🚀 Machine Learning Projects Showcase

Welcome to this repository containing two end-to-end Machine Learning projects, fully implemented in interactive Jupyter Notebooks. Each project is modular, self-contained, and features comprehensive Exploratory Data Analysis (EDA), data preprocessing pipelines, model selection, evaluation, and visualizations.

---

📂 Repository Structure

d:\Antigravity\
├── README.md                               # Project documentation (this file)
├── .gitignore                              # Git exclusion rules
│
├── 🚢 Titanic Survival Classification/
│   ├── Titanic_Survival_Classification.ipynb  # Comprehensive EDA, training & evaluation
│   └── data/
│       └── titanic.csv                      # Titanic dataset from Kaggle
│
└── 📈 Stock Price Prediction/
    ├── Stock_Price_Prediction.ipynb         # EDA, Linear Regression & LSTM training
    └── data/
        ├── AAPL_2020-01-01_2023-01-01.csv   # Historical Apple stock data (Yahoo Finance)
        └── AAPL_predictions_comparison.png  # Visual evaluation of stock forecasts

---

🚢 Project 1: Titanic Survival Classification

📝 Overview

This project constructs binary classification models to predict passenger survival on the Titanic based on passenger characteristics (age, gender, ticket class, fare, family size, etc.) sourced from Kaggle's classic dataset.

⚙️ Pipeline & Methods

1. Exploratory Data Analysis (EDA):
   • Visualized survival rates by demographics (gender, class, age distributions).
   • Examined correlation matrices of numerical attributes.
2. Data Cleansing & Imputation:
   • Handled missing data for Age (median imputation grouped by passenger class/gender) and Embarked.
   • Extracted social titles (e.g., Mr, Mrs, Miss, Master, Rare) from passenger names to infer status.
3. Feature Engineering:
   • Engineered new features such as FamilySize (Parch + SibSp + 1) and IsAlone.
   • Categorized fare ranges and applied one-hot encoding to categorical attributes.
4. Model Training & Evaluation:
   • Evaluated multiple classifier algorithms:
     • Logistic Regression (baseline classification)
     • Random Forest Classifier
     • Gradient Boosting Classifier
   • Compared performance metrics: Accuracy, Precision, Recall, F1-Score, and ROC-AUC curves.