Scribd
Upload
15 views9 pages

Crime Rate Prediction Using ML

Uploaded by

gelardbamchhor02
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
15 views9 pages

Crime Rate Prediction Using ML

Uploaded by

gelardbamchhor02
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 9

Crime Rate Prediction

Using ML

Submitted By: Abhay Rautela Submitted To: Ms. Sonali


University Roll No: 2018059 Designation: Assistant Professor,
GEHU
INTRODUCTION

 Crime is an act that is prohibited by law and is punishable by a fine,


imprisonment, or other legal action.

 Crime prediction is significant to determine increase or decrease in crime rate


from preceding years.

 Sophisticated ML algorithms and data-driven methods can be used to identify


patterns in data to detect and predict criminal activities.

 It can also be used to create predictive models that identify the likelihood of a
certain crime occurring in a particular area or time frame.

 Using machine learning, law enforcement professionals can gain insight into
criminal behavior, enabling them to better detect and prevent crime
THE DATASET

 The dataset is prepared manually based on the publication available on the


National Crime Rate Bureau (NCRB) official website.

 The data provides statistics on the number of crimes committed in 19 Indian


metropolitan cities during the year 2014 to 2021.

 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.
ALGORITHM – RANDOM FOREST

 Random Forest is an ensemble learning technique that combines multiple decision


trees to generate a more accurate prediction.

 It is a versatile and powerful algorithm used in a wide variety of applications.

 The algorithm can be illustrated as follows:


1. Begin with a dataset of observations and their associated labels.

2. Randomly select ‘k’ features from the dataset.

3. For each of the ‘k’ features, choose the best split point.

4. Create a tree using the chosen split points.

5. Repeat steps 2-4 for each of the ‘k’ features.

6. Combine the trees to form a forest.

7. Use the forest to make predictions on new data.

8. Calculate the accuracy of the predictions.


PROPOSED SYSTEM
 Data Preprocessing: The data is prepared in the correct format for analysis.
Some columns are removed or transformed, and label encoding is used to
convert the categorical data into numeric for better prediction.

 Random Sampling: After feature selection, the data has been splitted into two
parts: training data (70%) and testing data (30%).

 Model Creation: The model algorithms are imported from sklearn. The dataset
has been analyzed using five different models: support vector machine,
nearest neighbor, decision tree, random forest, and neural network.

 Model Selection: Based on the defined goals and model performance, random
forest model has been selected.

 Model Deployment: The model has been deployed for prediction using flask
web technology.
SYSTEM ARCHITECTURE
RESULTS & DISCUSSION
 The Random Forest Regression model demonstrates the best accuracy in
predicting test data among the five selected models.

 The model predicts the crime rate with an accuracy of 93.20% on the testing
data.

 The accuracy results obtained


Algorithm Meanafter testing
Meanare listedR2
below:
Score
Absolute Squared
Error Error
Support Vector Machine 10.3204 371.7907 0.17886

K-Nearest Neighbor 6.58181 140.8179 0.55349

Neural Networks 12.4248 307.5506 0.24823


MLPRegressor
Decision Tree Regressor 2.89024 34.95932 0.88915

Random Forest Regressor 2.49143 21.43956 0.93201


CONCLUSION
 Crime rate prediction has become an important tool for law enforcement
agencies to help them focus their resources in high-crime areas.

 By focusing their resources in the right areas, police officers can help reduce
the overall crime rate in a community.

 As a result of machine learning technology, finding relationships and patterns


between various data has become easier.

 The model prediction of crime rate and data visualization helps in analysis of
data set and prediction of crimes.

 The model helped in understanding different crime datasets that can be used
in implementing the factors that can help in keeping society safe.
THANK YOU

You might also like