Scientific Online Resource System

Varna Medical Forum

How Titanic continues to change the future: machine learning in respiratory practice

Evgeni Mekov, Rosen Petkov

Abstract

The application of artificial intelligence (AI) in all fields and in medicine in particular is gaining more and more popularity. Machine learning (ML) is part of AI, in which computers use statistical methods for self-learning without being explicitly programmed to do so. The challenge Titanic: Machine Learning from Disaster is probably the most popular among researchers entering the field of artificial intelligence and machine learning.

In almost all cases, ML goes through several successive stages: definition of the problem, data gathering, data preparation, exploratory analysis, feature engineering; model creation, and model evaluation. The following manuscript briefly describes each of them. With their help, a model for the challenge "Titanic: Machine Learning from Disaster" was created. This model outperformed most of the models featured in the competition. It achieved a 99.8% better result than the other models presented.

Machine learning could improve the accuracy of different models with the same data (equal start). In this way, proper data processing could lead to higher accuracy, which in medicine translates into improved clinically significant endpoints.

 


Keywords

machine learning, artificial intelligence, algorithm, accuracy, titanic, medicine

Full Text


References

James G, Witten D, Hastie T, Tibshirani R. An Introduction to Statistical Learning: with Applications in R. Springer; 2013. Available from: https://faculty.marshall.usc.edu/gareth-james/ISL/ [Last accessed 20.05.2022]

https://www.kaggle.com/c/titanic [Last accessed 20.05.2022]

https://www.kaggle.com/medaxone/titanic-challenge-0-82057-for-beginners-with-tune [Last accessed 20.05.2022]

Claesen M, De Moor B. Hyperparameter search in machine learning. arXiv preprint arXiv:1502.02127, 2015.

Rokach L. Ensemble-based classifiers. Artificial Intelligence Review 2010;33(1–2):1–39.

Polikar R. Ensemble based systems in decision making. IEEE Circuits and Systems Magazine 2006;6(3):21–45.

https://xgboost.readthedocs.io/en/latest/ [Last accessed 20.05.2022]

https://lightgbm.readthedocs.io/ [Last accessed 20.05.2022]

https://catboost.ai/ [Last accessed 20.05.2022]

Buch V. Artificial intelligence in medicine: current trends and future possibilities. Br J Gen Pract. 2018;68(668):143–144.

https://www.kaggle.com/c/covid19-global-forecasting-week-5 [Last accessed 20.05.2022]

https://www.kaggle.com/c/stanford-covid-vaccine [Last accessed 20.05.2022]

https://www.kaggle.com/c/ranzcr-clip-catheter-line-classification [Last accessed 20.05.2022]

https://www.kaggle.com/c/vinbigdata-chest-xray-abnormalities-detection [Last accessed 20.05.2022]




DOI: http://dx.doi.org/10.14748/vmf.v11i2.8650

Refbacks

Font Size


|