The Titanic Dataset Project

What are We Building?

In this article, we are building a predictive model to analyze whether a random passenger will survive based on their age, gender, cabin, Fare, and more. First, we will visualize data and then, based on it, adjust the parameter values according to our requirements.

Pre-Requisites

Below are some pre-requisites-

• Knowledge of Python
• Understanding of Utility Libraries like Pandas, Numpy
• Basic knowledge of visualization libraries like Seaborn, Matplotlib
• Implementable knowledge of SciKit Learn
• Knowledge of concepts like - Data Transformation, Binning, Feature Engineering, Model Training, Model Evaluation, and Model Testing

How Are We Going to Build This?

1. Environment - Choosing an IDE or an environment to work on, either online or offline
2. Importing - Loading the datasets and libraries required
3. Analysing - Understanding the structure of the data like the columns, values, and data types
4. Visualisation - Visualising the dataset to understand underlying patterns
5. Filling out values - Cleaning the data by completing the dataset and filling out the missing values
6. Feature Engineering - Understanding the features and creating new ones according to requirements
7. Traning the model - Using algorithms like logistic regression and random forest to train the model
8. Evaluation - Evaluating the predictive model
9. Testing - Testing the model and getting the accuracy

Final Output

The resultant output will be a prediction measure of Logistic regression and Random Forest accuracy.

Requirements

• Libraries - Pandas, Numpy, Matplotlib, Seaborn, SciKit Learn
• Algorithms - Logistic Regression, Random Forest

About the Data

We are using the Titanic Dataset here. It contains both numerical and string values. It is pre-distributed over 3 CSV files - Train, Test, and Gender.

The predefined columns are -

• Passenger ID - To identify unique passengers
• Survived - If they survived or not
• P Class - The class passengers travelled in
• Name - Passenger Name
• Sex - Gender of Passenger
• Age - Age of passenger
• SibSp - Number of siblings or spouse
• Parch - Parent or child
• Ticket - Ticket number
• Fare - Amount paid for the ticket
• Cabin - Cabin of residence
• Embarked - Point of embarkment

Building the Classifier

1. Load the Data

Output:

2. Analyze and Visualize the Dataset

Output:

Output:

Output:

Output:

Output:

3. Data Transformation

Output:

4. Data Cleaning

Output:

5. Feature Engineering

Output:

Output:

Output:

Output:

Output:

Output:

Output:

Output:

Output:

Output:

Output:

Output:

Output:

Output:

Output:

6. Training and Testing the Model

Output:

Output:

Output:

7. Model Evaluation

Output:

What’s Next

To expand this project further -

• Flask - To have an interface, we can use Flask.
• Algorithms - Other algorithms like SVM, Perceptron, Gradient Descent, etc., can be used.
• accuracy - The accuracy can be increased by changing features.