In this repository you can find the zip file "CodeAndData" which contains:
Code:
- Executable code in jupyter notebook: CodeCrimeLA.ipynb
Input files:
- Raw data: Crime_Data_from_2020_to_Present.csv
- Processed data: crime_clean_data.csv
Furtheremore, there is a compiled version of the code in HTML "DataAnalysis_CodeAndOutput.html".
Finally, the report is in PDF format in the file "LAPD_Analysis_Report".
In this project, we take on the role of data scientists whose mission involves collaborating closely with the Los Angeles Police Department (LAPD) to provide comprehensive insights aimed at enhancing their operational effectiveness. One of our goals is to investigate existing patrol boundaries; with the intention of establishing more intuitive ones, thereby improving the crime forecasting across a wide range of new “hotspot” areas. Additionally, our analysis involves assessing the safety of hourly time frames to better understand patterns and trends that emerge, with recommendations to hopefully strengthen efficient resource allocation. We also plan to delve into the dynamics of crime report responsiveness across various factors. The incentive of this report is two-pronged; to educate citizens on the importance of being attentive and reporting crimes that happen, and to inform police investigations for cases that might “fall between the cracks”. Furthermore, our objective includes empowering the LAPD with a daily anomaly detection tool, enabling proactive measures against irregularities in crime data and a monthly short-term forecasting to uncover the crime trend activity. Finally we conduct a deep-dive on victim demographics, to better understand what sort of individuals are most often targeted, with the intent of suggesting mitigation strategies.
