Password Strength Classification

Implemented a robust password strength classification system using data science and machine learning techniques. The project involved analyzing a dataset containing user passwords, extracting relevant features, and training a multinomial logistic regression model to classify passwords into three categories: weak, normal, and strong.

Conducted exploratory data analysis (EDA) on the password dataset to understand the distribution of password strengths and identify patterns.

Engineered features including password length, frequency of lowercase, uppercase, digits, and special characters to capture variations in password strength.

Visualized the relationships between different features and password strength using box plots and distribution plots to gain insights into the data.

Utilized TF-IDF (Term Frequency-Inverse Document Frequency) transformation to convert password strings into numerical vectors suitable for machine learning algorithms.

Trained a multinomial logistic regression classifier on the transformed data to predict password strength categories.