ML4N--Adversarial-Attacks-Polito-2023-24

Analysing Adversarial Attacks on Tabular Data Classifiers

AUTHORS: Alberto Ameglio Enrico Di Stasio Gianluca Di Bella Cosimo Vergari

Command to install the libraries required by the script

pip install -r requirements.txt

Data exploration and preprocessing

Analysis of german credit data Project purpose The german credit data contains financial and banking details of customers. The given dataset contains information about individuals who have applied for credit from a bank. Each entry in the dataset represents a person, and they are classified as either good or bad credit risks based on their attributes. The task involves predicting whether the customer will repay a credit.

The aim of the project was to perform exploratory data analysis of german credit data. The goal of the data exploration and preprocessing was to gain knowledge about the features that influence credit repayment.

The MAIN FILE Here we developed our project

Other files

- Exploratory Data Analysis

- Supervised Data Analysis Where we made the workflow to decide to use SMOTE and we compared the results

- Unsupervised Data Analyis