Today I officially kicked off my Titanic Survival Prediction project 🚀
As the first step, I focused on getting familiar with the data through basic EDA (Exploratory Data Analysis).

🔍 What I explored today

1. Loaded the train and test datasets
2. Inspected data structures (head(), info(), describe(include='all'))
3. Checked for missing values and noted columns needing attention (Age, Cabin, Embarked, Fare)
4. Visualized key features:
   • Survival distribution (Survived)
   • Gender vs. Survival rate (Sex)
   • Passenger Class vs. Survival rate (Pclass)
5. Set the initial roadmap for data cleaning:
   • Fill or drop missing values appropriately
   • Engineer new features if necessary (e.g., family size, title extraction)

🧠 Key Takeaway

Good EDA is about asking: “How will this help the model later?”
Rather than aimlessly plotting graphs, I tried to focus on features that might actually impact survival outcomes.

🧩 Next Steps

• Handle missing values carefully (especially Age, Cabin, Embarked, and Fare)
• Explore feature engineering opportunities (e.g., extracting titles from names, family-related features)
• Start simple baseline modeling to establish reference performance

Feels good to finally set sail 🚢
Let’s survive and thrive in this competition!