Titanic Survival Prediction

Overview

Titanic Survival Prediction is a machine learning project designed to predict whether a passenger survived the Titanic disaster based on various personal and travel-related features. This model demonstrates essential data science practices such as data preprocessing, feature engineering, and classification.

Dataset

The dataset used for training is Titanic.csv, which contains the following columns:

• Survived: Target variable (1 = Survived, 0 = Did not survive)
• Pclass: Ticket class (1st, 2nd, 3rd)
• Name: Passenger name
• Sex: Gender
• Age: Age of the passenger
• SibSp: Number of siblings/spouses aboard
• Parch: Number of parents/children aboard
• Ticket: Ticket number
• Fare: Ticket fare
• Cabin: Cabin number
• Embarked: Port of embarkation

Objectives

• Develop a classification model to predict passenger survival.
• Apply feature engineering techniques.
• Handle missing data and encode categorical variables.
• Evaluate model performance using accuracy, precision, recall, and F1-score.

Feature Engineering

The following new features were created:

• Title: Extracted from the Name column (e.g., Mr, Mrs, Miss, etc.)
• FamilySize: Sum of SibSp and Parch plus 1 (self)
• IsAlone: Indicates whether the passenger was traveling alone (FamilySize == 1)

Model Evaluation

The model is evaluated using the following metrics:

• Accuracy
• Precision
• Recall
• F1-score

Installation and Usage

Prerequisites

• Python 3.x
• Google Colab or Jupyter Notebook
• Required libraries:
  • pandas
  • numpy
  • scikit-learn
  • matplotlib
  • seaborn

Setup

Click COLAP Link And Downalod Dataset then You are Ready to start