Crime rate prediction has become an important tool for law enforcement agencies
to help them better understand patterns of crime and anticipate where crime is
likely to occur. By predicting future crime trends, law enforcement agencies can
better allocate resources to areas that are likely to experience increases in
criminal activity. This could lead to a decrease in crime overall, as well as
an increase in public safety. Additionally, crime rate prediction can help police
departments develop better strategies for responding to crime as it happens.
The dataset is prepared manually based on the publication available on the Indian
National Crime Rate Bureau (NCRB) official website. This data provides statistics
on crimes committed in 19 metropolitan cities during the year 2014 to 2021.
With the help of this application, we can predict the crime rates for 10 different
categories of crime that are likely to occur in 19 Indian metropolitan cities over
the next few years. It includes statistics on 10 different categories of crime,
including murder, kidnapping, crime against women, crime against children,
crime committed by juveniles, crime against senior citizens, crime against SC,
crime against ST, economic offences and cyber crimes.
The system uses scikit-learn's Random Forest Regression model, which takes year,
city name, and crime type data as inputs. Random Forest Regression is a type of
an ensemble learning techniques that can be used to predict continuous values
from a collection of data. It works by creating a large number of "decision trees"
which each make a prediction about the target variable. Then it averages all the
predictions to come up with a final prediction. This makes it more accurate than
a single decision tree. The model predicts the crime rate with an accuracy of 93.20%
on testing dataset.
Click here to access the project report.