Type: Exploratory Data Analysis (EDA) Contributor: Individual Dataset Size: 181,691 rows × 135 columns Tools Used: Python, Pandas, NumPy, Matplotlib, Seaborn, Google Colab
Terrorism is one of the most pressing global challenges, causing loss of life, economic disruption, and psychological trauma. This project dives deep into the Global Terrorism Database (GTD) to explore and uncover hidden patterns in terrorist activities over the last five decades (1970–2017).
📊 Goal: Extract actionable insights to support counter-terrorism strategies, smarter policymaking, and more efficient resource allocation using data.
The Global Terrorism Database (GTD) is an open-source dataset curated by the National Consortium for the Study of Terrorism and Responses to Terrorism (START), University of Maryland. It contains data on 180,000+ incidents of terrorism worldwide.
- 📍 Location: Country, Region, City
- 🎯 Targets: Civilians, Military, Police, Infrastructure
- 💣 Attack Types: Bombing, Armed Assault, Hostage Taking, etc.
- 🧠 Perpetrator Groups: Taliban, ISIL, Boko Haram, etc.
- 🔫 Weapons Used: Explosives, Firearms, Incendiaries, etc.
- 📊 Casualties: Killed (
nkill) and Wounded (nwound)
- 🔍 Identify patterns in terrorist attacks across time & regions
- 💡 Understand evolving tactics and attack types
- 🧑🤝🧑 Profile terrorist groups based on activity
- 📉 Indirectly assess counter-terrorism effectiveness via trends
- 🧠 Inform decisions on threat levels and resource deployment
- ✅ Handled encoding errors while loading data
- 🔧 Selected relevant columns:
country_txt,region_txt,attacktype1_txt,gname,nkill,nwound, etc. - ✍️ Renamed columns for clarity (e.g.,
iyear→year,nkill→no_of_kills) - 🧼 Checked for duplicates and missing values
- 🔢 Filtered down to high-impact insights
(Imagine each visualization coming alive through animated charts)
A sharp rise in terrorist incidents post-2010 📈 [Line chart with year-wise spike]
Bombing/Explosion tops the list by far 📊 [Bar chart shooting up like an explosion]
Middle East & North Africa (27.8%) and South Asia (24.8%) are the most impacted 🥧 [Pie chart animating slice-by-slice]
Private Citizens & Property are the most affected 👥 [Stacked bar chart with icons for homes, soldiers, police]
Positive correlation (r = 0.53) shows attacks with more fatalities also cause more injuries 🌡️ [Heatmap glowing with intensity]
Besides "Unknown", Taliban and ISIL dominate recent years (2008–2017) 👤 [Horizontal bar chart with group names pulsing]
Iraq, Pakistan, Afghanistan, and India top the charts 📍 [Map-style visual or ranked country flags]
Sustained attacks post-2008 with fluctuations in casualties 📉 [Line chart focused on India, animated by year]
- 🎯 Targeted Allocation: Prioritize high-risk zones like Iraq, Afghanistan, South Asia
- 🕵️ Intelligence Focus: Strengthen surveillance on groups like ISIL, Taliban
- 🛡️ Infrastructure Security: Fortify frequently targeted areas — public spaces, transport, etc.
- 📣 Community Preparedness: Launch public awareness & emergency response programs
- 🌍 International Coordination: Collaborate globally to share threat intelligence
- Python – Core programming language
- Pandas – Data cleaning and manipulation
- NumPy – Numerical analysis
- Matplotlib & Seaborn – Data visualization
- Google Colab – Jupyter notebook environment in the cloud
- Develop predictive models using ML (e.g., attack forecasting)
- Build an interactive dashboard (using Tableau/Power BI/Streamlit)
- Extend dataset with post-2017 information for trend continuation
This project demonstrates how data can serve as a powerful weapon in the fight against terrorism. The insights uncovered here are a small but meaningful step towards making our world a safer place.