Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fraud Detection with Logistic Regression #91

Open
akshaydubey05 opened this issue May 14, 2024 · 9 comments
Open

Fraud Detection with Logistic Regression #91

akshaydubey05 opened this issue May 14, 2024 · 9 comments

Comments

@akshaydubey05
Copy link

akshaydubey05 commented May 14, 2024

Objective : This model would predict whether a transaction is fraudulent or not based on transaction features.

This model would understand data, preprocess, train logistic regression; assess metrics, interpret coefficients; deploy for real-time security .

0-Normal
1-Fraud

1_Sd6cpc9Yrjmb588si4LyLA

@akshaydubey05
Copy link
Author

Hi, i created this issue, please assign it to me under GSSOC'24.

@RishiRaj292
Copy link

Hello @Akshat111111 , could you please also add the label 'status' to show whether the issue has been assigned or not to a person (And preferably to all the issues if possible)? This would save a lot of time for the others viewing the project for issues.
Thank you

@Saswatsusmoy
Copy link
Contributor

@Akshat111111 Instead of using logistic regression, better ML models which can capture intricate features and help in better classification can be used. Decision Trees and Random Forest are better options because they help in preventing overfitting in these cases which is a common problem in financial fraud detection models.

It would be better if you can change the issue to detect fraudulent transaction instead of specifying a single ML algorithm

@akshaydubey05
Copy link
Author

akshaydubey05 commented May 15, 2024

Hi @Saswatsusmoy thanks for the suggestion ,I thought of making this issue using Random Forest as it is the most efficient one, but at that time i was learning it. No worries, now i will do this task with both Linear Regression and Random Forest.

@Akshat111111
Copy link
Owner

Hello @Akshat111111 , could you please also add the label 'status' to show whether the issue has been assigned or not to a person (And preferably to all the issues if possible)? This would save a lot of time for the others viewing the project for issues. Thank you

I will advice you to create a new issue as currently all the issues are assigned and the peoples are working on it.

@Srijansarkar17
Copy link
Contributor

Hey @Akshat111111 I would like to solve this issue. Can you please assign it to me? I will use neural networks for this problem. Adding hidden layers will improve the accuracy too.

@Akshat111111
Copy link
Owner

Hey @Akshat111111 I would like to solve this issue. Can you please assign it to me? I will use neural networks for this problem. Adding hidden layers will improve the accuracy too.

I will suggest you to create a new issue and tell your approach in a brief way .

@Samik123Mit
Copy link

pls assign me the issue,i have prior experience with the model task.

@Akshat111111
Copy link
Owner

pls assign me the issue,i have prior experience with the model task.

create a new issue with your idea

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

6 participants