This repository contains several mini-projects that apply Logistic Regression models to various datasets. Each project demonstrates the application of logistic regression techniques to solve real-world classification problem. The mini-projects utilize different methodologies and data preprocessing techniques to enhance model performance and accuracy, showcasing the effectiveness of logistic regression in various contexts.
This project analyzes fake bills using logistic regression, employing two approaches for handling missing values: without Iterative Imputer and with Iterative Imputer. The dataset contains 1,500 rows and 7 columns related to bill characteristics.
Dataset: Fake Bills Dataset.
This project analyzes Halloween candy rankings using logistic regression to predict whether a candy contains chocolate. The dataset includes various attributes for each candy, such as flavor types and rankings.
Dataset: The Ultimate Halloween Candy Power Ranking.
This project aims to identify relevant risk factors for heart disease and predict overall risk using logistic regression. It highlights the importance of early prognosis in reducing complications associated with cardiovascular diseases.
Dataset: Heart Disease Prediction Dataset.
This project predicts Myers-Briggs Type Indicator (MBTI) personality types using a synthetic dataset from Kaggle, which includes over 100,000 samples of demographic information, interest areas, and personality scores.
Dataset: Predict People Personality Types Dataset.
The sinking of the Titanic was one of the deadliest maritime disasters in history, with over 1,500 passengers losing their lives. This dataset is designed to help predict whether a passenger survived the disaster based on various factors such as age, sex, class, and ticket fare.
Dataset: Link to the dataset on Kaggle
- pandas
- matplotlib
- seaborn
- scikit-learn
Feel free to explore each project to understand the methodologies and results in more detail!