akariri79/Road-Accidents-Kenya-2016-Analysis

This is a comprehensive data analysis on road acccidents in Kenya from a Kaggle Dataset with insights being drawn from key visualizations and correlation analysis. # Road-Accidents-Kenya-2016-Analysis This is a comprehensive data analysis on road acccidents in Kenya from a Kaggle Dataset with insights being drawn from key visualizations and correlation analysis. ## Road Accidents Analysis – Kenya --- ## **Overview** This repository contains a comprehensive end-to-end analysis of road traffic accidents in Kenya. The goal is to extract insights, identify high-risk locations, analyze temporal patterns, and examine behavioral, environmental, and demographic drivers of road accidents. The project includes data cleaning, exploratory analysis, temporal trends, geospatial mapping, summary insights, and recommendations. --- ## 1. Project Summary Road accidents in Kenya continue to impose a significant human and economic burden, despite ongoing mitigation efforts. This project analyzes 1,118 accident records from a compiled 2016–2017 dataset to: 1. Clean and harmonize manually-recorded accident data 2. Identify high-risk counties and hotspots 3. Examine accident patterns across time, roads, vehicle types, victim demographics, and cause codes 4. Extract meaning from unstructured accident narratives 5. Visualize findings through charts and a folium-based choropleth map 6. Establish a foundation for future predictive models or dashboards Source reference: Kaggle’s accidents_kenya dataset, curated by Waweru Fidelis --- ## 2. Dataset Description The dataset includes 15 raw attributes, later renamed, cleaned, and standardized. Key fields include: - Time and Date of the accident - County, Road, and Place - Vehicles involved - Accident narrative (“Brief Accident Details”) - Victim demographics (Gender, Age, Victim Type) - Cause Code mapped to Cause Description and Category - Number of victims An additional dataset containing Cause Codes → Descriptions → Categories is merged during preprocessing to enrich the analysis. ## 3. Key Objectives 1. Clean and standardize a highly irregular, manually recorded dataset 2. A …