Titanic Survival Prediction Enhanced by Innovative Feature Engineering and Multi-Model Ensemble Optimization

DOI

10.2991/978-94-6463-823-3_19How to use a DOI?

Keywords

Titanic Survival Prediction; Feature Engineering; Ensemble Learning; Hyperparameter Tuning

Abstract

This study enhances Titanic survival prediction through advanced feature engineering and ensemble model optimization. The Titanic dataset presents a classic binary classification problem requiring the prediction of passenger survival based on demographic and ticket information. Our methodology employs systematic preprocessing where missing values are intelligently imputed, including eXtreme Gradient Boosting (XGBoost) predictions for age values. Novel features were extracted from passenger names, cabin information, and family relationships to improve predictive power. Feature importance was evaluated using Random Forest and XGBoost algorithms, with SelectFromModel and Recursive Feature Elimination applied for effective feature selection. Three classification algorithms—logistic regression, random forest, and XGBoost—were systematically optimized using grid search, random search, and Bayesian optimization techniques. Based on these optimized models, soft voting and stacking ensemble approaches were implemented. Cross-validation results demonstrate that ensemble methods achieved superior accuracy (0.8361) compared to individual models: logistic regression (0.8339), random forest (0.8350), and XGBoost (0.8204). This research provides valuable optimization strategies for similar classification tasks, particularly highlighting how feature engineering combined with ensemble methods can substantially enhance predictive performance while maintaining computational efficiency.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Hanzhi Li
PY  - 2025
DA  - 2025/08/31
TI  - Titanic Survival Prediction Enhanced by Innovative Feature Engineering and Multi-Model Ensemble Optimization
BT  - Proceedings of the 2025 3rd International Conference on Image, Algorithms, and Artificial Intelligence (ICIAAI 2025)
PB  - Atlantis Press
SP  - 207
EP  - 217
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6463-823-3_19
DO  - 10.2991/978-94-6463-823-3_19
ID  - Li2025
ER  -