Chicago Crime Analysis - Big Data

Dataset - https://data.cityofchicago.org/Public-Safety/Crimes-2001-to-present/ijzp-q8t2

This dataset contained records from 2001 - 2019. Size - 1.6GB data (6.6M unique records).

Modules covered -

1. ETL (Cleaning and Filtering) and Data operations (child tables creation and write to Cassandra DB).
2. Interactive visualizations for Hourly, Monthly and Yearly Crime Analysis (Line Charts and GEO maps)
3. Wordcloud implementation for most occuring crime type and crime location.
4. Forecasting of crime for 2019.
5. Static visualizations for crime severity, Successful arrests (area-wise).

Tech Stack -

1. Data Cleaning & Filtering - Spark Dataframes and Spark SQL functions
2. Data Manipulation - Spark,Cassandra, Python, Pandas
3. Data Visualization - Plotly, Dash
4. Forecasting - FBprophet
5. Webapp - Dash, Flask

Illustration from Webapp -