Skip to content

Latest commit

 

History

History
127 lines (96 loc) · 6.84 KB

README.md

File metadata and controls

127 lines (96 loc) · 6.84 KB
Raw

Data Visualization Assignment 3 Report

Contents

Introduction

This README accompanies the Data Visualization Assignment 3 report, detailing the analysis, tasks performed, visualizations created, and insights derived from crime and arrest data in Los Angeles.

Dataset

The dataset includes crime and arrest data from Los Angeles spanning from January 2020 to December 2023.

Data Pre-Processing

The data underwent preprocessing steps, including cleaning and filtering out entries with unknown location coordinates.

Objective of Data Analysis

The primary objective was to determine the impact of location and time on crime counts within various areas of Los Angeles.

Tasks

Two main tasks were identified:

  1. Identifying temporal periods when crime counts surged in different areas.
  2. Examining the impact of crime surges on the quantity of arrests within specific areas.

Analytics Workflow

The workflow encompassed various stages, including data loading, preprocessing, analysis, visualization, and deriving insights.

Visualizations

Detailed visualizations were created to analyze crime patterns, including bar charts, heatmaps, seasonal comparisons, yearly trends, moving averages, and spatial distribution maps.

Bar Chart - Distribution of Crime Counts Across Areas

This visualization illustrates the total number of reported crimes across different areas in Los Angeles. Each bar represents a distinct area, providing a visual comparison of crime frequency among various areas.

Heatmap of Crime Incidents Across Los Angeles

The heatmap visualizes the spatial distribution and intensity of reported crime incidents across different areas in Los Angeles. It showcases the geographical concentration of crime incidents, offering insights into crime hotspots or patterns based on location coordinates.

Seasonal Crime Comparison Across Areas

Small multiples showcase the comparison of seasonal crime incidents across different areas in Los Angeles. The plot comprises subplots representing distinct seasons—Fall, Spring, Summer, and Winter—depicting the count of reported crimes within various areas during these seasons.

Yearly Crime Trends Over Days

This visualization presents the yearly trends of reported crime incidents from 2020 to 2023, demonstrating how the count of reported crimes fluctuates over the days of the respective years.

Moving Average - Small Multiples of Crimes Across Months Per Area

The moving average plots visualize the 7-day average of crimes in each area for each year as a line chart. These plots help in understanding trends by smoothing out short-term fluctuations.

Bubble Map within each Area per Year

This visualization uses unsupervised clustering to identify crime clusters within each area for each year. Each cluster represents the concentration of crimes, plotted on the Los Angeles map based on location coordinates.

Bar Graph of Arrests in the Detected Area Name

Bar graphs depict the number of arrests in specific areas during particular time frames, helping understand the impact of crime surges on arrest quantities.

Inferences

Insights were derived, highlighting specific periods and areas experiencing notable spikes in crime counts and analyzing the correlation between crime spikes and arrest quantities.

Contributors

  • Munagala Kalyan Ram
  • Vikas Kalyanapuram
  • Ramsai Koushik

References

  1. Crime Data 2020 to Present in Los Angeles: Crime Data Link
  2. Crime Dataset Description: Crime Dataset Description Link
  3. Arrest Data 2020 to Present in Los Angeles: Arrest Data Link
  4. Arrest Dataset Description: Arrest Dataset Description Link
  5. Matplotlib Tutorial: Matplotlib Tutorial Link
  6. Seaborn Tutorial: Seaborn Tutorial Link
  7. Folium Tutorial: Folium Tutorial Link
  8. A1 Report: A1 Report Link

How to Run

Prerequisites

Ensure you have the following installed:

  • Python (version 3.x)
  • Jupyter Notebook
  • Required libraries: pandas, matplotlib, seaborn, folium, scikit-learn

Steps

  1. Clone the Repository:
git clone https://github.com/DataViz-Trio/Assignment-3.git

cd Assignment-3
  1. Install Dependencies: Ensure the required libraries are installed using the following command:
pip install pandas matplotlib seaborn folium scikit-learn

  1. Run the Notebook: Open Jupyter Notebook and navigate to the cloned repository. Run the DataVisualizationAssignment3.ipynb notebook.

  2. Exploring Visualizations:

  • The notebook contains different sections, each generating specific visualizations. Execute each cell to generate the visualizations.
  • Modify parameters or input data as needed to explore different aspects of the data or adjust the visualizations.
  1. Understanding Outputs:
  • The notebook generates various visualizations showcasing crime patterns, seasonal trends, spatial distributions, and more.
  • Explore each visualization to understand the insights derived from the data.