#Does anyone have a code/project of fraud analysis using python?

44 messages · Page 1 of 1 (latest)

plucky obsidian
#

i want it for my college project please help

real minnow
#

What do you mean by you "want it for your college project"?

plucky obsidian
#

tomorrow is my due date of project submission

real minnow
#

We're not just gonna give you a finished project tf?

#

!rule academics

ivory vesselBOT
#
Academic Honesty

You may ask for help with homework but asking others to help on tests/quizzes or to do your coursework for you is not allowed.

plucky obsidian
#

i will do my project myself just want a python code

#

how its gonna be performed

#

used chat gpt and all but didn't understand anything

real minnow
#

I'm not sure how we can help you when your project is due tomorrow and you have nothing?

plucky obsidian
#

sorry broo

#

i am trying from last few days

#

and someone suggest me this

real minnow
#

Otherwise you won't be learning

plucky obsidian
#

i can send you the file?

#

i am unable to run it

#

😕

real minnow
#

Why can't you run it?

#

That's weird that you have a file you can't run

plucky obsidian
#

i am unable to set up path

#

in vs code

real minnow
#

We can help you with that

#

You just need to learn to ask

ivory vesselBOT
#

@plucky obsidian

File Attachments Not Allowed

For safety reasons we do not allow file and video attachments.

Code Formatting

You can share your code using triple backticks like this:
```
YOUR CODE
```

Large Portions of Code

For longer scripts use Hastebin or GitHub Gists and share the link here

Ignored these files
  • Fraud_Project.ipynb
plucky obsidian
real minnow
#

A FN means it is fraudulent, but your model classified it as legitimate

plucky obsidian
#

where am i wrong

#

😭

real minnow
#

The Bayes model has the highest FN rate; however, this error is less serious (in a practical sense, these transactions would be flagged and fraud, and later resolved by the user, which is better than totally missing a fraudulent transaction), so we give more weight in our decision to the low FP.

#

A fraudulent transaction should be labeled positive

#

A false negative error, or false negative, is a test result which wrongly indicates that a condition does not hold. For example, when a pregnancy test indicates a woman is not pregnant, but she is, or when a person guilty of a crime is acquitted, these are false negatives (from wikipedia)

#

I.e. your fraud detection model indicates a transaction is not fraudulent, but it is. That is a FN which your Naive Bayes has a lot of

#

Not a good look if I do say so myself

#

This is easily remedied by adjusting the priors or doing some sampling method as your dataset is most likely highly imbalanced (how effective these will be we won't know unless they've been tried)

#

594,643 rows of data, including 587,443 normal payments and 7,200 fraudulent
yup highly unbalanced

#

Actually I think a lot of your methodology and conclusions are sus

#

But it's getting late for me, so idk if anyone else will be able to help you especially as you haven't actually asked for any specific help or explained your problem

real minnow
#

Wait is that even your repo???

#

Last commit 6 years ago

#

Is blud cheating and trying to pass off someone else's analysis as their own

plucky obsidian
#

yaa

real minnow
#

You are cheating?