NYC_risk

The process

1. Business Understanding
2. Find data
3. Understand data

• Understand all columns
• Run tests involving aggregate data: - Neighborhoods with most accidents - Day of the week with most accidents - Time of the day with most accidents - Type of car involved in most accidents

4. Define business rules (criteria)

• Criteria for dangerous routes
• Define a scale to measure risk
• Define the influence radius of an accident

5. Prepare data/matrix

• Choose the columns/criteria that make more sense (dimensionality reduction)
• Fill missing data by combining other data

6. Develop logic/work flow
7. Develop code
8. Test
9. Implement solution