,872 | 2,062 | 3,048 | 932 | 7.60 |
| 6 | Nigeria | 1,718 | 10,191 | 6,089 | 1,701 | 8.50 |
Note: Nigeria exhibits an exceptionally high Fatality-to-Incident ratio compared to India, highlighting variations in regional tactical lethality.
By computing the net shift in incidents over the 10-year gap, we observe a massive geographical migration of terrorism.
- The De-escalation Zone (Middle East & South Asia): Extensive military campaigns have yielded significant incident reductions in Iraq (-887 incidents), Pakistan (-340), and Afghanistan (-282).
- The Escalation Zone (Southeast Asia & The Sahel): Severe destabilization is actively occurring in Myanmar (+747 incidents), Burkina Faso (+310), and Mali (+256), indicating the new frontiers of global terrorism.
To understand terrorist methodology, a correlation matrix was applied to the four tactical metrics:
-
Fatalities & Injuries (
$r = 0.913$ ): Near-perfect positive correlation. Incidents designed to yield high fatalities reliably produce mass injuries (e.g., explosive devices in dense civilian areas). -
Incidents & Fatalities (
$r = 0.860$ ): Strong direct correlation. -
The Hostage Anomaly (
$r = 0.282$ ): Hostage counts show minimal correlation with mass casualty metrics. This proves that hostage-taking and kidnapping-for-ransom operations are tactically isolated events, utilized for strategic leverage rather than immediate lethality.
- Data Manipulation:
pandas,numpy(Aggregations, Pivot Tables, Time-Series Grouping) - Statistical Visualization:
matplotlib.pyplot,seaborn(Correlation Heatmaps, Dual-Axis Line Plots, Categorical Bar Charts)
- Data Ingestion & Hygiene: Preprocessing 1.7K+ records for zero-null integrity.
- Time-Series Analysis: Granular year-over-year mapping of global incidents.
-
Shift Calculus: Pivot-table operations calculating net-escalation Deltas (2012
$\rightarrow$ 2022). - Comparative Analysis: Dual-line plotting of Iraq vs. Afghanistan conflict curves.
- Statistical Modeling: Seaborn-driven Pearson correlation heatmaps.
Based on the empirical data, the following recommendations are provided for national security and international humanitarian bodies:
- Resource Reallocation to the Sahel: The data definitively proves conflict is migrating from the Middle East to Sub-Saharan Africa (Mali, Burkina Faso). International peacekeeping and counter-terrorism funding must pivot geographically to match this emerging threat vector.
- Asymmetric Tactical Preparation in Myanmar: The unprecedented spike in incidents in Myanmar (+747) requires immediate geopolitical intervention and localized intelligence gathering.
- Differentiated Response Protocols: Because hostage operations do not correlate with mass-casualty bombings, defense agencies must maintain specialized, surgically-trained hostage rescue units distinct from standard counter-insurgency military forces.
Osman Ahmed Khan
- Role: Data Analyst & Software Engineer
- Institution: Malla Reddy University
- Program Affiliation: Edunet Foundation & VOIS for Tech (AICTE Internship)
- Connect: GitHub Profile | LinkedIn
"In God we trust, all others must bring data." — W. Edwards Deming