23-2_ML_TEAM_Titanic

📝 Project Info

Overview

Our project delves into the intricate realm of combating Website Fingerprinting—a menace to user privacy. As technology advances, adversaries exploit traffic patterns, posing a threat to anonymity, especially within the Tor network. Our research aims to unveil vulnerabilities inherent in Tor's defense mechanisms against fingerprinting attacks, crucial for preserving user privacy.

Problem Definition

Website Fingerprinting, a sophisticated form of attack, jeopardizes user privacy by allowing adversaries to deduce visited websites based on traffic patterns. Despite Tor's robust anonymity, it struggles to fully defend against such attacks, necessitating thorough research. Our focus is to identify and understand these vulnerabilities, paving the way for the development of effective defenses, ensuring the resilience of Tor's anonymity.

Experiment Settings and Environment

Executing experiments in the Colab environment, we harnessed the power of T4 GPU, 12.7GB RAM, and 78.2GB disk space. To ensure robustness, we employed an INTEL i9-12900K as an alternative during downtimes. Initial experiments involved feature visualization and subsequent refinement through the removal of less significant features, resulting in improved accuracy.

⚙️ How to Run

Setup on Google Colab

1. Open the provided Colab notebook in Google Colab.

2. Mount your Google Drive to access the dataset. Run the following code snippet:

   from google.colab import drive
   drive.mount('/content/drive')

3. Import necessary libraries and load the dataset:

   import numpy as np
   import pandas as pd
   import matplotlib.pyplot as plt
   
   from sklearn.model_selection import train_test_split
   from sklearn.preprocessing import StandardScaler
   from sklearn.metrics import accuracy_score, confusion_matrix

4. Load the dataset:

   mon_features_modified = pd.read_csv('/content/drive/MyDrive/ML_Project/mon_features_modified.csv')
   mon_labels = pd.read_csv('/content/drive/MyDrive/ML_Project/mon_labels.csv')
   
   unmon_features_modified = pd.read_csv('/content/drive/MyDrive/ML_Project/unmon_features_modified.csv')
   unmon_labels = pd.read_csv('/content/drive/MyDrive/ML_Project/unmon_labels.csv')

   ⭐ Note: If you have the dataset stored in a different location, modify the paths accordingly.

5. Run the code cells in the Colab notebook sequentially to reproduce the experiments.

🛳️ Team Titanic Member

HWANG JAEEUN	SONG JIMIN	SONG CHAEWON	AN CHAEYEON	LEE NAMYOUNG	KIM RINA