Predicting-Crime-and-Proposing-Safer-Neighborhoods-Using-Machine-Learning-Models

The project targets the limitations of conventional crime prevention measures in effectively reducing criminal activities in dynamic urban environments. Traditional methods, which rely heavily on law enforcement presence and reactive punitive actions, are often unable to adapt quickly enough to emerging threats and evolving patterns of unlawful behavior. This initiative seeks to implement an innovative approach that utilizes data analysis for early detection of high-risk areas and times, allowing targeted interventions before incidents happen. Through these efforts, we aspire to revolutionize current crime prevention strategies, leading to safer communities and optimized allocation of law enforcement resources. Moreover, the advantages of the model extend to multiple sectors. Law enforcement agencies can effectively allocate their officers, city administrations can invest in crime prevention measures more efficiently, businesses can enhance their security protocols, and individuals can evaluate the risks present in their neighborhoods and take appropriate actions for personal and property safety. This initiative offers a groundbreaking approach to community safety by leveraging cutting-edge technology and adopting data-driven methodologies.

Technical report pdf file

Data_270_GWAR_Group_6.pdf

Presentation slides

Data270-Final-Team_Prsentation-Group-6.pdf

crime_dataset.csv is the original csv file

Located in the drive - https://drive.google.com/drive/u/1/folders/17mRl3pv7zmY3RM93jyG24a7E72NBckv-

Data_270_Data Collection_Preprocessing_Transformation_Feature_Engineering.ipynb

It has all the code till feature engineering

preprocessed_crimes_data.csv

This is the file after preprocessing our original dataset

ANN model

Code is divided into 2 parts i)Baseline ANN.ipynb - Baseline Model ii)Hyperparametertuned_ANN_model.ipynb - Hyperparametertuned model

Random Forest model

Random_Forest_Model.ipynb

Decision Tree model

DATA270_Decision_Tree.ipynb

XGBoost

XGBOOST.ipynb

Saved best model .

dt_hyperparamter_tuning_model.joblib

Note: Every file has .py file as well.