Classifying-Criminal-Offenses

Classification Application in Python Using scikit-learn and TensorFlow-Keras

(PySpark program included to show data exploration/manipulation and descriptive statistics analyses)

This repository contains the prediction of more serious crimes using Chicago crime data accessed via Google BigQuery Storage API.

Goals

• Using Chicago Police Department crime data from 2001 to present, summarize and examine crime statistics.

• Build and train Classification models to predict index offenses (more serious crimes) implementing the following approaches:

  • Logistic Regression
  • Random Forests Classification
  • Support Vector Machines Classification
  • XGBoost Classification
  • Neural Networks Classification

• Apply regularization and cross-validation techniques for model evaluation, selection, and optimization.

Business Use Case

By implementing a machine learning project for classifying criminal offenses, law enforcement agencies can optimize resource allocation, enhance their effectiveness, improve public safety, and achieve better outcomes for communities.

1. Improved Resource Allocation: By accurately predicting the severity of crimes, law enforcement agencies can allocate their resources more efficiently. High-risk areas and severe crime predictions can be prioritized, ensuring that more officers and resources are directed to where they are most needed.

2. Proactive Crime Prevention: Machine learning models can help identify patterns and trends in criminal activity. This allows law enforcement to anticipate and prevent severe crimes before they occur, leading to a reduction in overall crime rates and enhancing community safety.

3. Enhanced Decision-Making: With reliable predictions of crime severity, policymakers and law enforcement leaders can make better-informed decisions regarding crime prevention strategies, community policing efforts, and emergency response plans.

4. Judicial Efficiency: Accurate classification of crime severity can assist the judicial system in prioritizing cases, streamlining court processes, and ensuring that severe crimes are addressed promptly. This can lead to faster resolutions and a more efficient legal system.

5. Community Trust and Transparency: Demonstrating the use of advanced technologies to tackle crime can increase public trust in law enforcement. Transparent use of machine learning models shows a commitment to leveraging innovative solutions for community safety.

6. Cost Savings: Efficient resource allocation and crime prevention strategies can lead to significant cost savings for law enforcement agencies. Reduced crime rates can also lower the economic burden on society, including costs related to property damage, healthcare, and legal proceedings.

7. Data-Driven Insights: The project can provide valuable insights into the underlying factors contributing to severe crimes. These insights can inform targeted interventions and policies aimed at addressing the root causes of crime, leading to long-term societal benefits.